{"thread":{"id":"23060","subject":"What's in a name? Let's use a (uuid,name,email) triplet","startedAt":"2010-03-18T13:23:11Z","lastAt":"2010-03-24T19:23:38Z","messageCount":104,"participants":["Michael Witten","Jon Smirl","Linus Torvalds","Matthieu Moy","Nicolas Pitre","Martin Langhoff","Wincent Colaiuta","Reece Dunn","tytso@mit.edu","A Large Angry SCM","Sitaram Chamarty","Nazri Ramliy","Michael Haggerty","david@lang.hm","Mike Hommey","Paolo Bonzini","Jakub Narebski","Michael J Gruber","Erik Faye-Lund","Junio C Hamano","Mark Brown","Avi Kivity"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"137089","messageId":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","threadId":"23060","inReplyTo":null,"subject":"What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T13:23:11Z","receivedAt":"2010-03-18T13:23:11Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"Short Version:\n-------------\n\n\nRather than use a (name,email) pair to identify people, let's use\na (uuid,name,email) triplet.\n\nThe uuid can be any piece of information that a user of git determines\nto be reasonably unique across space and time and that is intended to\nbe used by that user virtually forever (at least within a project's\nhistory).\n\nFor instance, the uuid could be an OSF DCE 1.1 UUID or the SHA-1 of\nsome easily remembered, already reasonably unique information.\n\nThis could really help keep identifications clean, and it is rather\nstraightforward and possibly quite efficient.\n\n\nLong Version:\n------------\n\n\nThere are 2 reasons why people contribute (pro bono) to projects:\n\n (0) To improve the project.\n (1) To garner recognition.\n\nand in my experience, (0) is not as sweet without (1).\n\nOne of the great boons of distributed systems like git is that they\nseparate author (contributor) identities from committer identities,\nthereby maintaining (some semblance of) proper attribution in an\nofficial, structured format that is amenable to parsing by tools.\n\nWhile git's use of (name,email) pairs to identify each person is\nextremely practical, it turns out that it's rather `unstable';\nconsider the following information gleaned from a clone of the\nofficial git repository:\n\n $ git shortlog -se origin/master | grep Linus\n 3 Linus Torvalds \n 122 Linus Torvalds \n 235 Linus Torvalds \n 276 Linus Torvalds \n 9 Linus Torvalds \n 439 Linus Torvalds \n 9 Linus Torvalds \n\n $ git shortlog -se origin/master | grep Junio\n 3658 Junio C Hamano \n 2 Junio C Hamano \n 3 Junio C Hamano \n 3 Junio C Hamano \n 8 Junio C Hamano \n 4167 Junio C Hamano \n 2 Junio C Hamano \n 2 Junio Hamano \n\nor using a clone of Linus's Linux repo:\n\n $ git shortlog -se origin/master | grep Linus\n 2 Linus Luessing \n 2 Linus Lüssing \n 2 Linus Nilsson \n 2 Linus Nilsson \n 32 Linus Torvalds \n 1522 Linus Torvalds \n 4174 Linus Torvalds \n 7 Linus Torvalds \n 2 Linus Torvalds \n 8 Linus Torvalds \n 4 Linus Torvalds \n 166 Linus Torvalds \n 1 Linus Torvalds \n 1606 Linus Torvalds \n 174 Linus Torvalds \n 1 Linus Walleij (LD/EAB \n 3 Linus Walleij \n 1 Linus Walleij \n 81 Linus Walleij \n 9 Linus Walleij \n\n $ git shortlog -se origin/master | grep Morton\n 581 Andrew Morton \n 836 Andrew Morton \n 1 Andrew Morton \n\nFrom these few examples it seems pretty clear that the most volatile\nportion of the (name,email) pair is the email, which is unfortunate\nbecause the email is the most uniquely identifying information. Are\nwe really reasonably certain that these two are the same person?\n\n Linus Walleij \n Linus Walleij \n\nThus, I propose a more stable form of identification; rather than\nusing just a (name,email) pair, let's use a (uuid,name,email) triplet,\nwhere the uuid can be any piece of information that a user of git\ndetermines to be reasonably unique across space and time and that is\nintended to be used by that user virtually forever (at least within a\nproject's history).\n\nFor instance, Linus is always stuck in his basement with the same\nancient computers, so he chooses to set up his few ~/.gitconfig\nfiles with an OSF DCE 1.1 conforming UUID (generated by, say, uuidgen):\n\nLinus Torvalds \n\n [user]\n uuid = 6b202ed1-e8ec-4048-84c2-ae0dd3b2df47\n name = Linus Torvalds\n email = torvalds@linux-foundation.org\n\nOn the other hand, Junio is infatuated with the latest palmtop\ncomputing gadgets and finds himself setting up a ~/.gitconfig file\nseveral times each month; he doesn't want to bother remembering\nsome long human-hostile string, so he adopts as his uuid the\nSHA-1 of some easily remembered piece of information like the\nvery first (name,email) pair that he used for git\n(Junio C Hamano ):\n\n [user]\n uuid = 6e99d26860f0b87ef4843fa838df2a918b85d1f7\n name = Junio C Hamano\n email = gitster@pobox.com\n\nI'm sure that some optimizations could made for certain choices like\nUUID and SHA-1 strings.\n\nAnyway, I think this could really help keep identifications clean,\nand it is rather straightforward and possibly quite efficient.\n\nSincerely,\nMichael Witten\n"},{"id":"137091","messageId":"9e4733911003180648w74f48194w1d75b0d58a309f1e@mail.gmail.com","threadId":"23060","inReplyTo":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T13:48:57Z","receivedAt":"2010-03-18T13:48:57Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"You can't go back and edit the history in git so a map of the aliases\nis needed. The easy fix is a .mailmap file. However, the .mailmap\nentries need a mechanism to track which entries are correct and which\nhave been fixed. Read this long and painful thread...\nhttp://lkml.org/lkml/2008/7/28/134\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137095","messageId":"b4087cc51003180726s4d5c679egf76f5179383459c5@mail.gmail.com","threadId":"23060","inReplyTo":"9e4733911003180648w74f48194w1d75b0d58a309f1e@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T14:26:07Z","receivedAt":"2010-03-18T14:26:07Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 08:48, Jon Smirl wrote:\n> You can't go back and edit the history in git so a map of the aliases\n> is needed.  The easy fix is a .mailmap file. However, the .mailmap\n> entries need a mechanism to track which entries are correct and which\n> have been fixed. Read this long and painful thread...\n> http://lkml.org/lkml/2008/7/28/134\n\nThe addition of a uuid would not only likely decrease future trouble\ntremendously, but also allow for a much more efficient remapping of\nold (name,email) pairs.\n"},{"id":"137105","messageId":"alpine.LFD.2.00.1003181022040.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T17:27:50Z","receivedAt":"2010-03-18T17:27:50Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Michael Witten wrote:\n>\n> Short Version:\n> -------------\n> \n> Rather than use a (name,email) pair to identify people, let's use\n> a (uuid,name,email) triplet.\n\nEven shorter version: NO.\n\n> Long Version:\n> ------------\n\nUUID's are some total crazy shit. It's like XML. If you think you need \nthem, you're almost certainly wrong. If it's about identifying a unique \npiece of hardware, ok. If it's about identifying people, no.\n\nHow about you walk around with a bar-code tattooed to your forehead? Don't \nlike the idea? Then think about having to care about a uuid in your \nprojects. Same deal.\n\nNobody is going to associate themselves with a uuid. It's not how humans \nwork. It's degrading, and it's work-for-no-gain to anybody who doesn't \nhave OCD.\n\nSo in practice, the only thing that would happen is that people make up \nrandom uuid's and they'd be different for every single machine they have, \nbecause absolutely NOBODY would ever bother to try to save and move their \nuuids around.\n\nSo when you point out that emails aren't unique, or that people change \ntheir emails over time, please realize that the emails are _more_ stable \nthan a uuid would ever be. Because an email actually has some emotional \nattachment to the person in question. Yes, they change. So do real names \ntoo (which change more seldom, exactly because people are way _more_ \nemotionally attached to their real names).\n\nuuid's? I can pretty much guarantee that for me, it would be different for \nevery single machine I have. Because I could just not be bothered to care.\n\n\t\t\tLinus\n"},{"id":"137107","messageId":"4ba27424.9804cc0a.4b1f.11b1@mx.google.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181022040.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T18:42:44Z","receivedAt":"2010-03-18T18:42:44Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"\n\nLinus: Don't skim; read.\n\n\n\nOn Thu, Mar 18, 2010 at 12:27, Linus Torvalds wrote:\n> So in practice, the only thing that would happen\n> is that people make up random uuid's and they'd\n> be different for every single machine they have,\n> because absolutely NOBODY would ever bother to\n> try to save and move their uuids around.\n>\n> ...\n>\n> please realize that the emails are _more_ stable\n> than a uuid would ever be. Because an email\n> actually has some emotional attachment to the\n> person in question.\n\nMy anticipation of your response was uncanny:\n\n >> For instance, the uuid could be... the SHA-1\n >> of some easily remembered, already reasonably\n >> unique information.\n >>\n >> ...\n >>\n >> ...he doesn't want to bother remembering some\n >> long human-hostile string, so he adopts as\n >> his uuid the SHA-1 of some easily remembered\n >> piece of information like the very first\n >> (name,email) pair that he used for git\n >> (Junio C Hamano )\n\nSo, forget the original generality and let's\ndefine the uuid as a SHA-1 of some EASILY\nREMEMBERED, already reasonably unique piece of\ninformation such as an old (name,email) pair.\n\nTo make life easier on people, git tools could automate\nthat process; to Junio, his just uuid is an old,\nunchanging (name,email) pair:\n\n $ git config --global user.name \"Junio C Hamano\"\n $ git config --global user.email \"gitster@pobox.com\"\n $ git config --global --uuid \"Junio C Hamano \"\n\nwhich produces something like:\n\n [user]\n name = Junio C Hamano\n email = gitster@pobox.com\n uuid = 6e99d26860f0b87ef4843fa838df2a918b85d1f7\n\nIn fact those three steps should probably be\nfurther automated anyway:\n\n $ git config --global --init\n Full Name? Junio C Hamano\n Email? gitster@pobox.com\n UUID [Junio C Hamano ]? Junio C Hamano \n\nSet it and forget it in a completely human way.\n\nCould people still bungle the uuid or enter trash?\nSure, but that's essentially no different than the\ncurrent situation. This would be an improvement,\nbecause at least some people would take advantage\nof it; in fact, I bet most people would use it\nproperly because:\n\n * The information required is easily remembered\n and reproduced; it has that emotional aspect.\n\n * People have an emotional attachment to getting\n proper attribution for their work, and this\n helps.\n\nMoreover, storing and using the SHA-1 uuid would be\nvery efficient and allow for saner .mailmap hacks.\n\nSincerely,\nMichael Witten\n"},{"id":"137108","messageId":"vpq8w9po4az.fsf@bauges.imag.fr","threadId":"23060","inReplyTo":"4ba27424.9804cc0a.4b1f.11b1@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Matthieu Moy","fromEmail":"matthieu.moy@grenoble-inp.fr","sentAt":"2010-03-18T18:47:16Z","receivedAt":"2010-03-18T18:47:16Z","isPatch":false,"sender":{"key":"matthieu.moy@grenoble-inp.fr","avatar":"https://gravatar.com/avatar/72c8a2705971a25dfaff23cece15130d405685845d911aedd5667ace277f3fc5?d=mp&s=160"},"body":"Michael Witten writes:\n\n> So, forget the original generality and let's\n> define the uuid as a SHA-1 of some EASILY\n> REMEMBERED, already reasonably unique piece of\n> information such as an old (name,email) pair.\n\nWhat's the added value of the \"SHA-1\" thing, here? A hash of a pair\n(a, b) is exactly as unique as the pair itself (well, actually even a\nbit less if you consider collisions).\n\n-- \nMatthieu Moy\nhttp://www-verimag.imag.fr/~moy/\n"},{"id":"137112","messageId":"b4087cc51003181157i7c8b46favb52d4ccda75be799@mail.gmail.com","threadId":"23060","inReplyTo":"vpq8w9po4az.fsf@bauges.imag.fr","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T18:57:38Z","receivedAt":"2010-03-18T18:57:38Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 13:47, Matthieu Moy\n wrote:\n> What's the added value of the \"SHA-1\" thing, here? A hash of a pair\n> (a, b) is exactly as unique as the pair itself (well, actually even a\n> bit less if you consider collisions).\n\nYour observation is correct, but I'm pushing for the SHA-1 string\nbecause it could be efficiently parsed, stored, and used; it's\nessentially an optimization (or a preparation for an optimization).\n\nIf that's not a good way to approach it, then I'd be satisifed with\njust a straight (name,email) pair or any other reasonably unique\nstring.\n\nOn a more general note, the idea of a uuid is to distribute the\nprocess of canonicalizing identities. Does that not make perfect\nsense?\n"},{"id":"137114","messageId":"9e4733911003181202h23b3d702xce8fc0670a994d52@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181022040.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T19:02:59Z","receivedAt":"2010-03-18T19:02:59Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 1:27 PM, Linus Torvalds\n wrote:\n>\n>\n> On Thu, 18 Mar 2010, Michael Witten wrote:\n>>\n>> Short Version:\n>> -------------\n>>\n>> Rather than use a (name,email) pair to identify people, let's use\n>> a (uuid,name,email) triplet.\n>\n> Even shorter version: NO.\n>\n>> Long Version:\n>> ------------\n>\n> UUID's are some total crazy shit. It's like XML. If you think you need\n> them, you're almost certainly wrong. If it's about identifying a unique\n> piece of hardware, ok. If it's about identifying people, no.\n\nWe could hash people emails and then build a .mailmap equivalent thus\nhiding their identity.\n\nSeveral things needed to be combined to build that mailmap.\n1) a lot of hand work to identify aliases and misspellings\n2) work with google to translate email addresses into human names when\nnames were missing\n3) a list of all of the email addresses that had been checked, to make\nit easy to identify new ones.\n\nThe trouble with hashing it is that all of the tools that use it will\nneed to be rewritten.\n\nI'd really like to see a more global database constructed that links\ncommits, lkml discussions and the various distribution bug databases\nbut apparently it is too much of a threat to developer privacy. You\ncan achieve the same effect with a few hours in google throwing out\nbunches of false positives. It would be cool to be looking at a\nsubroutine, poke a button and then see all of the human oriented\nhistory around it instead of just the diffs.\n\n>\n> How about you walk around with a bar-code tattooed to your forehead? Don't\n> like the idea? Then think about having to care about a uuid in your\n> projects. Same deal.\n>\n> Nobody is going to associate themselves with a uuid. It's not how humans\n> work. It's degrading, and it's work-for-no-gain to anybody who doesn't\n> have OCD.\n>\n> So in practice, the only thing that would happen is that people make up\n> random uuid's and they'd be different for every single machine they have,\n> because absolutely NOBODY would ever bother to try to save and move their\n> uuids around.\n>\n> So when you point out that emails aren't unique, or that people change\n> their emails over time, please realize that the emails are _more_ stable\n> than a uuid would ever be. Because an email actually has some emotional\n> attachment to the person in question. Yes, they change. So do real names\n> too (which change more seldom, exactly because people are way _more_\n> emotionally attached to their real names).\n>\n> uuid's? I can pretty much guarantee that for me, it would be different for\n> every single machine I have. Because I could just not be bothered to care.\n>\n>                        Linus\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at  http://vger.kernel.org/majordomo-info.html\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137118","messageId":"alpine.LFD.2.00.1003181203370.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"9e4733911003181202h23b3d702xce8fc0670a994d52@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T19:07:23Z","receivedAt":"2010-03-18T19:07:23Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Jon Smirl wrote:\n> \n> We could hash people emails and then build a .mailmap equivalent thus\n> hiding their identity.\n\nSo? Why? What's the advantage?\n\nI literally _only_ see disadvantages to the whole thing. If the uuid has \nsome meaning (ie it's related to actual _real_ information), then it is \nnothing but a really inconvenient placeholder for the real information, \nadn another source of new problems (like \"how do we know they are in \nsync? I edit the .gitconfig file by hand all the time\").\n\nAnd if it doesn't have meaning, then it's just annoying and will never \never be attached to anything relevant long-term.\n\nEither way, there are only downsides, no upsides. There is absolutely _no_ \nway that teh uuid would ever actually encode any real meaningful \ninformation that isn't better represented by the name/email.\n\n\t\t\tLinus\n"},{"id":"137119","messageId":"alpine.LFD.2.00.1003181454160.31128@xanadu.home","threadId":"23060","inReplyTo":"4ba27424.9804cc0a.4b1f.11b1@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-18T19:12:10Z","receivedAt":"2010-03-18T19:12:10Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Michael Witten wrote:\n\n> So, forget the original generality and let's\n> define the uuid as a SHA-1 of some EASILY\n> REMEMBERED, already reasonably unique piece of\n> information such as an old (name,email) pair.\n\nEven with _that_, I bet many people will simply no bother. You may as \nwell just use your current name and email address. Oh wait, Git is \nusing just that already.\n\n> To make life easier on people, git tools could automate\n> that process; to Junio, his just uuid is an old,\n> unchanging (name,email) pair:\n> \n> $ git config --global user.name \"Junio C Hamano\"\n> $ git config --global user.email \"gitster@pobox.com\"\n> $ git config --global --uuid \"Junio C Hamano \"\n> \n> which produces something like:\n> \n> [user]\n> name = Junio C Hamano\n> email = gitster@pobox.com\n> uuid = 6e99d26860f0b87ef4843fa838df2a918b85d1f7\n\nEven then, some people _will_ manage to screw up some of their UUID \nconfigs. And you'll inevitably end up in the same situation that we \nhave today i.e. different identification credentials that have to be \nmapped to the same individual.\n\n> Could people still bungle the uuid or enter trash?\n> Sure, but that's essentially no different than the\n> current situation.\n\nExact. So why bother?\n\n> This would be an improvement, because at least some people would take \n> advantage of it; in fact, I bet most people would use it properly \n> because:\n[...]\n\nMost people _already_ use their name/email configuration properly. And \nthose who really care are managing a stable email address already. so \nthis is not an improvement at all but only some added complexity.\n\n> Moreover, storing and using the SHA-1 uuid would be\n> very efficient and allow for saner .mailmap hacks.\n\nI don't buy that either. If anything, it is way better to fix the \ncurrent .mailmap mechanism to catter for changing email addresses. \nThat's what people use to contact people anyway as I doubt you could \nsend any congratulations or job offers solely by using the Git's UUID. \nSo you must link back to some form of email address in the end, and \npreferably the current one, otherwise the UUID is useless. In that case \nthen why not simply using that email address in the first place?\n\nThe real solution is actually to improve the .mailmap so that any \nindividual could decide that for this or that name/email pair to be \nfound in the repository then here's the current email that should be \ndisplayed instead. Currently this applies partially and only to \ngit-shortlog.\n\n\nNicolas\n"},{"id":"137121","messageId":"9e4733911003181216m59a09ad6y6c7c890714d0d20b@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181203370.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T19:16:43Z","receivedAt":"2010-03-18T19:16:43Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 3:07 PM, Linus Torvalds\n wrote:\n>\n>\n> On Thu, 18 Mar 2010, Jon Smirl wrote:\n>>\n>> We could hash people emails and then build a .mailmap equivalent thus\n>> hiding their identity.\n>\n> So? Why? What's the advantage?\n\nI happen to think that the concept of privacy and working on an open\nsource project are fairly incompatible. But apparently their are\npeople who think otherwise. The use would be to reconstruct that\nmailmap I made, but with the email addresses replaced with SHA1 hashes\nof the email. No human would use the SHA1s, they're just there to\nobscure the emails.\n\n>\n> I literally _only_ see disadvantages to the whole thing. If the uuid has\n> some meaning (ie it's related to actual _real_ information), then it is\n> nothing but a really inconvenient placeholder for the real information,\n> adn another source of new problems (like \"how do we know they are in\n> sync? I edit the .gitconfig file by hand all the time\").\n>\n> And if it doesn't have meaning, then it's just annoying and will never\n> ever be attached to anything relevant long-term.\n>\n> Either way, there are only downsides, no upsides. There is absolutely _no_\n> way that teh uuid would ever actually encode any real meaningful\n> information that isn't better represented by the name/email.\n>\n>                        Linus\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137120","messageId":"alpine.LFD.2.00.1003181218400.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"9e4733911003181216m59a09ad6y6c7c890714d0d20b@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T19:20:10Z","receivedAt":"2010-03-18T19:20:10Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Jon Smirl wrote:\n> \n> I happen to think that the concept of privacy and working on an open\n> source project are fairly incompatible. But apparently their are\n> people who think otherwise. The use would be to reconstruct that\n> mailmap I made, but with the email addresses replaced with SHA1 hashes\n> of the email. No human would use the SHA1s, they're just there to\n> obscure the emails.\n\nI really see that as a bad thing, not a good thing. It's like enabling \nsome crazy shit and making it official.\n\nIf you don't want to reveal your real name, use a fake address. Just don't \nexpect anybody to want to work with you. \n\nThe LAST thing we want is built-in git support for doing f*cking stupid \nthings. You can do stupid things with it on your own without us helping \nand encouraging you.\n\n\t\tLinus\n"},{"id":"137123","messageId":"b4087cc51003181232q6b8ec056lad79c62926b65a10@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181203370.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T19:32:38Z","receivedAt":"2010-03-18T19:32:38Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 14:07, Linus Torvalds\n wrote:\n\n> And if it doesn't have meaning, then it's just\n> annoying and will never ever be attached to\n> anything relevant long-term.\n\nYou've actually just described the current name/email system.\n\nWhat a uuid provides is that very property of long-term attachment; a\ngit user can change the name/email pair but keep the same uuid.\n\nYou see, the problem is that the name/email pair isn't really an\nidentifier; it's actually just info about the user's current email\naccount, which is very useful for everyday workflow, but pretty naive\nfor historical identification over long periods of time.\n\nAs previously discussed in my original email, the 'email' portion of\nthe name/email pair is the most volatile portion, and that's because\nit's only tangentially related to identity (and it certainly has\nnothing to do with long-term identity).\n\n>There is absolutely _no_ way that teh uuid would\n> ever actually encode any real meaningful\n> information that isn't better represented by the\n> name/email.\n\nIt IS a name/email pair (if you want or if that is enforced); it's\njust one that isn't as volatile.\n\nThis notion of a uuid is an attempt to adopt a BETTER MODEL for\nidentity: The user get's to choose a piece of information that he\nhimself deems a longterm identifier; it's not about what address you\ncurrently use for email, it's solely about who you are over a long\nperiod of time.\n"},{"id":"137125","messageId":"9e4733911003181237i56a00a78k31fac2c1afe5a674@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181218400.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T19:37:58Z","receivedAt":"2010-03-18T19:37:58Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 3:20 PM, Linus Torvalds\n wrote:\n>\n>\n> On Thu, 18 Mar 2010, Jon Smirl wrote:\n>>\n>> I happen to think that the concept of privacy and working on an open\n>> source project are fairly incompatible. But apparently their are\n>> people who think otherwise.  The use would be to reconstruct that\n>> mailmap I made, but with the email addresses replaced with SHA1 hashes\n>> of the email. No human would use the SHA1s, they're just there to\n>> obscure the emails.\n>\n> I really see that as a bad thing, not a good thing. It's like enabling\n> some crazy shit and making it official.\n>\n> If you don't want to reveal your real name, use a fake address. Just don't\n> expect anybody to want to work with you.\n\nGo ahead and commit that .mailmap I made. It really cleans up the\nstatistics by fixing 500 errors is people's names. Just don't point\nthe ensuing flame war at me, your hide is tougher.\n\n> The LAST thing we want is built-in git support for doing f*cking stupid\n> things.  You can do stupid things with it on your own without us helping\n> and encouraging you.\n>\n>                Linus\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137128","messageId":"alpine.LFD.2.00.1003181237450.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"b4087cc51003181232q6b8ec056lad79c62926b65a10@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T19:40:07Z","receivedAt":"2010-03-18T19:40:07Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Michael Witten wrote:\n> \n> What a uuid provides is that very property of long-term attachment; a\n> git user can change the name/email pair but keep the same uuid.\n\nI don't think you understand what \"attachment\" means.\n\nThink about your wife, your kids, or your pet. THAT is attachment.\n\nRandom 16-letter letter-jumble? No. People will _never_ care. They'll \nsimply not care. \n\nIt's true that people _already_ don't care too much about their emails, \nand that typos and simply job changes (or annoying ISP's) will change \nthem. But that would be orders of magnitude _worse_ with something like a \nuuid.\n\n> It IS a name/email pair (if you want or if that is enforced); it's\n> just one that isn't as volatile.\n\nDon't be an idiot.\n\nTry to think like a HUMAN. Not a computer scientist. And ponder.\n\nIt's a _social_ issue, not a \"let's tattoo this uuid on everybody\".\n\n\t\tLinus\n"},{"id":"137135","messageId":"4B445491-99C8-49CD-910F-7F3CC1C67984@wincent.com","threadId":"23060","inReplyTo":"b4087cc51003181232q6b8ec056lad79c62926b65a10@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Wincent Colaiuta","fromEmail":"win@wincent.com","sentAt":"2010-03-18T19:40:51Z","receivedAt":"2010-03-18T19:40:51Z","isPatch":false,"sender":{"key":"greg@hurrell.net","avatar":"https://avatars.githubusercontent.com/u/7074?v=4"},"body":"El 18/03/2010, a las 20:32, Michael Witten escribió:\n\n> On Thu, Mar 18, 2010 at 14:07, Linus Torvalds\n> wrote:\n> \n>> And if it doesn't have meaning, then it's just\n>> annoying and will never ever be attached to\n>> anything relevant long-term.\n> \n> You've actually just described the current name/email system.\n> \n> What a uuid provides is that very property of long-term attachment; a\n> git user can change the name/email pair but keep the same uuid.\n> \n> You see, the problem is that the name/email pair isn't really an\n> identifier; it's actually just info about the user's current email\n> account, which is very useful for everyday workflow, but pretty naive\n> for historical identification over long periods of time.\n\nThis whole thing is a stupid idea.\n\nIf users can't even be bothered keeping a stable email address, what makes you think that they can be assed \"doing the right thing\" with respect to a meaningless UUID string?\n\nThe idea is complicated, over-engineered, brings no benefit and adds only cruft.\n\nW\n"},{"id":"137126","messageId":"46a038f91003181242ifb3d0bao553c5be5e6a4f25c@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003181232q6b8ec056lad79c62926b65a10@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Martin Langhoff","fromEmail":"martin.langhoff@gmail.com","sentAt":"2010-03-18T19:42:25Z","receivedAt":"2010-03-18T19:42:25Z","isPatch":false,"sender":{"key":"martin.langhoff@gmail.com","avatar":"https://gravatar.com/avatar/1e3f311b6c4c15836501901ca58f8c0b0667246488084ba524d8bc9867e22fd9?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 3:32 PM, Michael Witten wrote:\n> On Thu, Mar 18, 2010 at 14:07, Linus Torvalds\n> wrote:\n>\n>> And if it doesn't have meaning, then it's just\n>> annoying and will never ever be attached to\n>> anything relevant long-term.\n>\n> You've actually just described the current name/email system.\n\nWTH are you drinking? I have been using my current name and email\naccounts for many years.\n\nThey are useful for git and for some things that are even more useful\n-- like addressing emails! My best CV is googing for my name / email\naddresses -- it will show you my professional career. Including the\ntime that Linus called my patch \"idiotic\" :-)\n\nSo, these things are attached to something meaningful: my long term\npersonal identity. A git-only \"uuid\"? Screw that, I hack on too many\nphysically different machines, I am not going to be carrying around a\nmagic string.\n\ncheers,\n\n\nm\n-- \n martin.langhoff@gmail.com\n martin@laptop.org -- School Server Architect\n - ask interesting questions\n - don't get distracted with shiny stuff - working code first\n - http://wiki.laptop.org/go/User:Martinlanghoff\n"},{"id":"137129","messageId":"b4087cc51003181247m26125efft3d03cd34a2dc9280@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181237450.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T19:47:27Z","receivedAt":"2010-03-18T19:47:27Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 14:40, Linus Torvalds\n wrote:\n> Random 16-letter letter-jumble? No. People will _never_ care. They'll\n> simply not care.\n\nI don't think you've read one word that I've written.\n"},{"id":"137130","messageId":"alpine.LFD.2.00.1003181241080.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"9e4733911003181237i56a00a78k31fac2c1afe5a674@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T19:47:48Z","receivedAt":"2010-03-18T19:47:48Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Jon Smirl wrote:\n> \n> Go ahead and commit that .mailmap I made. It really cleans up the\n> statistics by fixing 500 errors is people's names. Just don't point\n> the ensuing flame war at me, your hide is tougher.\n\nHow hard is it to understand the notion of \"people just don't _care_ \nenough\"?\n\nLook at CVS. Look at three _decades_ of CVS. Then look at the \n\"identifiers\" that thing used. \n\nGit is much better. Git is better for two reasons:\n\n - We allow/encourage people to use way more meaningful identifiers\n\n - Exactly _because_ what we use is meaningful to people, most people \n bother to try.\n\nAnd you don't seem to understand that whole \"meaningful\" part. If you \ndon't have the social understanding of how people actually _work_, then \nnothing I say can explain it.\n\nLet me try one more time: do the statistics on \"committer information\" vs \n\"author information\" on the Linux kernel repository, and count the types \nof errors that happen. I can explain the errors and why they happen, and \nit has everything to do with how _humans_work_ (*).\n\nIf you don't understand that, then there's no point in arguing.\n\n\t\t\tLinus\n\n(*) I'll give you one answer in the next email. But before you read that \nemail, try to think about it, and see if you can guess at patterns.\n"},{"id":"137131","messageId":"alpine.LFD.2.00.1003181248040.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181241080.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T19:50:14Z","receivedAt":"2010-03-18T19:50:14Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Linus Torvalds wrote:\n> \n> (*) I'll give you one answer in the next email. But before you read that \n> email, try to think about it, and see if you can guess at patterns.\n\nLookie here:\n\n [torvalds@i5 linux]$ git log --pretty=full | grep '^Commit: ' | sort | uniq -c | sort -n | grep localdomain\n 1 Commit: Jeff Garzik \n 2 Commit: Dave Airlie \n 3 Commit: James Bottomley \n 3 Commit: James Morris \n 3 Commit: James Morris \n 4 Commit: James Bottomley \n 32 Commit: Thomas Graf \n 410 Commit: James Bottomley \n [torvalds@i5 linux]$ git log --pretty=full | grep '^Author: ' | sort | uniq -c | sort -n | grep localdomain\n 1 Author: Alex Deucher \n 1 Author: Dave Airlie \n 1 Author: Eduardo Habkost \n 1 Author: Grzegorz Nosek \n 1 Author: Izik Eidus \n 1 Author: Jeff Garzik \n 2 Author: Esti Kummer \n 2 Author: James Bottomley \n 3 Author: Dave Airlie \n 3 Author: James Bottomley \n 3 Author: James Bottomley \n 4 Author: Cindy H Kao \n 4 Author: Kristian Høgsberg \n\nSee? Mistakes happen. But look at what happens to the committer \ninformation? Think about it. Really _think_ about it. There is absolutely \nzero _technical_ difference between the two fields. The only difference is \nthat \"git log\" by default shows one, and not the other.\n\nSo as a human, which one do you think people care about and fix more \nquickly?\n\nAnd look at the numbers once more.\n\n\t\t\tLinus\n"},{"id":"137132","messageId":"alpine.LFD.2.00.1003181250200.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"b4087cc51003181247m26125efft3d03cd34a2dc9280@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T19:52:29Z","receivedAt":"2010-03-18T19:52:29Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Michael Witten wrote:\n\n> On Thu, Mar 18, 2010 at 14:40, Linus Torvalds\n> wrote:\n> > Random 16-letter letter-jumble? No. People will _never_ care. They'll\n> > simply not care.\n> \n> I don't think you've read one word that I've written.\n\nOh, I read them. They make no sense.\n\nIf the uuid isn't random, but tied to the email address, then it's \nworthless. \n\nIf you like the random 16-letter jumbles, then for christ sake JUST CHANGE \n\"git log\" to hash the author name for you. You'll get the uuid's. What I'm \ntelling you is that NOBODY SANE WANTS TO EVER SEE THEM.\n\nAnd if nobody wants them, then nobody will maintain them, and they'll be \nmuch _less_ useful than the emails we already have.\n\n\t\tLinus\n"},{"id":"137137","messageId":"EE13AB7A-1AE0-4741-AB3F-F8880D2BD385@wincent.com","threadId":"23060","inReplyTo":"b4087cc51003181247m26125efft3d03cd34a2dc9280@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Wincent Colaiuta","fromEmail":"win@wincent.com","sentAt":"2010-03-18T19:52:42Z","receivedAt":"2010-03-18T19:52:42Z","isPatch":false,"sender":{"key":"greg@hurrell.net","avatar":"https://avatars.githubusercontent.com/u/7074?v=4"},"body":"El 18/03/2010, a las 20:47, Michael Witten escribió:\n\n> On Thu, Mar 18, 2010 at 14:40, Linus Torvalds\n> wrote:\n>> Random 16-letter letter-jumble? No. People will _never_ care. They'll\n>> simply not care.\n> \n> I don't think you've read one word that I've written.\n\nOn the contrary I get the impression he has waded through everything you've written, and has even been patient enough to put together (now several) replies explaining exactly why it is a misguided idea.\n\nNow it's time for you to read and actually reflect on what's been said. If you're sane and have a modicum of intelligence you'll come to the conclusion that your idea doesn't solve any problem, and in fact only adds a bunch of new ones.\n\nW\n"},{"id":"137133","messageId":"b4087cc51003181300x5aab9033yde20bff12b7ff289@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181250200.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T20:00:21Z","receivedAt":"2010-03-18T20:00:21Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 14:52, Linus Torvalds\n wrote:\n>\n> If you like the random 16-letter jumbles, then for christ sake JUST CHANGE\n> \"git log\" to hash the author name for you. You'll get the uuid's. What I'm\n> telling you is that NOBODY SANE WANTS TO EVER SEE THEM.\n\nNo, I'm reasonably certain you didn't.\n"},{"id":"137134","messageId":"alpine.LFD.2.00.1003181256310.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181248040.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T20:01:51Z","receivedAt":"2010-03-18T20:01:51Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Linus Torvalds wrote:\n> \n> So as a human, which one do you think people care about and fix more \n> quickly?\n\nBtw, one other thing you can take away from it is that even when they \n_are_ shown, and even when they _are_ meaningful, people still don't care. \n\nThere's absolutely tons of \"(none)\" emails even in the _visible_ parts, \nwhich is really really sad. But it does tell a lot about humans - they \nwon't be noticing even _obvious_ mistakes like that.\n\n(And yes, it does say that git should probably have errored out way more \naggressively about badly set up host/domain names in the \"guess at email \naddress\" code. My bad. Maybe it's still worth fixing for the future)\n\n\t\t\tLinus\n"},{"id":"137136","messageId":"3f4fd2641003181331t4e70bed0yf1d03e20d1127705@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181248040.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Reece Dunn","fromEmail":"msclrhd@googlemail.com","sentAt":"2010-03-18T20:31:08Z","receivedAt":"2010-03-18T20:31:08Z","isPatch":false,"sender":{"key":"msclrhd@googlemail.com","avatar":null},"body":"On 18 March 2010 19:50, Linus Torvalds wrote:\n>\n> On Thu, 18 Mar 2010, Linus Torvalds wrote:\n>>\n>> (*) I'll give you one answer in the next email. But before you read that\n>> email, try to think about it, and see if you can guess at patterns.\n>\n> Lookie here:\n>\n>  [torvalds@i5 linux]$ git log --pretty=full | grep '^Commit: ' | sort | uniq -c | sort -n | grep localdomain\n>      1 Commit: Jeff Garzik \n>      2 Commit: Dave Airlie \n>      3 Commit: James Bottomley \n>      3 Commit: James Morris \n>      3 Commit: James Morris \n>      4 Commit: James Bottomley \n>     32 Commit: Thomas Graf \n>    410 Commit: James Bottomley \n>  [torvalds@i5 linux]$ git log --pretty=full | grep '^Author: ' | sort | uniq -c | sort -n | grep localdomain\n>      1 Author: Alex Deucher \n>      1 Author: Dave Airlie \n>      1 Author: Eduardo Habkost \n>      1 Author: Grzegorz Nosek \n>      1 Author: Izik Eidus \n>      1 Author: Jeff Garzik \n>      2 Author: Esti Kummer \n>      2 Author: James Bottomley \n>      3 Author: Dave Airlie \n>      3 Author: James Bottomley \n>      3 Author: James Bottomley \n>      4 Author: Cindy H Kao \n>      4 Author: Kristian Høgsberg \n>\n> See? Mistakes happen. But look at what happens to the committer\n> information? Think about it. Really _think_ about it. There is absolutely\n> zero _technical_ difference between the two fields. The only difference is\n> that \"git log\" by default shows one, and not the other.\n>\n> So as a human, which one do you think people care about and fix more\n> quickly?\n>\n> And look at the numbers once more.\n\nSo... going back to the original problem, we have:\n\n 1/ people making mistakes in the commit logs for whatever reason\n(e.g. re-installation or a new computer);\n 2/ people changing name (e.g. getting married) or changing email\n(e.g. gmail.com to googlemail.com).\n\nThe problem is that it may be beneficial to see all the changes Cindy\nH Kao made for example, including the ones made\n@localhost.localdomain.\n\nHaving (user, email, uuid) will not solve the problem (even if the\nuuid is from a memorable string) -- consider case 1. If you forget to\nsetup git, uuid will be blank or some random data, so this will be\nworse than the (user, email) identity. As noted, there is also the\nissue that git is used in a lot of places and not all git clone\ninstances are running the same version (e.g. pushing to an older git\nclient that does not support this new data).\n\nWhat would be better is having a concept of identity aliases. That is,\na user can say that (for this git project), (user1,email1) is the same\nperson as (user2,email2). This would allow someone who has\nmis-configured their git instance to say what the (user,email) pair\nshould have been. It also allows people to say that they used to be\ncalled someone and they are now called somebody.\n\nThis information should ideally be in some form of (user,email) ->\n(user,email) map that is versioned and tracked by git (in a way that\nis also backward compatible, which could be tricky).\n\nIt also needs to be changeable and version tracked (i.e. with history)\nto allow people to undo this; for example, this system would allow me\nto say that Linus' (user,email) id is actually an alias for my\n(user,email) which is bad. I don't know of a decent way to prevent\nthis (or someone using the uuid of someone else in the original\nproposal), but this approach would at least allow it to be corrected.\n\nThere will need to be the related plumbing and porcelain to access and\nmanipulate this data/meta-data.\n\nWould this be a better approach? Or is there a fatal flaw I am missing\n(like people being able to alias themselves as other people, for\nexample)?\n\n- Reece\n"},{"id":"137139","messageId":"20100318204416.GM8256@thunk.org","threadId":"23060","inReplyTo":"4ba27424.9804cc0a.4b1f.11b1@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"","fromEmail":"tytso@mit.edu","sentAt":"2010-03-18T20:44:16Z","receivedAt":"2010-03-18T20:44:16Z","isPatch":false,"sender":{"key":"tytso@mit.edu","avatar":"https://avatars.githubusercontent.com/u/51416?v=4"},"body":"On Thu, Mar 18, 2010 at 11:42:44AM -0700, Michael Witten wrote:\n> Could people still bungle the uuid or enter trash?\n> Sure, but that's essentially no different than the\n> current situation. This would be an improvement,\n> because at least some people would take advantage\n> of it; in fact, I bet most people would use it\n> properly because:\n> \n> * The information required is easily remembered\n> and reproduced; it has that emotional aspect.\n> \n> * People have an emotional attachment to getting\n> proper attribution for their work, and this\n> helps.\n\nThe problem is that people don't get emotionally attached to a UUID.\nAnd even if the UUID is generated algorithmically, they need to\nremember, gee, was my UUID generated using:\n\n\tTheodore Y. Ts'o \n\tTheodore Tso \n\tTheodore T'so (*) \n\tTheodore Y Tso \n\tTheodore Tso \n\t\n\n(*) The VA Linux folks screwed up where the apostrophe goes in some\npress release, and the mispelling of my last name has followed me for\nthe last ten years since then.\n\nMore importantly, there's a lot more to someone's reputation than just\nGit. What about reviews of other people's patches on LKML? Can you\n**honestly** expect people to say,\n\n Hi, I'm and here are the things\n that you need to fix with your patch....\n\nPeople who give thoughtful reviews of other people's code count for a\nlot, and people are not going to track that sort of thing by UUID.\nThey are going to track it by name and e-mail address.\n\nOr what about papers? Can you honestly expect that it would matter\neven one iota if someone put in a bibliography of a paper\n\nR. Card (14a8da4b-0231-497b-aa66-1809cc9727f9), T. Y. Ts'o\n(dd1b51a1-ce2a-41fd-ae89-f68b7f0ace85), and S. Tweedie\n(9052e458-32cc-11df-93b8-0016eb0fac40), \"Design and implementation of\nthe second extended filesystem,\" in Proceedings of the 1994 Amsterdam\nLinux Conference, 1994.\n\nIs that going to contribute to my identity any? I don't think so.\n\n\nFinally, if someone misses one of my commits in a git changelog, so\nwhat? People don't guage impact by the number of commits. There are\nsome people who have huge numbers commits, but they are all spelling\ncorrections. A developer's reputation is developed over many months\nor years of contributions; of interactions over e-mail; interactions\nin hallway discussions at conferences; papers which they author; etc.\nIt's not just about git commits.\n\n \t \t \t \t \t \t - Ted\n"},{"id":"137141","messageId":"alpine.LFD.2.00.1003181348250.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"3f4fd2641003181331t4e70bed0yf1d03e20d1127705@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T20:59:14Z","receivedAt":"2010-03-18T20:59:14Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Reece Dunn wrote:\n> \n> What would be better is having a concept of identity aliases. That is,\n> a user can say that (for this git project), (user1,email1) is the same\n> person as (user2,email2). This would allow someone who has\n> mis-configured their git instance to say what the (user,email) pair\n> should have been. It also allows people to say that they used to be\n> called someone and they are now called somebody.\n\nYeah. And that's what '.mailmap' is, really.\n\nDoes mailmap get annoying? Yes. Is it going to be incomplete? Yes. Do we \never even _bother_ to try to make it perfect? No.\n\nIn the kernel, for example, we tend to use it _only_ to fix up the real \nname. It's much more capable than that (ie you can use it to fix up email \naddresses too), but we literally haven't cared enough to bother. So you \nstill see the \"localhost\" emails or the \"(none)\" domains - even if you use \none of the formats that ask for a \"fixed\" name and email.\n\nAnd git itself only fixes up names for certain commands (git blame, git \nshortlog) and with specific format specifiers (%aN and %aE).\n\nThe _default_ pretty log format printouts don't do it, for example. Should \nthey? Maybe. Or maybe we should have a flag and/or config option to do so \nby default.\n\n\t\t\t\tLinus\n"},{"id":"137142","messageId":"b4087cc51003181412s78d153c8t48984eb87eb75e0e@mail.gmail.com","threadId":"23060","inReplyTo":"20100318204416.GM8256@thunk.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T21:12:48Z","receivedAt":"2010-03-18T21:12:48Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 15:44, wrote:\n>   Hi, I'm and here are the things\n>   that you need to fix with your patch....\n\nLook, there is a huge misunderstanding.\n\nThis is all that I'm saying: Keep git exactly the way it is, but add\none extra piece of identifying information for each person.\n\nThat's it.\n\nNothing is being taken away.\n\nYou can still see/grep/access the full names and email addresses just\nas before, only now there will be another piece of information on\nwhich to filter (or ignore it if you want).\n\nIn the most general form of my proposal, the idea is to let the user\nchoose some piece of information that he himself deems to be uniquely\nidentifying over a long period of time. However, I think it would be\nsmart to reduce that information to a SHA-1 (at least when it's\nrecorded in, say, a commit).\n\nEssentially, the goal is to distribute the task of maintaining aliases.\n"},{"id":"137143","messageId":"46a038f91003181419n5d076bdbv18e32e292d7937a8@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003181412s78d153c8t48984eb87eb75e0e@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Martin Langhoff","fromEmail":"martin.langhoff@gmail.com","sentAt":"2010-03-18T21:19:19Z","receivedAt":"2010-03-18T21:19:19Z","isPatch":false,"sender":{"key":"martin.langhoff@gmail.com","avatar":"https://gravatar.com/avatar/1e3f311b6c4c15836501901ca58f8c0b0667246488084ba524d8bc9867e22fd9?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 5:12 PM, Michael Witten wrote:\n> This is all that I'm saying: Keep git exactly the way it is, but add\n> one extra piece of identifying information for each person.\n\nWhat's the value? For me it'll be \"Martin Langhoff\". I already have that.\n\n> Nothing is being taken away.\n\nBut something is added.\n\nGood design is not when there's nothing more to add, it's when there's\nnothing left _to remove_.\n\nGit is what it is thanks to removing superfluous crud from its core\ndatamodel. Don't be surprised that there is a very strong resistance\nto adding anything to that datamodel.\n\n> Essentially, the goal is to distribute the task of maintaining aliases.\n\nAlready achieved with mailcap. No need to mess with the secret of\ngit's success (the tight datamodel).\n\ncheers,\n\n\n\nm\n-- \n martin.langhoff@gmail.com\n martin@laptop.org -- School Server Architect\n - ask interesting questions\n - don't get distracted with shiny stuff - working code first\n - http://wiki.laptop.org/go/User:Martinlanghoff\n"},{"id":"137145","messageId":"alpine.LFD.2.00.1003181411400.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"b4087cc51003181412s78d153c8t48984eb87eb75e0e@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-18T21:27:24Z","receivedAt":"2010-03-18T21:27:24Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Michael Witten wrote:\n> \n> This is all that I'm saying: Keep git exactly the way it is, but add\n> one extra piece of identifying information for each person.\n\nThe thing is, you don't seem to realize that most authorship is over \nemail.\n\nLet's take some numbers from the kernel archive, for example. Here's _one_ \ntrivial way to count it:\n\n - number of commits where author/committer email matches (presumably \n _not_ emailed, although sometimes people commit their own patches that \n were emailed around):\n\n\t[torvalds@i5 linux]$ git log --no-merges \"--pretty=format:%h-%ae%n%h-%ce\" | uniq -d | wc\n\t 33473 33473 959167\n\n - total number of commits:\n\n\t[torvalds@i5 linux]$ git rev-list --no-merges HEAD | wc\n\t 176415 176415 7233015\n\nIOW, less than a fifth of the patches were done by the person who actually \ncommitted things. 80%+ of all changes were committed by somebody else than \nthe author.\n\nHow do you think the authorship information can be transferred sanely, \nconsidering that the author didn't even use git in the first place? \nReally?\n\nThat's where the typos/mistakes/missing-info really happens. And it often \nstarts out with incomplete information, because the person has a bad email \nsetup, and the thing only has an email address to begin with, ie the \n\"From:\" might literally say just \"tytso@mit.edu\" or something (to pick an \nexample from the Cc list in this discussion - when Ted sends real emails, \nthey tend to have proper naming).\n\nSometimes we'll edit the messages to have the \"From: xyz \" thing at \nthe top, fixing up the incomplete thing then. Typos happen there. Or the \npatch will simply come in two different ways, so there's no typo, yet \nthere are two different emails that get author attribution.\n\nThe thing is, development really is about human interaction. Yes, there's \na tool involved (git), and once the data is in the tool we won't lose it \nany more, but this is about getting the data _into_ the tool in the first \nplace.\n\nAnd the data you want to add simply DOES NOT EXIST. And we can't make it \nexist. The fact that even the trivial and obvious data that git _does_ ask \nfor gets to be incomplete should tell you something.\n\n\t\t\tLinus\n"},{"id":"137144","messageId":"b4087cc51003181429v2998b95ei3f08360e6d2a5aa7@mail.gmail.com","threadId":"23060","inReplyTo":"46a038f91003181419n5d076bdbv18e32e292d7937a8@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T21:29:28Z","receivedAt":"2010-03-18T21:29:28Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 16:19, Martin Langhoff\n wrote:\n> On Thu, Mar 18, 2010 at 5:12 PM, Michael Witten wrote:\n>> This is all that I'm saying: Keep git exactly the way it is, but add\n>> one extra piece of identifying information for each person.\n>\n> What's the value? For me it'll be \"Martin Langhoff\". I already have that.\n\nWell, that's rather egotistical considering you're probably not the\nonly Martin Langhoff in this world. I'd advocate something like\n\"Martin Langhoff \".\n\nAt worst, things will be just like they have always been.\n\nMost likely, all that will happen is identification entropy won't\nincrease nearly so rapidly and there might be other benefits such as\nshortlog speed improvements.\n"},{"id":"137146","messageId":"46a038f91003181439lc343dafl6b9321a0b620de84@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003181429v2998b95ei3f08360e6d2a5aa7@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Martin Langhoff","fromEmail":"martin.langhoff@gmail.com","sentAt":"2010-03-18T21:39:02Z","receivedAt":"2010-03-18T21:39:02Z","isPatch":false,"sender":{"key":"martin.langhoff@gmail.com","avatar":"https://gravatar.com/avatar/1e3f311b6c4c15836501901ca58f8c0b0667246488084ba524d8bc9867e22fd9?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 5:29 PM, Michael Witten wrote:\n> On Thu, Mar 18, 2010 at 16:19, Martin Langhoff\n>> What's the value? For me it'll be \"Martin Langhoff\". I already have that.\n>\n> Well, that's rather egotistical considering you're probably not the\n> only Martin Langhoff in this world. I'd advocate something like\n> \"Martin Langhoff \".\n\nSo you are saying we should change the core datamodel of git to say...\nwhat we already can say?\n\n> At worst, things will be just like they have always been.\n\nNo, we'll have another way to have data mismatches. There are _more_\nmoving parts in your model. That's what Linus is pointing out.\n\nThis is a case where an ancillary \"fixup table\", in the form of\nmailmap, works best. Don't move the fixup table to the core of the\ndatamodel, it just doesn't belong there.\n\nHere's a hint: using your \"uuid\" model, I'll get some commits into a\nproject with the wrong uuid. Because I made a typo, or changed\nmachines (and a random uuid got created), whatever reason. So now in\nmy project I appear under 2 uuids.\n\nWhat should we do in that case? Use mailmap to map the stray uuid to\nthe \"real\" one?... Have we done a lot of work to get back to square 0?\n\ncheers,\n\n\nm\n-- \n martin.langhoff@gmail.com\n martin@laptop.org -- School Server Architect\n - ask interesting questions\n - don't get distracted with shiny stuff - working code first\n - http://wiki.laptop.org/go/User:Martinlanghoff\n"},{"id":"137147","messageId":"b4087cc51003181444i7b7dea5bgc8f4f43b365526f4@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181411400.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T21:44:06Z","receivedAt":"2010-03-18T21:44:06Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 16:27, Linus Torvalds\n wrote:\n>\n> On Thu, 18 Mar 2010, Michael Witten wrote:\n>>\n>> This is all that I'm saying: Keep git exactly the way it is, but add\n>> one extra piece of identifying information for each person.\n>\n> The thing is, you don't seem to realize that most authorship is [sent\n> over email with incomplete information].\n\nThat is a really good point, and something I'll have to consider more\nthoroughly.\n\nHowever, I do NOT claim that my proposal will add information where\nthere is none, only that it will reduce the rate at which entropy\nincreases.\n"},{"id":"137148","messageId":"b4087cc51003181446r6bd89371q9f2e7a14bfa4d557@mail.gmail.com","threadId":"23060","inReplyTo":"46a038f91003181439lc343dafl6b9321a0b620de84@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T21:46:50Z","receivedAt":"2010-03-18T21:46:50Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 16:39, Martin Langhoff\n wrote:\n>\n> Here's a hint: using your \"uuid\" model, I'll get some commits into a\n> project with the wrong uuid. Because I made a typo, or changed\n> machines (and a random uuid got created), whatever reason. So now in\n> my project I appear under 2 uuids.\n>\n> What should we do in that case? Use mailmap to map the stray uuid to\n> the \"real\" one?... Have we done a lot of work to get back to square 0?\n\nAgain:\n\n>> At worst, things will be just like they have always been.\n>>\n>> Most likely, all that will happen is identification entropy won't\n>> increase nearly so rapidly and there might be other benefits\n>> such as shortlog speed improvements.\n"},{"id":"137153","messageId":"46a038f91003181455u7136e011u5352c996384f086d@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003181446r6bd89371q9f2e7a14bfa4d557@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Martin Langhoff","fromEmail":"martin.langhoff@gmail.com","sentAt":"2010-03-18T21:55:08Z","receivedAt":"2010-03-18T21:55:08Z","isPatch":false,"sender":{"key":"martin.langhoff@gmail.com","avatar":"https://gravatar.com/avatar/1e3f311b6c4c15836501901ca58f8c0b0667246488084ba524d8bc9867e22fd9?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 5:46 PM, Michael Witten wrote:\n>> What should we do in that case? Use mailmap to map the stray uuid to\n>> the \"real\" one?... Have we done a lot of work to get back to square 0?\n>\n> Again:\n>\n>>> At worst, things will be just like they have always been.\n\nOf course we all read that line. You are proposing a change that will\nmean a flag day -- that is, old versions of git won't be able to read\n\"new\" repositories (and \"new\" git will have to be backwards compat for\nX releases...). This is major breakage.\n\nInflict a painful change on our userbase for... what exactly? Ah, \"At\nworst, things will be just like they have always been.\"\n\nI don't think you understand what you've been proposing.\n\nIs it clearer now why you get a clear \"no\" from all quarters? Huge\ncost, no upside?\n\n\n\nm\n-- \n martin.langhoff@gmail.com\n martin@laptop.org -- School Server Architect\n - ask interesting questions\n - don't get distracted with shiny stuff - working code first\n - http://wiki.laptop.org/go/User:Martinlanghoff\n"},{"id":"137149","messageId":"b4087cc51003181457s76560b02y84aafbe77f45af87@mail.gmail.com","threadId":"23060","inReplyTo":"46a038f91003181439lc343dafl6b9321a0b620de84@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T21:57:52Z","receivedAt":"2010-03-18T21:57:52Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 16:39, Martin Langhoff\n wrote:\n> On Thu, Mar 18, 2010 at 5:29 PM, Michael Witten wrote:\n>> On Thu, Mar 18, 2010 at 16:19, Martin Langhoff\n>>> What's the value? For me it'll be \"Martin Langhoff\". I already have that.\n>>\n>> Well, that's rather egotistical considering you're probably not the\n>> only Martin Langhoff in this world. I'd advocate something like\n>> \"Martin Langhoff \".\n>\n> So you are saying we should change the core datamodel of git to say...\n> what we already can say?\n\nYou see, Martin, you might want/need to stop using \"Martin Langhoff\n\" as your email account, but there's no\nreason why you can't continue to use it for your UUID.\n\n>> At worst, things will be just like they have always been.\n>\n> No, we'll have another way to have data mismatches. There are _more_\n> moving parts in your model. That's what Linus is pointing out.\n\nMismatches in UUIDs will be the only thing worth worrying about;\nfortunately, UUIDs won't change as frequently because they would be\nrarely typed by git users and they are not subject to changing email\nsystems or changing names.\n"},{"id":"137150","messageId":"b4087cc51003181502y1f339cf8u268fa5388ace098@mail.gmail.com","threadId":"23060","inReplyTo":"46a038f91003181455u7136e011u5352c996384f086d@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T22:02:06Z","receivedAt":"2010-03-18T22:02:06Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 16:55, Martin Langhoff\n wrote:\n> On Thu, Mar 18, 2010 at 5:46 PM, Michael Witten wrote:\n>>> What should we do in that case? Use mailmap to map the stray uuid to\n>>> the \"real\" one?... Have we done a lot of work to get back to square 0?\n>>\n>> Again:\n>>\n>>>> At worst, things will be just like they have always been.\n>\n> Of course we all read that line.\n\nYou missed the other line (probably gmail's fault):\n\nMost likely, all that will happen is identification entropy won't\nincrease nearly so rapidly and there might be other benefits\nsuch as shortlog speed improvements.\n"},{"id":"137151","messageId":"3f4fd2641003181506s7797f01eub2d74e6fd556c990@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003181446r6bd89371q9f2e7a14bfa4d557@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Reece Dunn","fromEmail":"msclrhd@googlemail.com","sentAt":"2010-03-18T22:06:28Z","receivedAt":"2010-03-18T22:06:28Z","isPatch":false,"sender":{"key":"msclrhd@googlemail.com","avatar":null},"body":"On 18 March 2010 21:46, Michael Witten wrote:\n> On Thu, Mar 18, 2010 at 16:39, Martin Langhoff\n> wrote:\n>>\n>> Here's a hint: using your \"uuid\" model, I'll get some commits into a\n>> project with the wrong uuid. Because I made a typo, or changed\n>> machines (and a random uuid got created), whatever reason. So now in\n>> my project I appear under 2 uuids.\n>>\n>> What should we do in that case? Use mailmap to map the stray uuid to\n>> the \"real\" one?... Have we done a lot of work to get back to square 0?\n>\n> Again:\n>\n>>> At worst, things will be just like they have always been.\n>>>\n>>> Most likely, all that will happen is identification entropy won't\n>>> increase nearly so rapidly and there might be other benefits\n>>> such as shortlog speed improvements.\n\nYou have 3 pieces of information that can change by adding uuid instead of 2.\n\nAre people going to remember that they need to set a uuid when\nchecking things into git? Different uuids? Forgetting the key string\nto generate the hash for the uuid?\n\nThe uuid is another source of permutations that will see an increase\nin identity triples. It is also another thing that needs to be stored\nin a commit on disk and in memory, printed out in the shortlog and\nchecked by people.\n\nEven if you generate a SHA-1 hash from a memorable bit of data, the\nresulting hash is not readable. It is something that could cause\ncollisions with partial hashes in treeish queries (does 12ab34 refer\nto a commit, or to a persons uuid?). It is also meaningless to the\nuser: I want to find Ted Ts'o's (I hope I've got the apostrophe in the\ncorrect place) commits - how do I know what uuid refers to his\ncommits? How can I find it out?\n\nIt is just adding more resistance, whereas with a well-configured\n.mailmap I could use one of his known email addresses, something that\nis easy to find and remember.\n\n>From what Linus and others have said, .mailmap is the way to fix name\nand/or email changes. It may need more work to expose it to more\ncommands, but that is the simplest, cleanest and most elegant approach\nto fixing the problem you specified.\n\nWhat about .mailmap does not solve your problem? Is it that it does\nnot work for `git log`? If so, then write a patch to allow `git log`\nto use that information when you specify a certain flag (or pretty\nformat string).\n\nNOTE: It is not just the author/committer that needs to remember/use\nthe uuid - it is people doing analysis on commits, curious people,\nautomated scripts and many others.\n\n- Reece\n"},{"id":"137152","messageId":"4BA2A68C.1040708@gmail.com","threadId":"23060","inReplyTo":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"A Large Angry SCM","fromEmail":"gitzilla@gmail.com","sentAt":"2010-03-18T22:17:48Z","receivedAt":"2010-03-18T22:17:48Z","isPatch":false,"sender":{"key":"gitzilla@gmail.com","avatar":"https://gravatar.com/avatar/354625c442439908ff3dd99757dee330e29e9df7847472384faf7a00add247fb?d=mp&s=160"},"body":"Michael Witten wrote:\n> Short Version:\n> -------------\n> \n> \n> Rather than use a (name,email) pair to identify people, let's use\n> a (uuid,name,email) triplet.\n> \n> The uuid can be any piece of information that a user of git determines\n> to be reasonably unique across space and time and that is intended to\n> be used by that user virtually forever (at least within a project's\n> history).\n> \n> For instance, the uuid could be an OSF DCE 1.1 UUID or the SHA-1 of\n> some easily remembered, already reasonably unique information.\n> \n> This could really help keep identifications clean, and it is rather\n> straightforward and possibly quite efficient.\n> \n> \n> Long Version:\n> ------------\n\n[Much text deleted]\n\nThe formatting of the information in the author & committer fields are a \n_social_ convention (with a little help from the tools).You can actually \nuse this proposed \"feature\" now for your own commits by appending the \nUUID string to you name config setting, environment variable and/or GCOS \nfield today and everything will work. You can even make it a requirement \nfor projects that you control. But don't expect all other projects to do \nso also as they may not care.\n"},{"id":"137154","messageId":"46a038f91003181536ib3b74f8o40603a4cee13d62b@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181022040.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Martin Langhoff","fromEmail":"martin.langhoff@gmail.com","sentAt":"2010-03-18T22:36:40Z","receivedAt":"2010-03-18T22:36:40Z","isPatch":false,"sender":{"key":"martin.langhoff@gmail.com","avatar":"https://gravatar.com/avatar/1e3f311b6c4c15836501901ca58f8c0b0667246488084ba524d8bc9867e22fd9?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 1:27 PM, Linus Torvalds\n wrote:\n> Even shorter version: NO.\n\nOne thing we all forgot to mention here is that even if it was a good\nidea (which it is not), implementing it means a flag day: changing in\nthe pack format, wire protocol and APIs, messing up with compatibility\nwith users of pre-flag-day git, and rippling out to all the GUIs,\nfrontends and integration scripts out there.\n\nA veritable mess that would reberberate for years.\n\nAny proposal that touches the core git datamodel... better implement\nsomething that is outrageously wondrously good and impossible to do\nany other way.\n\nMy guess is that people that parachute into this list and propose\ndatamodel changes haven't thought this aspect through.\n\ncheers,\n\n\nm\n-- \n martin.langhoff@gmail.com\n martin@laptop.org -- School Server Architect\n - ask interesting questions\n - don't get distracted with shiny stuff - working code first\n - http://wiki.laptop.org/go/User:Martinlanghoff\n"},{"id":"137155","messageId":"9e4733911003181612m217e805cl95888b4eb0d36a18@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181411400.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T23:12:02Z","receivedAt":"2010-03-18T23:12:02Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 5:27 PM, Linus Torvalds\n wrote:\n>\n>\n> On Thu, 18 Mar 2010, Michael Witten wrote:\n>>\n>> This is all that I'm saying: Keep git exactly the way it is, but add\n>> one extra piece of identifying information for each person.\n>\n> The thing is, you don't seem to realize that most authorship is over\n> email.\n>\n> Let's take some numbers from the kernel archive, for example. Here's _one_\n> trivial way to count it:\n>\n>  - number of commits where author/committer email matches (presumably\n>   _not_ emailed, although sometimes people commit their own patches that\n>   were emailed around):\n>\n>        [torvalds@i5 linux]$ git log --no-merges \"--pretty=format:%h-%ae%n%h-%ce\" | uniq -d | wc\n>          33473   33473  959167\n>\n>  - total number of commits:\n>\n>        [torvalds@i5 linux]$ git rev-list --no-merges HEAD | wc\n>         176415  176415 7233015\n>\n> IOW, less than a fifth of the patches were done by the person who actually\n> committed things. 80%+ of all changes were committed by somebody else than\n> the author.\n>\n> How do you think the authorship information can be transferred sanely,\n> considering that the author didn't even use git in the first place?\n> Really?\n>\n> That's where the typos/mistakes/missing-info really happens. And it often\n> starts out with incomplete information, because the person has a bad email\n> setup, and the thing only has an email address to begin with, ie the\n> \"From:\" might literally say just \"tytso@mit.edu\" or something (to pick an\n> example from the Cc list in this discussion - when Ted sends real emails,\n> they tend to have proper naming).\n\nIf I recall correctly the top source of errors is variations in the\ndomain name of the email address. Second place was mangling of names\nfrom non-ASCII charsets. Third place was human typos. Fourth was\ninconsistency in the human name, like Ted's example.\n\nA really simple check would be for git to say - I've never seen this\nname/email combo before, are you sure it is correct before I commit\nit.\n\nPS - I am not in favor of the UUID scheme.\n\n>\n> Sometimes we'll edit the messages to have the \"From: xyz \" thing at\n> the top, fixing up the incomplete thing then. Typos happen there. Or the\n> patch will simply come in two different ways, so there's no typo, yet\n> there are two different emails that get author attribution.\n>\n> The thing is, development really is about human interaction. Yes, there's\n> a tool involved (git), and once the data is in the tool we won't lose it\n> any more, but this is about getting the data _into_ the tool in the first\n> place.\n>\n> And the data you want to add simply DOES NOT EXIST. And we can't make it\n> exist. The fact that even the trivial and obvious data that git _does_ ask\n> for gets to be incomplete should tell you something.\n>\n>                        Linus\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at  http://vger.kernel.org/majordomo-info.html\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137156","messageId":"alpine.LFD.2.00.1003181909180.31128@xanadu.home","threadId":"23060","inReplyTo":"46a038f91003181536ib3b74f8o40603a4cee13d62b@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-18T23:17:46Z","receivedAt":"2010-03-18T23:17:46Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Martin Langhoff wrote:\n\n> On Thu, Mar 18, 2010 at 1:27 PM, Linus Torvalds\n> wrote:\n> > Even shorter version: NO.\n> \n> One thing we all forgot to mention here is that even if it was a good\n> idea (which it is not), implementing it means a flag day: changing in\n> the pack format, wire protocol and APIs, messing up with compatibility\n> with users of pre-flag-day git, and rippling out to all the GUIs,\n> frontends and integration scripts out there.\n\nAnd nobody yet mentioned what should happen when someone sends a patch \nby email. Most commits in git.git originated from a patch sent via \nemail. Should we start pasting UUIDs in the email body? What if the \ncut & paste was quickly done and the UUID is missing a character or two? \nBecause this does happen. And because this UUID thing is supposed to be \na perfect identity representation then we'll need a .uuidmap to correct \nsuch mistakes of course.\n\nBetter improve on the existing .mailmap instead.\n\n\nNicolas\n"},{"id":"137157","messageId":"9e4733911003181626t7d143903mbc5737ff2fa5100f@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181909180.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T23:26:13Z","receivedAt":"2010-03-18T23:26:13Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 7:17 PM, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Martin Langhoff wrote:\n>\n>> On Thu, Mar 18, 2010 at 1:27 PM, Linus Torvalds\n>> wrote:\n>> > Even shorter version: NO.\n>>\n>> One thing we all forgot to mention here is that even if it was a good\n>> idea (which it is not), implementing it means a flag day: changing in\n>> the pack format, wire protocol and APIs, messing up with compatibility\n>> with users of pre-flag-day git, and rippling out to all the GUIs,\n>> frontends and integration scripts out there.\n>\n> And nobody yet mentioned what should happen when someone sends a patch\n> by email.  Most commits in git.git originated from a patch sent via\n> email.  Should we start pasting UUIDs in the email body?  What if the\n> cut & paste was quickly done and the UUID is missing a character or two?\n> Because this does happen.  And because this UUID thing is supposed to be\n> a perfect identity representation then we'll need a .uuidmap to correct\n> such mistakes of course.\n>\n> Better improve on the existing .mailmap instead.\n\nIf anyone is interested I can send them a .mailmap that fixes a lot of\nthe problems in the kernel tree. It's two years old so it will need\nupdating.\n\n>\n>\n> Nicolas\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at  http://vger.kernel.org/majordomo-info.html\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137158","messageId":"alpine.LFD.2.00.1003181930230.31128@xanadu.home","threadId":"23060","inReplyTo":"9e4733911003181626t7d143903mbc5737ff2fa5100f@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-18T23:34:32Z","receivedAt":"2010-03-18T23:34:32Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Jon Smirl wrote:\n\n> If anyone is interested I can send them a .mailmap that fixes a lot of\n> the problems in the kernel tree. It's two years old so it will need\n> updating.\n\nPlease just make a patch with it, and post it to lkml and CC Linus and \nAndrew Morton. Repost a month later if no one picked it up.\n\nI think that 'git log' should really consider the .mailmap by default. \nOtherwise what's the point? The only time when .mailmap should not be \nconsidered is when using --pretty=raw or when explicitly told not to.\n\n\nNicolas\n"},{"id":"137159","messageId":"b4087cc51003181634u102cacaey3cb3a870a3f9fc0f@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181909180.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T23:34:57Z","receivedAt":"2010-03-18T23:34:57Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 18:17, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Martin Langhoff wrote:\n>\n>> On Thu, Mar 18, 2010 at 1:27 PM, Linus Torvalds\n>> wrote:\n>> > Even shorter version: NO.\n>>\n>> One thing we all forgot to mention here is that even if it was a good\n>> idea (which it is not), implementing it means a flag day: changing in\n>> the pack format, wire protocol and APIs, messing up with compatibility\n>> with users of pre-flag-day git, and rippling out to all the GUIs,\n>> frontends and integration scripts out there.\n>\n> And nobody yet mentioned what should happen when someone sends a patch\n> by email.  Most commits in git.git originated from a patch sent via\n> email.  Should we start pasting UUIDs in the email body?  What if the\n> cut & paste was quickly done and the UUID is missing a character or two?\n> Because this does happen.  And because this UUID thing is supposed to be\n> a perfect identity representation then we'll need a .uuidmap to correct\n> such mistakes of course.\n>\n> Better improve on the existing .mailmap instead.\n\nActually, those points were touched upon earlier (including my rebuttals).\n"},{"id":"137160","messageId":"alpine.LFD.2.00.1003181935070.31128@xanadu.home","threadId":"23060","inReplyTo":"b4087cc51003181502y1f339cf8u268fa5388ace098@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-18T23:37:25Z","receivedAt":"2010-03-18T23:37:25Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Michael Witten wrote:\n\n> You missed the other line (probably gmail's fault):\n> \n> Most likely, all that will happen is identification entropy won't\n> increase nearly so rapidly and there might be other benefits\n> such as shortlog speed improvements.\n\nThe shortlog speed improvement is certainly not going to compensate for \nall the added human time needed to process the extra piece of \ninformation.\n\n\nNicolas\n"},{"id":"137161","messageId":"9e4733911003181641n400704c9r1a0addd6fce6fce0@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181930230.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-18T23:41:02Z","receivedAt":"2010-03-18T23:41:02Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 7:34 PM, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Jon Smirl wrote:\n>\n>> If anyone is interested I can send them a .mailmap that fixes a lot of\n>> the problems in the kernel tree. It's two years old so it will need\n>> updating.\n>\n> Please just make a patch with it, and post it to lkml and CC Linus and\n> Andrew Morton.  Repost a month later if no one picked it up.\n\nBeen there, done that. 1000 message flame war ensued about privacy\nconcerns over people's email address in the file.\n\n>\n> I think that 'git log' should really consider the .mailmap by default.\n> Otherwise what's the point?   The only time when .mailmap should not be\n> considered is when using --pretty=raw or when explicitly told not to.\n>\n>\n> Nicolas\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137162","messageId":"b4087cc51003181644j45722911u87d4cb62a9b4ea09@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181935070.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-18T23:44:17Z","receivedAt":"2010-03-18T23:44:17Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 18:37, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Michael Witten wrote:\n>\n>> You missed the other line (probably gmail's fault):\n>>\n>> Most likely, all that will happen is identification entropy won't\n>> increase nearly so rapidly and there might be other benefits\n>> such as shortlog speed improvements.\n>\n> The shortlog speed improvement is certainly not going to compensate for\n> all the added human time needed to process the extra piece of\n> information.\n\nWhat added human time?\n"},{"id":"137164","messageId":"alpine.LFD.2.00.1003181953010.31128@xanadu.home","threadId":"23060","inReplyTo":"9e4733911003181641n400704c9r1a0addd6fce6fce0@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-18T23:58:28Z","receivedAt":"2010-03-18T23:58:28Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Jon Smirl wrote:\n\n> On Thu, Mar 18, 2010 at 7:34 PM, Nicolas Pitre wrote:\n> > On Thu, 18 Mar 2010, Jon Smirl wrote:\n> >\n> >> If anyone is interested I can send them a .mailmap that fixes a lot of\n> >> the problems in the kernel tree. It's two years old so it will need\n> >> updating.\n> >\n> > Please just make a patch with it, and post it to lkml and CC Linus and\n> > Andrew Morton.  Repost a month later if no one picked it up.\n> \n> Been there, done that. 1000 message flame war ensued about privacy\n> concerns over people's email address in the file.\n\nWell, you used git itself as the data source to fix up those email \naddresses, right? If so there is simply no privacy concerns as the data \nis already there and public. Just don't venture adding emails that are \nnot already present in the whole Git history/content at all without \nconsent.\n\n\nNicolas\n"},{"id":"137165","messageId":"alpine.LFD.2.00.1003181958430.31128@xanadu.home","threadId":"23060","inReplyTo":"b4087cc51003181644j45722911u87d4cb62a9b4ea09@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-19T00:03:59Z","receivedAt":"2010-03-19T00:03:59Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Michael Witten wrote:\n\n> On Thu, Mar 18, 2010 at 18:37, Nicolas Pitre wrote:\n> > On Thu, 18 Mar 2010, Michael Witten wrote:\n> >\n> >> You missed the other line (probably gmail's fault):\n> >>\n> >> Most likely, all that will happen is identification entropy won't\n> >> increase nearly so rapidly and there might be other benefits\n> >> such as shortlog speed improvements.\n> >\n> > The shortlog speed improvement is certainly not going to compensate for\n> > all the added human time needed to process the extra piece of\n> > information.\n> \n> What added human time?\n\nThe time that humans will have to spend on this UUID \nsetup/fixing/whatnot.\n\n\nNicolas\n"},{"id":"137166","messageId":"9e4733911003181716q7f141d5eqd18218c749ca4624@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181953010.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-19T00:16:31Z","receivedAt":"2010-03-19T00:16:31Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 7:58 PM, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Jon Smirl wrote:\n>\n>> On Thu, Mar 18, 2010 at 7:34 PM, Nicolas Pitre wrote:\n>> > On Thu, 18 Mar 2010, Jon Smirl wrote:\n>> >\n>> >> If anyone is interested I can send them a .mailmap that fixes a lot of\n>> >> the problems in the kernel tree. It's two years old so it will need\n>> >> updating.\n>> >\n>> > Please just make a patch with it, and post it to lkml and CC Linus and\n>> > Andrew Morton.  Repost a month later if no one picked it up.\n>>\n>> Been there, done that. 1000 message flame war ensued about privacy\n>> concerns over people's email address in the file.\n>\n> Well, you used git itself as the data source to fix up those email\n> addresses, right?  If so there is simply no privacy concerns as the data\n> is already there and public.  Just don't venture adding emails that are\n> not already present in the whole Git history/content at all without\n> consent.\n\nI'll sent you the file and you can commit it. Please take full credit for it.\nhttp://lkml.org/lkml/2008/7/28/134\n\nAll of the data came out of git tree.\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137167","messageId":"alpine.LFD.2.00.1003181715490.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"9e4733911003181716q7f141d5eqd18218c749ca4624@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-19T00:17:04Z","receivedAt":"2010-03-19T00:17:04Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Jon Smirl wrote:\n> \n> I'll sent you the file and you can commit it. Please take full credit for it.\n\nUmm. You do realize that what people complained about was mostly that they \nfelt a lot of the entries were totally pointless.\n\nFor example, you included names whether they were mistyped or not, and \nclaimed that everybody needed to always be in the mailmap if they ever \nmade any commit.\n\nSo I think 99% of the flames were due to just the patch being stupid.\n\n\t\tLinus\n"},{"id":"137168","messageId":"b4087cc51003181727k37dabd15y12ba34762b67c4d7@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181958430.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T00:27:30Z","receivedAt":"2010-03-19T00:27:30Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Thu, Mar 18, 2010 at 19:03, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Michael Witten wrote:\n>\n>> On Thu, Mar 18, 2010 at 18:37, Nicolas Pitre wrote:\n>> > On Thu, 18 Mar 2010, Michael Witten wrote:\n>> >\n>> >> You missed the other line (probably gmail's fault):\n>> >>\n>> >> Most likely, all that will happen is identification entropy won't\n>> >> increase nearly so rapidly and there might be other benefits\n>> >> such as shortlog speed improvements.\n>> >\n>> > The shortlog speed improvement is certainly not going to compensate for\n>> > all the added human time needed to process the extra piece of\n>> > information.\n>>\n>> What added human time?\n>\n> The time that humans will have to spend on this UUID\n> setup/fixing/whatnot.\n\nCompatibility concerns aside, there is virtually no overhead. Indeed,\nthere would be less overhead than there is now in terms of fixing.\n"},{"id":"137169","messageId":"alpine.LFD.2.00.1003182030480.31128@xanadu.home","threadId":"23060","inReplyTo":"b4087cc51003181727k37dabd15y12ba34762b67c4d7@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-19T00:32:43Z","receivedAt":"2010-03-19T00:32:43Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Michael Witten wrote:\n\n> On Thu, Mar 18, 2010 at 19:03, Nicolas Pitre wrote:\n> > On Thu, 18 Mar 2010, Michael Witten wrote:\n> >\n> >> On Thu, Mar 18, 2010 at 18:37, Nicolas Pitre wrote:\n> >> > On Thu, 18 Mar 2010, Michael Witten wrote:\n> >> >\n> >> >> You missed the other line (probably gmail's fault):\n> >> >>\n> >> >> Most likely, all that will happen is identification entropy won't\n> >> >> increase nearly so rapidly and there might be other benefits\n> >> >> such as shortlog speed improvements.\n> >> >\n> >> > The shortlog speed improvement is certainly not going to compensate for\n> >> > all the added human time needed to process the extra piece of\n> >> > information.\n> >>\n> >> What added human time?\n> >\n> > The time that humans will have to spend on this UUID\n> > setup/fixing/whatnot.\n> \n> Compatibility concerns aside, there is virtually no overhead. Indeed,\n> there would be less overhead than there is now in terms of fixing.\n\nIn a perfect world maybe. Let's talk about it again when we get there.\n\n\nNicolas\n"},{"id":"137172","messageId":"9e4733911003181739m2f605dd7g9406aaecc296749f@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181715490.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-19T00:39:00Z","receivedAt":"2010-03-19T00:39:00Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 8:17 PM, Linus Torvalds\n wrote:\n>\n>\n> On Thu, 18 Mar 2010, Jon Smirl wrote:\n>>\n>> I'll sent you the file and you can commit it. Please take full credit for it.\n>\n> Umm. You do realize that what people complained about was mostly that they\n> felt a lot of the entries were totally pointless.\n>\n> For example, you included names whether they were mistyped or not, and\n> claimed that everybody needed to always be in the mailmap if they ever\n> made any commit.\n>\n> So I think 99% of the flames were due to just the patch being stupid.\n\nI had all of the names in the list so that I could regenerate the list\nand diff it against the old version to know which new names needed to\nbe checked. Looking back I could have eliminated the names without\nerrors and then added a comment to the file as to the last date all of\nthe names were checked. But that is less reliable than recording\nwhich were checked. The problem is that if you lose track of what has\nbeen checked, you are forced to recheck everything and it takes a long\ntime to recheck everything.\n\nI'll send you a copy and you can unstupify it.\n\n>\n>                Linus\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137171","messageId":"alpine.LFD.2.00.1003181739310.18017@i5.linux-foundation.org","threadId":"23060","inReplyTo":"9e4733911003181739m2f605dd7g9406aaecc296749f@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Linus Torvalds","fromEmail":"torvalds@linux-foundation.org","sentAt":"2010-03-19T00:50:58Z","receivedAt":"2010-03-19T00:50:58Z","isPatch":false,"sender":{"key":"torvalds@linux-foundation.org","avatar":"https://avatars.githubusercontent.com/u/1024025?v=4"},"body":"\n\nOn Thu, 18 Mar 2010, Jon Smirl wrote:\n> \n> I had all of the names in the list so that I could regenerate the list\n> and diff it against the old version to know which new names needed to\n> be checked. Looking back I could have eliminated the names without\n> errors and then added a comment to the file as to the last date all of\n> the names were checked. But that is less reliable than recording\n> which were checked. The problem is that if you lose track of what has\n> been checked, you are forced to recheck everything and it takes a long\n> time to recheck everything.\n\nThe part you keep missing is that NOBODY CARES!\n\nFor example, I exist in the current git kernel tree with 11 different \nnames for just the authorship information:\n\n 32 Linus Torvalds torvalds@evo.osdl.org\n 1522 Linus Torvalds torvalds@g5.osdl.org\n 4194 Linus Torvalds torvalds@linux-foundation.org\n 7 Linus Torvalds torvalds@macmini.osdl.org\n 2 Linus Torvalds torvalds@merom.osdl.org\n 8 Linus Torvalds torvalds@osdl.org\n 166 Linus Torvalds torvalds@ppc970.osdl.org\n 4 Linus Torvalds torvalds@ppc970.osdl.org.(none)\n 1 Linus Torvalds torvalds@quad.osdl.org\n 1606 Linus Torvalds torvalds@woody.linux-foundation.org\n 174 Linus Torvalds torvalds@woody.osdl.org\n\n(that's counts, in case you care). And then if you check signed-off lines, \nyou'll find some _additional_ oddities where things just got misspelled, \nlike\n\n\tLinus Torvalds \n\tLinus Torvalds \n\nwhere in one case there's a missing 'r', and in the other it's some odd \nperverse incestuous relationship between me and Andrew (in reality, it's \nme doing a stupid \"search-and-replace\" on the emails, adding my own \nsign-off to Andrew's and that got a bit too much copy-paste issues)\n\nThere's a few other mistakes like that in the sign-offs.\n\nDoes anybody care? Certainly not I. There is absolutely zero reason to \nworry about it. I used to find it convenient to see what machines I had \nworked on, so I actually included that. And one of them was clearly \nmis-configured, or git did something wrong when the hostname was already \nin FQDN format. Whatever.\n\nThere is no real _value_ in making a .mailcap for each such buggy entry is \nwhat I'm trying to tell you. Those things are maybe used for statistics. \nOn the whole, they are correct. \n\n\t\t\tLinus\n"},{"id":"137173","messageId":"9e4733911003181812k1cf08833ke683226377527b56@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181739310.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-19T01:12:59Z","receivedAt":"2010-03-19T01:12:59Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 8:50 PM, Linus Torvalds\n wrote:\n> Does anybody care? Certainly not I. There is absolutely zero reason to\n> worry about it. I used to find it convenient to see what machines I had\n> worked on, so I actually included that. And one of them was clearly\n> mis-configured, or git did something wrong when the hostname was already\n> in FQDN format. Whatever.\n>\n> There is no real _value_ in making a .mailcap for each such buggy entry is\n> what I'm trying to tell you. Those things are maybe used for statistics.\n> On the whole, they are correct.\n\nI was trying to track how many real people were working on the kernel.\n If we don't collapse the 13 different versions of you down to one\nperson the number numbers are way off.\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137175","messageId":"alpine.LFD.2.00.1003182141050.31128@xanadu.home","threadId":"23060","inReplyTo":"9e4733911003181812k1cf08833ke683226377527b56@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nicolas Pitre","fromEmail":"nico@fluxnic.net","sentAt":"2010-03-19T01:45:01Z","receivedAt":"2010-03-19T01:45:01Z","isPatch":false,"sender":{"key":"nico@fluxnic.net","avatar":"https://avatars.githubusercontent.com/u/702790?v=4"},"body":"On Thu, 18 Mar 2010, Jon Smirl wrote:\n\n> On Thu, Mar 18, 2010 at 8:50 PM, Linus Torvalds\n> > There is no real _value_ in making a .mailcap for each such buggy entry is\n> > what I'm trying to tell you. Those things are maybe used for statistics.\n> > On the whole, they are correct.\n> \n> I was trying to track how many real people were working on the kernel.\n> If we don't collapse the 13 different versions of you down to one\n> person the number numbers are way off.\n\nIf you have a cleaned up .mailmap file which doesn't include unneeded \nentries then just submit it for inclusion. If someone else eventually \ncares to check and update it then another patch should come forth at \nthat point. That doesn't have to be any more complicated than that.\n\n\nNicolas\n"},{"id":"137176","messageId":"9e4733911003181905i25e908f9u472d0467f68d3b00@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003182141050.31128@xanadu.home","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-19T02:05:52Z","receivedAt":"2010-03-19T02:05:52Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Thu, Mar 18, 2010 at 9:45 PM, Nicolas Pitre wrote:\n> On Thu, 18 Mar 2010, Jon Smirl wrote:\n>\n>> On Thu, Mar 18, 2010 at 8:50 PM, Linus Torvalds\n>> > There is no real _value_ in making a .mailcap for each such buggy entry is\n>> > what I'm trying to tell you. Those things are maybe used for statistics.\n>> > On the whole, they are correct.\n>>\n>> I was trying to track how many real people were working on the kernel.\n>>  If we don't collapse the 13 different versions of you down to one\n>> person the number numbers are way off.\n>\n> If you have a cleaned up .mailmap file which doesn't include unneeded\n> entries then just submit it for inclusion.  If someone else eventually\n> cares to check and update it then another patch should come forth at\n> that point.  That doesn't have to be any more complicated than that.\n\nI sent you a copy, feel free to do whatever you want with it. The\nacademics doing statistics on Linux will love you for submitting it.\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137179","messageId":"2e24e5b91003181947t48d7cb31jd6512d9b53bf0b32@mail.gmail.com","threadId":"23060","inReplyTo":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Sitaram Chamarty","fromEmail":"sitaramc@gmail.com","sentAt":"2010-03-19T02:47:23Z","receivedAt":"2010-03-19T02:47:23Z","isPatch":false,"sender":{"key":"sitaramc@gmail.com","avatar":"https://avatars.githubusercontent.com/u/43316?v=4"},"body":"On Thu, Mar 18, 2010 at 6:53 PM, Michael Witten wrote:\n> Short Version:\n> -------------\n\n[all snipped]\n\nGreat Gods above... 50+ emails, including many from Linus himself,\ntrying to respond to a non-solution to a non-problem...\n\nslow day?\n"},{"id":"137184","messageId":"544dda351003182217g3395c327v7be9c92e4ede6280@mail.gmail.com","threadId":"23060","inReplyTo":"2e24e5b91003181947t48d7cb31jd6512d9b53bf0b32@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Nazri Ramliy","fromEmail":"ayiehere@gmail.com","sentAt":"2010-03-19T05:17:09Z","receivedAt":"2010-03-19T05:17:09Z","isPatch":false,"sender":{"key":"ayiehere@gmail.com","avatar":"https://avatars.githubusercontent.com/u/164756?v=4"},"body":"On Fri, Mar 19, 2010 at 10:47 AM, Sitaram Chamarty wrote:\n> On Thu, Mar 18, 2010 at 6:53 PM, Michael Witten wrote:\n>> Short Version:\n>> -------------\n>\n> [all snipped]\n>\n> Great Gods above... 50+ emails, including many from Linus himself,\n> trying to respond to a non-solution to a non-problem...\n>\n> slow day?\n\nNah.. just wait until someone mentions either Hitler or Nazis.\n\nnazri.\n"},{"id":"137190","messageId":"4BA338C1.7030803@alum.mit.edu","threadId":"23060","inReplyTo":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Haggerty","fromEmail":"mhagger@alum.mit.edu","sentAt":"2010-03-19T08:41:37Z","receivedAt":"2010-03-19T08:41:37Z","isPatch":false,"sender":{"key":"mhagger@alum.mit.edu","avatar":"https://avatars.githubusercontent.com/u/119718?v=4"},"body":"Michael Witten wrote:\n> Rather than use a (name,email) pair to identify people, let's use\n> a (uuid,name,email) triplet.\n> [...]\n\nA UUID doesn't need to be a big hex number. All it has to be is a\n\"Universally Unique Identifier\". Like, oh, for example, your\n\n *** EMAIL ADDRESS ***\n\n[1]. There is even already a way to fix up mistakes or unavoidable\nemail address changes, namely the .mailmap file.\n\nSo if you are exercised about having a persistent identity, simply find\nan email provider that is unlikely to ever give your email address to\nsomebody else, and use that address consistently. Encourage other\npeople to do the same and to keep their .mailmap entries up to date.\n\n(Not that it's likely to happen, but having people maintain opaque UUIDs\nis even *less* likely.)\n\nMichael\n\n[1] The only non-UUID property of legitimate email addresses is that the\nusername part or even the domain name part of an email address can be\nrecycled. But with a reputable email provider this shouldn't be a\nproblem. For the purpose of the UUID it is not even a problem if the\nemail address becomes defunct, as long as it is not taken over by\nsomebody else.\n"},{"id":"137197","messageId":"b4087cc51003190439x3c9ff269g35d11432bd2a3d60@mail.gmail.com","threadId":"23060","inReplyTo":"4BA338C1.7030803@alum.mit.edu","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T11:39:04Z","receivedAt":"2010-03-19T11:39:04Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 02:41, Michael Haggerty wrote:\n> Michael Witten wrote:\n>> Rather than use a (name,email) pair to identify people, let's use\n>> a (uuid,name,email) triplet.\n>> [...]\n>\n> A UUID doesn't need to be a big hex number.  All it has to be is a\n> \"Universally Unique Identifier\".  Like, oh, for example, your\n>\n>                   *** EMAIL ADDRESS ***\n>\n> [1].  There is even already a way to fix up mistakes or unavoidable\n> email address changes, namely the .mailmap file.\n\n*facepalm*\n\nYou've just repeated everything that I've said; go look at the rest of\nthe thread, where I spend plenty of time correcting the same hangups\nabout my choice of the word UUID and my use of hex digits.\n\nI'm only observing that the current name/email system pair conflates\nan individual with his current email system and that it would be\nworthwhile to ALLOW an individual to FURTHER describe himself by\nincluding another piece of information that is solely meant as\nidentification within git. That piece of information could be whatever\na user deems to be uniquely identifying for himself. You could use\n\"Michael Haggerty \" as your uuid, and you could\nstill use it after you change the `email' config variable to something\nelse.\n\nThere is MUCH LESS CHANCE of such a uuid getting trashed by typos,\nchanging names, and changing email addresses; of course it can still\nget messed up, but the rate at which something like .mailmap would\nneed to be updated would likely be greatly decreased and it would make\ngathering statistics easier (especially for the individuals who take\nadvantage of such a uuid for describing themselves---and it only\nrequires setting one config variable to something easily remembered by\nthat person).\n\nI cover all of this numerous times in numerous rebuttals; don't\ncontribute to a thread with more than 60 emails without having read at\nleast some of them. If you don't care to read so much, then perhaps\njump here:\n\n http://marc.info/?l=git&m=126894679711600&w=2\n\nIn the end, there is probably only one legitimate problem with my\nproposal: It might break compatibility with older repo formats/tools.\nI'm not sure about that.\n\nSincerely,\nMichael Witten\n"},{"id":"137198","messageId":"alpine.DEB.2.00.1003190441530.3821@asgard.lang.hm","threadId":"23060","inReplyTo":"b4087cc51003190439x3c9ff269g35d11432bd2a3d60@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"","fromEmail":"david@lang.hm","sentAt":"2010-03-19T11:45:38Z","receivedAt":"2010-03-19T11:45:38Z","isPatch":false,"sender":{"key":"david@lang.hm","avatar":null},"body":"On Fri, 19 Mar 2010, Michael Witten wrote:\n\n> On Fri, Mar 19, 2010 at 02:41, Michael Haggerty wrote:\n>> Michael Witten wrote:\n>>> Rather than use a (name,email) pair to identify people, let's use\n>>> a (uuid,name,email) triplet.\n>>> [...]\n>>\n>> A UUID doesn't need to be a big hex number.  All it has to be is a\n>> \"Universally Unique Identifier\".  Like, oh, for example, your\n>>\n>>                   *** EMAIL ADDRESS ***\n>>\n>> [1].  There is even already a way to fix up mistakes or unavoidable\n>> email address changes, namely the .mailmap file.\n>\n> *facepalm*\n>\n> You've just repeated everything that I've said; go look at the rest of\n> the thread, where I spend plenty of time correcting the same hangups\n> about my choice of the word UUID and my use of hex digits.\n>\n> I'm only observing that the current name/email system pair conflates\n> an individual with his current email system and that it would be\n> worthwhile to ALLOW an individual to FURTHER describe himself by\n> including another piece of information that is solely meant as\n> identification within git. That piece of information could be whatever\n> a user deems to be uniquely identifying for himself. You could use\n> \"Michael Haggerty \" as your uuid, and you could\n> still use it after you change the `email' config variable to something\n> else.\n>\n> There is MUCH LESS CHANCE of such a uuid getting trashed by typos,\n> changing names, and changing email addresses; of course it can still\n> get messed up, but the rate at which something like .mailmap would\n> need to be updated would likely be greatly decreased and it would make\n> gathering statistics easier (especially for the individuals who take\n> advantage of such a uuid for describing themselves---and it only\n> requires setting one config variable to something easily remembered by\n> that person).\n\nhere is where you are missing the point.\n\nno, there is not 'much less chance' of it getting messed up.\n\nyou seem to assume that people would never need to set the UUID on \nmultiple machines.\n\nif they don't need to set it on multiple machines, then the e-mail/userid \nis going to be reliable anyway\n\nif they do need to set it on multiple machines and can't be bothered to \nkeep their e-mail consistant, why would they bother keeping this \nadditional thing considtant? Linus is pointing out that people don't care \nnow about their e-mail and name, and will care even less about some \nabstract UUID\n\npeople who care will already make their e-mail consistant.\n\nDavid Lang\n\n\n> I cover all of this numerous times in numerous rebuttals; don't\n> contribute to a thread with more than 60 emails without having read at\n> least some of them. If you don't care to read so much, then perhaps\n> jump here:\n>\n> http://marc.info/?l=git&m=126894679711600&w=2\n>\n> In the end, there is probably only one legitimate problem with my\n> proposal: It might break compatibility with older repo formats/tools.\n> I'm not sure about that.\n>\n> Sincerely,\n> Michael Witten\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at http://vger.kernel.org/majordomo-info.html\n>"},{"id":"137199","messageId":"20100319115445.GA12986@glandium.org","threadId":"23060","inReplyTo":"alpine.DEB.2.00.1003190441530.3821@asgard.lang.hm","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Mike Hommey","fromEmail":"mh@glandium.org","sentAt":"2010-03-19T11:54:45Z","receivedAt":"2010-03-19T11:54:45Z","isPatch":false,"sender":{"key":"mh@glandium.org","avatar":"https://avatars.githubusercontent.com/u/1038527?v=4"},"body":"On Fri, Mar 19, 2010 at 04:45:38AM -0700, david@lang.hm wrote:\n> here is where you are missing the point.\n> \n> no, there is not 'much less chance' of it getting messed up.\n> \n> you seem to assume that people would never need to set the UUID on\n> multiple machines.\n> \n> if they don't need to set it on multiple machines, then the\n> e-mail/userid is going to be reliable anyway\n> \n> if they do need to set it on multiple machines and can't be bothered\n> to keep their e-mail consistant, why would they bother keeping this\n> additional thing considtant? Linus is pointing out that people don't\n> care now about their e-mail and name, and will care even less about\n> some abstract UUID\n> \n> people who care will already make their e-mail consistant.\n\nWhile I don't agree with the need for that uuid thing, I'd like to\npinpoint that people who care can't necessarily make their e-mail\nconsistant. For example, Linus used to use an @osdl.org address, and\nhe now uses an @linux-foundation.org address. It's still the same Linus,\nbut the (name, email) pair has legitimately changed.\n\nMike\n"},{"id":"137200","messageId":"b4087cc51003190508m3651e929w2dc16156a84d09f@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.DEB.2.00.1003190441530.3821@asgard.lang.hm","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T12:08:04Z","receivedAt":"2010-03-19T12:08:04Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 05:45, wrote:\n> On Fri, 19 Mar 2010, Michael Witten wrote:\n>\n>> On Fri, Mar 19, 2010 at 02:41, Michael Haggerty \n>> wrote:\n>>>\n>>> Michael Witten wrote:\n>>>>\n>>>> Rather than use a (name,email) pair to identify people, let's use\n>>>> a (uuid,name,email) triplet.\n>>>> [...]\n>>>\n>>> A UUID doesn't need to be a big hex number.  All it has to be is a\n>>> \"Universally Unique Identifier\".  Like, oh, for example, your\n>>>\n>>>                   *** EMAIL ADDRESS ***\n>>>\n>>> [1].  There is even already a way to fix up mistakes or unavoidable\n>>> email address changes, namely the .mailmap file.\n>>\n>> *facepalm*\n>>\n>> You've just repeated everything that I've said; go look at the rest of\n>> the thread, where I spend plenty of time correcting the same hangups\n>> about my choice of the word UUID and my use of hex digits.\n>>\n>> I'm only observing that the current name/email system pair conflates\n>> an individual with his current email system and that it would be\n>> worthwhile to ALLOW an individual to FURTHER describe himself by\n>> including another piece of information that is solely meant as\n>> identification within git. That piece of information could be whatever\n>> a user deems to be uniquely identifying for himself. You could use\n>> \"Michael Haggerty \" as your uuid, and you could\n>> still use it after you change the `email' config variable to something\n>> else.\n>>\n>> There is MUCH LESS CHANCE of such a uuid getting trashed by typos,\n>> changing names, and changing email addresses; of course it can still\n>> get messed up, but the rate at which something like .mailmap would\n>> need to be updated would likely be greatly decreased and it would make\n>> gathering statistics easier (especially for the individuals who take\n>> advantage of such a uuid for describing themselves---and it only\n>> requires setting one config variable to something easily remembered by\n>> that person).\n>\n> here is where you are missing the point.\n>\n> no, there is not 'much less chance' of it getting messed up.\n>\n> you seem to assume that people would never need to set the UUID on multiple\n> machines.\n\nI covered that in the first email, highlighting the importance of\nusing an easily remembered, already reasonably unique piece of\ninformation (like a name/email pair) that you don't need to change.\n\n> if they don't need to set it on multiple machines, then the e-mail/userid is\n> going to be reliable anyway\n\nThe problem is that the name/email pair (as in the 'name' and 'email'\nconfig variables) is NOT ONLY subject to typos, but it is ALSO subject\nto changing email accounts and changing real life names.\n\nIf you don't use the uuid `field' that I propose, then everything\nwould be just like it was before. If you do use it, then you can\neasily identify all of your own contributions regardless of what your\nname/email du jour is.\n\n> if they do need to set it on multiple machines and can't be bothered to keep\n> their e-mail consistant, why would they bother keeping this additional thing\n> considtant? Linus is pointing out that people don't care now about their\n> e-mail and name, and will care even less about some abstract UUID\n\nThe user doesn't have a damn choice!\n\nThe email can't be kept consistent over time because the tools expect\nit to be and/or use the actual physical email used to send/receive\nstuff. It's information that CONFLATES identity with whatever\ntool/system you're using.\n\nFor instance, Michael Haggerty cannot reasonably use\n\n [user]\n name = Michael Haggerty\n email = mhagger@MIT.EDU\n\nbecause he likely no longer has that email account to use. He is\nforced to change it and therefore forced to make his identity\nconfused.\n\nI'm proposing ALLOWING him to say:\n\n [user]\n uuid = Michael Haggerty \n name = Michael Haggerty\n email = mhagger@ALUM.mit.edu\n\nHeck, let's say he works at Red Hat as well; he might make some\ncommits under this config AT WORK:\n\n [user]\n uuid = Michael Haggerty \n name = Michael Haggerty\n email = mhagger@redhat.com\n\nThen, he can make, say, commits to the Linux kernel repo for both work\nand hobby related issues and still be recognized as the same person.\nThat is, he can have some commits under \"Michael Haggerty\n\" and other commits under \"Michael Haggerty\n\".\n\nSincerely,\nMichael Witten\n"},{"id":"137201","messageId":"3f4fd2641003190509j6e4908e4kab87b601d2ce9c00@mail.gmail.com","threadId":"23060","inReplyTo":"20100319115445.GA12986@glandium.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Reece Dunn","fromEmail":"msclrhd@googlemail.com","sentAt":"2010-03-19T12:09:26Z","receivedAt":"2010-03-19T12:09:26Z","isPatch":false,"sender":{"key":"msclrhd@googlemail.com","avatar":null},"body":"On 19 March 2010 11:54, Mike Hommey wrote:\n> On Fri, Mar 19, 2010 at 04:45:38AM -0700, david@lang.hm wrote:\n>> here is where you are missing the point.\n>>\n>> no, there is not 'much less chance' of it getting messed up.\n>>\n>> you seem to assume that people would never need to set the UUID on\n>> multiple machines.\n>>\n>> if they don't need to set it on multiple machines, then the\n>> e-mail/userid is going to be reliable anyway\n>>\n>> if they do need to set it on multiple machines and can't be bothered\n>> to keep their e-mail consistant, why would they bother keeping this\n>> additional thing considtant? Linus is pointing out that people don't\n>> care now about their e-mail and name, and will care even less about\n>> some abstract UUID\n>>\n>> people who care will already make their e-mail consistant.\n>\n> While I don't agree with the need for that uuid thing, I'd like to\n> pinpoint that people who care can't necessarily make their e-mail\n> consistant. For example, Linus used to use an @osdl.org address, and\n> he now uses an @linux-foundation.org address. It's still the same Linus,\n> but the (name, email) pair has legitimately changed.\n\nSo create an aliases list that maps one (name,email) to another that\nis from the same person. There is no need for an additional item (a\nuuid) to solve this problem. It also means that searching on any\n(name,email) pair will find the others, so you only need to\nremember/find one of the identities for the person you are interested\nin finding the commits for.\n\nAFAICS, mailmap is about correcting mistakes (primarily in the\nreported name for a given email address). In this case, mailmap and\nthis aliases-map will work in conjunction with each other to give what\nthe original poster wanted. However, I haven't seen any of his replies\nthat answer this (or sufficiently address why mailmap does not solve\nhis problem).\n\n- Reece\n"},{"id":"137202","messageId":"b4087cc51003190509y6ce7ad9dy992254cfe6ac2fb3@mail.gmail.com","threadId":"23060","inReplyTo":"20100319115445.GA12986@glandium.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T12:09:34Z","receivedAt":"2010-03-19T12:09:34Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 05:54, Mike Hommey wrote:\n> While I don't agree with the need for that uuid thing, I'd like to\n> pinpoint that people who care can't necessarily make their e-mail\n> consistant. For example, Linus used to use an @osdl.org address, and\n> he now uses an @linux-foundation.org address. It's still the same Linus,\n> but the (name, email) pair has legitimately changed.\n\nIndeed.\n\nThis is because the name/email pair (as in the 'name' and 'email'\nconfig variables) CONFLATES the idea of identity and current email\naccount.\n"},{"id":"137203","messageId":"b4087cc51003190516h42202e34k598a163c246cb9f2@mail.gmail.com","threadId":"23060","inReplyTo":"3f4fd2641003190509j6e4908e4kab87b601d2ce9c00@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T12:16:27Z","receivedAt":"2010-03-19T12:16:27Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 06:09, Reece Dunn wrote:\n> On 19 March 2010 11:54, Mike Hommey wrote:\n>> On Fri, Mar 19, 2010 at 04:45:38AM -0700, david@lang.hm wrote:\n>>> here is where you are missing the point.\n>>>\n>>> no, there is not 'much less chance' of it getting messed up.\n>>>\n>>> you seem to assume that people would never need to set the UUID on\n>>> multiple machines.\n>>>\n>>> if they don't need to set it on multiple machines, then the\n>>> e-mail/userid is going to be reliable anyway\n>>>\n>>> if they do need to set it on multiple machines and can't be bothered\n>>> to keep their e-mail consistant, why would they bother keeping this\n>>> additional thing considtant? Linus is pointing out that people don't\n>>> care now about their e-mail and name, and will care even less about\n>>> some abstract UUID\n>>>\n>>> people who care will already make their e-mail consistant.\n>>\n>> While I don't agree with the need for that uuid thing, I'd like to\n>> pinpoint that people who care can't necessarily make their e-mail\n>> consistant. For example, Linus used to use an @osdl.org address, and\n>> he now uses an @linux-foundation.org address. It's still the same Linus,\n>> but the (name, email) pair has legitimately changed.\n>\n> So create an aliases list that maps one (name,email) to another that\n> is from the same person. There is no need for an additional item (a\n> uuid) to solve this problem. It also means that searching on any\n> (name,email) pair will find the others, so you only need to\n> remember/find one of the identities for the person you are interested\n> in finding the commits for.\n>\n> AFAICS, mailmap is about correcting mistakes (primarily in the\n> reported name for a given email address). In this case, mailmap and\n> this aliases-map will work in conjunction with each other to give what\n> the original poster wanted. However, I haven't seen any of his replies\n> that answer this (or sufficiently address why mailmap does not solve\n> his problem).\n\nSee:\n\n http://marc.info/?l=git&m=126900051102958&w=2\n\nThe idea is to distribute the responsibility for maintaining a\nconsistent identity AND to make that responsibility EASY.\n\nThe extra uuid `field' can only suffer from typos, while the\nname/email pair can suffer from typos, changing email accounts, and\nchanging real life names. If the uuid `field' does get bungled by a\ntypo or is not used, then we're no worse off than we were before.\n"},{"id":"137204","messageId":"b4087cc51003190518y2a7004c2pf8effaa9f5adf502@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190516h42202e34k598a163c246cb9f2@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T12:18:26Z","receivedAt":"2010-03-19T12:18:26Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 06:16, Michael Witten wrote:\n> The extra uuid `field' can only suffer from typos\n\nI should add that because the uuid `field' would be typed pretty much\nonly as a config variable and then used by git tools from thenceforth,\nthe rate at which typos can occur is much less than for the name/email\npair.\n"},{"id":"137205","messageId":"9e4733911003190525m552a400exdd7fb3f6c88cc4ca@mail.gmail.com","threadId":"23060","inReplyTo":"3f4fd2641003190509j6e4908e4kab87b601d2ce9c00@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-19T12:25:28Z","receivedAt":"2010-03-19T12:25:28Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Fri, Mar 19, 2010 at 8:09 AM, Reece Dunn wrote:\n> On 19 March 2010 11:54, Mike Hommey wrote:\n>> On Fri, Mar 19, 2010 at 04:45:38AM -0700, david@lang.hm wrote:\n>>> here is where you are missing the point.\n>>>\n>>> no, there is not 'much less chance' of it getting messed up.\n>>>\n>>> you seem to assume that people would never need to set the UUID on\n>>> multiple machines.\n>>>\n>>> if they don't need to set it on multiple machines, then the\n>>> e-mail/userid is going to be reliable anyway\n>>>\n>>> if they do need to set it on multiple machines and can't be bothered\n>>> to keep their e-mail consistant, why would they bother keeping this\n>>> additional thing considtant? Linus is pointing out that people don't\n>>> care now about their e-mail and name, and will care even less about\n>>> some abstract UUID\n>>>\n>>> people who care will already make their e-mail consistant.\n>>\n>> While I don't agree with the need for that uuid thing, I'd like to\n>> pinpoint that people who care can't necessarily make their e-mail\n>> consistant. For example, Linus used to use an @osdl.org address, and\n>> he now uses an @linux-foundation.org address. It's still the same Linus,\n>> but the (name, email) pair has legitimately changed.\n>\n> So create an aliases list that maps one (name,email) to another that\n> is from the same person. There is no need for an additional item (a\n> uuid) to solve this problem. It also means that searching on any\n> (name,email) pair will find the others, so you only need to\n> remember/find one of the identities for the person you are interested\n> in finding the commits for.\n\ngit already supports aliases via the .mailmap file. Pick one\nname/address pair that you like and then use .mailmap to map all of\nthe variations into the primary one. Granted some git tools don't\nprocess .mailmap, but it is easier to fix the tools that create a new\nID system.\n\nLook at the .mailmap in the current kernel tree. It fixes a few\nproblems. I have a much larger one that fixes most address issues.\n\nYou don't need to reimplement this aliases, they are already in git.\n\n\n>\n> AFAICS, mailmap is about correcting mistakes (primarily in the\n> reported name for a given email address). In this case, mailmap and\n> this aliases-map will work in conjunction with each other to give what\n> the original poster wanted. However, I haven't seen any of his replies\n> that answer this (or sufficiently address why mailmap does not solve\n> his problem).\n>\n> - Reece\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at  http://vger.kernel.org/majordomo-info.html\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137206","messageId":"4BA36F5F.9080706@gnu.org","threadId":"23060","inReplyTo":"b4087cc51003181457s76560b02y84aafbe77f45af87@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Paolo Bonzini","fromEmail":"bonzini@gnu.org","sentAt":"2010-03-19T12:34:39Z","receivedAt":"2010-03-19T12:34:39Z","isPatch":false,"sender":{"key":"bonzini@gnu.org","avatar":"https://avatars.githubusercontent.com/u/42082?v=4"},"body":"On 03/18/2010 10:57 PM, Michael Witten wrote:\n> On Thu, Mar 18, 2010 at 16:39, Martin Langhoff\n> wrote:\n>> On Thu, Mar 18, 2010 at 5:29 PM, Michael Witten wrote:\n>>> On Thu, Mar 18, 2010 at 16:19, Martin Langhoff\n>>>> What's the value? For me it'll be \"Martin Langhoff\". I already have that.\n>>>\n>>> Well, that's rather egotistical considering you're probably not the\n>>> only Martin Langhoff in this world. I'd advocate something like\n>>> \"Martin Langhoff\".\n>>\n>> So you are saying we should change the core datamodel of git to say...\n>> what we already can say?\n>\n> You see, Martin, you might want/need to stop using \"Martin Langhoff\n> \" as your email account, but there's no\n> reason why you can't continue to use it for your UUID.\n\nWhile a gnu.org or gmail.com will (most likely) stay with some person \nforever, hindsight is 20/20 and many people may generate his UUID from a \nwork email. So, suppose I make my UUID based on \nwhat will guarantee that in 20 years I won't find a new career as a \nbartender, and Red Hat wouldn't hire someone with my same name, and give \nhim the same email address?\n\nHeck, some people use gmail only for their personal email, and they \nrightly cannot be bothered to create another account to solve a problem \nthey don't understand and they probably do not have.\n\nFor the UUID to make sense, it would need to be what the acronym says: \nuniversally unique. An SHA-1 value is _not_ universally unique, it is \njust a one-way function. There are tons of git repos out there with a \nblob hashing to e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 or \n257cc5642cb1a054f08cc83f2d943e56fd3ebe99.\n\nI have an idea. Start your own website uuidemail.com. One registers \nand gets an alias for their email, something like \n8aacc35ffca0d34fccf8a750e84e3a81bdcb940b@uuidemail.com. Then people can \nstart using \n8aacc35ffca0d34fccf8a750e84e3a81bdcb940b+pbonzini--redhat.com@uuidemail.com \nas their git user.email. I bet nobody will.\n\nPaolo\n\nps: Yes, in a perfect world it would be nice for people to know that I \nam the same person independent of whether I contribute as \nbonzini@gnu.org or pbonzini@redhat.com. But we're not in a perfect \nworld, so amen.\n"},{"id":"137207","messageId":"3f4fd2641003190540y79685d69x7270c3aa84cdf6f7@mail.gmail.com","threadId":"23060","inReplyTo":"9e4733911003190525m552a400exdd7fb3f6c88cc4ca@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Reece Dunn","fromEmail":"msclrhd@googlemail.com","sentAt":"2010-03-19T12:40:15Z","receivedAt":"2010-03-19T12:40:15Z","isPatch":false,"sender":{"key":"msclrhd@googlemail.com","avatar":null},"body":"On 19 March 2010 12:25, Jon Smirl wrote:\n> On Fri, Mar 19, 2010 at 8:09 AM, Reece Dunn wrote:\n>> On 19 March 2010 11:54, Mike Hommey wrote:\n>>> On Fri, Mar 19, 2010 at 04:45:38AM -0700, david@lang.hm wrote:\n>>>> here is where you are missing the point.\n>>>>\n>>>> no, there is not 'much less chance' of it getting messed up.\n>>>>\n>>>> you seem to assume that people would never need to set the UUID on\n>>>> multiple machines.\n>>>>\n>>>> if they don't need to set it on multiple machines, then the\n>>>> e-mail/userid is going to be reliable anyway\n>>>>\n>>>> if they do need to set it on multiple machines and can't be bothered\n>>>> to keep their e-mail consistant, why would they bother keeping this\n>>>> additional thing considtant? Linus is pointing out that people don't\n>>>> care now about their e-mail and name, and will care even less about\n>>>> some abstract UUID\n>>>>\n>>>> people who care will already make their e-mail consistant.\n>>>\n>>> While I don't agree with the need for that uuid thing, I'd like to\n>>> pinpoint that people who care can't necessarily make their e-mail\n>>> consistant. For example, Linus used to use an @osdl.org address, and\n>>> he now uses an @linux-foundation.org address. It's still the same Linus,\n>>> but the (name, email) pair has legitimately changed.\n>>\n>> So create an aliases list that maps one (name,email) to another that\n>> is from the same person. There is no need for an additional item (a\n>> uuid) to solve this problem. It also means that searching on any\n>> (name,email) pair will find the others, so you only need to\n>> remember/find one of the identities for the person you are interested\n>> in finding the commits for.\n>\n> git already supports aliases via the .mailmap file. Pick one\n> name/address pair that you like and then use .mailmap to map all of\n> the variations into the primary one. Granted some git tools don't\n> process .mailmap, but it is easier to fix the tools that create a new\n> ID system.\n>\n> Look at the .mailmap in the current kernel tree. It fixes a few\n> problems. I have a much larger one that fixes most address issues.\n>\n> You don't need to reimplement this aliases, they are already in git.\n\nIndeed. I wasn't aware that mailmap catered for this as well.\n\n- Reece\n"},{"id":"137208","messageId":"b4087cc51003190543mafc403q6260a3774653e2d5@mail.gmail.com","threadId":"23060","inReplyTo":"4BA36F5F.9080706@gnu.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T12:43:44Z","receivedAt":"2010-03-19T12:43:44Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 06:34, Paolo Bonzini wrote:\n> On 03/18/2010 10:57 PM, Michael Witten wrote:\n>>\n>> On Thu, Mar 18, 2010 at 16:39, Martin Langhoff\n>>  wrote:\n>>>\n>>> On Thu, Mar 18, 2010 at 5:29 PM, Michael Witten\n>>>  wrote:\n>>>>\n>>>> On Thu, Mar 18, 2010 at 16:19, Martin Langhoff\n>>>>>\n>>>>> What's the value? For me it'll be \"Martin Langhoff\". I already have\n>>>>> that.\n>>>>\n>>>> Well, that's rather egotistical considering you're probably not the\n>>>> only Martin Langhoff in this world. I'd advocate something like\n>>>> \"Martin Langhoff\".\n>>>\n>>> So you are saying we should change the core datamodel of git to say...\n>>> what we already can say?\n>>\n>> You see, Martin, you might want/need to stop using \"Martin Langhoff\n>> \" as your email account, but there's no\n>> reason why you can't continue to use it for your UUID.\n>\n> While a gnu.org or gmail.com will (most likely) stay with some person\n> forever, hindsight is 20/20 and many people may generate his UUID from a\n> work email.  So, suppose I make my UUID based on what\n> will guarantee that in 20 years I won't find a new career as a bartender,\n> and Red Hat wouldn't hire someone with my same name, and give him the same\n> email address?\n\nFirstly, the UUID need not be a name/email pair.\n\nSecondly, you're being ridiculous; even if that ridiculous scenario\nplayed out not-infrequently, there would still be less identity\nconfusion in git repos over time, because changing real life names,\nand changing email accounts do happen frequently and are not\nridiculous events.\n\n> Heck, some people use gmail only for their personal email, and they rightly\n> cannot be bothered to create another account to solve a problem they don't\n> understand and they probably do not have.\n\nThis doesn't make any sense. Why does anybody need to create another\naccount? Are you still confused about what a uuid is this context?\n\n> For the UUID to make sense, it would need to be what the acronym says:\n> universally unique.  An SHA-1 value is _not_ universally unique, it is just\n> a one-way function.  There are tons of git repos out there with a blob\n> hashing to e69de29bb2d1d6434b8b29ae775ad8c2e48c5391 or\n> 257cc5642cb1a054f08cc83f2d943e56fd3ebe99.\n\nThe SHA-1 is supposed to be an optimization; it's not essential, as\nI've already explained; I also get the feeling that you're being\nridiculous again. In particular, I don't see your point.\n\n> I have an idea.  Start your own website uuidemail.com.  One registers and\n> gets an alias for their email, something like\n> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b@uuidemail.com.  Then people can\n> start using\n> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b+pbonzini--redhat.com@uuidemail.com\n> as their git user.email.  I bet nobody will.\n\nThis is nonsense that betrays your misunderstanding.\n"},{"id":"137209","messageId":"4BA373C0.2070506@gnu.org","threadId":"23060","inReplyTo":"b4087cc51003190543mafc403q6260a3774653e2d5@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Paolo Bonzini","fromEmail":"bonzini@gnu.org","sentAt":"2010-03-19T12:53:20Z","receivedAt":"2010-03-19T12:53:20Z","isPatch":false,"sender":{"key":"bonzini@gnu.org","avatar":"https://avatars.githubusercontent.com/u/42082?v=4"},"body":"\n>> While a gnu.org or gmail.com will (most likely) stay with some\n>> person forever, hindsight is 20/20 and many people may generate\n>> his UUID from a work email. So, suppose I make my UUID based\n>> on what will guarantee that in 20 years I\n>> won't find a new career as a bartender, and Red Hat wouldn't hire\n>> someone with my same name, and give him the same email address?\n>\n> Firstly, the UUID need not be a name/email pair.\n\nThat's what you lastly proposed generating it from.\n\n> Secondly, you're being ridiculous; even if that ridiculous scenario\n> played out not-infrequently\n\nIt's not a matter of frequency. If you want a \"UU\" identification,\ncollisions must not even happen *once*.\n\n>> I have an idea. Start your own website uuidemail.com. One\n>> registers and gets an alias for their email, something like\n>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b@uuidemail.com. Then\n>> people can start using\n>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b+pbonzini--redhat.com@uuidemail.com\n>> as their git user.email. I bet nobody will.\n>\n> This is nonsense that betrays your misunderstanding.\n\nWhy? What does (name, email, uuid) provide over (name, concat(uuid, \nemail))? Nothing.\n\nBut the point is, neither really provides anything over (name, email).\n\nPaolo\n"},{"id":"137210","messageId":"b4087cc51003190603v259b833embb40b7360a07cf1f@mail.gmail.com","threadId":"23060","inReplyTo":"4BA373C0.2070506@gnu.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T13:03:25Z","receivedAt":"2010-03-19T13:03:25Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 06:53, Paolo Bonzini wrote:\n>\n>>> While a gnu.org or gmail.com will (most likely) stay with some\n>>> person forever, hindsight is 20/20 and many people may generate\n>>> his UUID from a work email.  So, suppose I make my UUID based\n>>> on  what will guarantee that in 20 years I\n>>> won't find a new career as a bartender, and Red Hat wouldn't hire\n>>> someone with my same name, and give him the same email address?\n>>\n>> Firstly, the UUID need not be a name/email pair.\n>\n> That's what you lastly proposed generating it from.\n\nNo. Please go read.\n\n>> Secondly, you're being ridiculous; even if that ridiculous scenario\n>> played out not-infrequently\n>\n> It's not a matter of frequency.  If you want a \"UU\" identification,\n> collisions must not even happen *once*.\n\nI've got news for you. The UUIDs generated by uuidgen CAN collide:\n\n The new UUID can reasonably be considered unique\n among all UUIDs created on the local system, and\n among UUIDs created on other systems in the past\n and in the future.\n\nYou're creating a straw man argument; conceptually, what I propose is\nbetter than what the current system provides because it would decrease\nthe rate at which identity entropy increases.\n\n>>> I have an idea.  Start your own website uuidemail.com.  One\n>>> registers and gets an alias for their email, something like\n>>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b@uuidemail.com.  Then\n>>> people can start using\n>>>\n>>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b+pbonzini--redhat.com@uuidemail.com\n>>> as their git user.email.  I bet nobody will.\n>>\n>> This is nonsense that betrays your misunderstanding.\n>\n> Why?  What does (name, email, uuid) provide over (name, concat(uuid,\n> email))?  Nothing.\n\nGo read the thread until you understand.\n"},{"id":"137211","messageId":"4BA3776B.5040706@gnu.org","threadId":"23060","inReplyTo":"b4087cc51003190603v259b833embb40b7360a07cf1f@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Paolo Bonzini","fromEmail":"bonzini@gnu.org","sentAt":"2010-03-19T13:08:59Z","receivedAt":"2010-03-19T13:08:59Z","isPatch":false,"sender":{"key":"bonzini@gnu.org","avatar":"https://avatars.githubusercontent.com/u/42082?v=4"},"body":"On 03/19/2010 02:03 PM, Michael Witten wrote:\n\n>>> Secondly, you're being ridiculous; even if that ridiculous scenario\n>>> played out not-infrequently\n>>\n>> It's not a matter of frequency. If you want a \"UU\" identification,\n>> collisions must not even happen *once*.\n>\n> I've got news for you. The UUIDs generated by uuidgen CAN collide:\n>\n> The new UUID can reasonably be considered unique\n> among all UUIDs created on the local system, and\n> among UUIDs created on other systems in the past\n> and in the future.\n\nPlease read the UUID generation algorithm.\n\n> You're creating a straw man argument; conceptually, what I propose is\n> better than what the current system provides because it would decrease\n> the rate at which identity entropy increases.\n\nMaybe you have to define entropy. For human consumers, \"Paolo Bonzini \n\" has considerably less \"entropy\" than \n8aacc35ffca0d34fccf8a750e84e3a81bdcb940b, as does even \"Paolo Bonzini \n\". For non-human consumers, a \ngood mailmap will do.\n\n>>>> I have an idea. Start your own website uuidemail.com. One\n>>>> registers and gets an alias for their email, something like\n>>>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b@uuidemail.com. Then\n>>>> people can start using\n>>>>\n>>>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b+pbonzini--redhat.com@uuidemail.com\n>>>> as their git user.email. I bet nobody will.\n>>>\n>>> This is nonsense that betrays your misunderstanding.\n>>\n>> Why? What does (name, email, uuid) provide over (name, concat(uuid,\n>> email))? Nothing.\n>\n> Go read the thread until you understand.\n\nI am not alone.\n\nPaolo\n"},{"id":"137212","messageId":"b4087cc51003190613y2fe32c26jd27cb009e64f197c@mail.gmail.com","threadId":"23060","inReplyTo":"4BA3776B.5040706@gnu.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T13:13:25Z","receivedAt":"2010-03-19T13:13:25Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 07:08, Paolo Bonzini wrote:\n> Maybe you have to define entropy.  For human consumers, \"Paolo Bonzini\n> \" has considerably less \"entropy\" than\n> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b, as does even \"Paolo Bonzini\n> \".  For non-human consumers, a good\n> mailmap will do.\n\nAs I've stated before many times, the SHA-1 is not necessary to the proposal.\n\nPlease go read.\n"},{"id":"137214","messageId":"873DD7AF-6D5C-4F5F-BDC1-1F26480D412F@wincent.com","threadId":"23060","inReplyTo":"b4087cc51003190613y2fe32c26jd27cb009e64f197c@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Wincent Colaiuta","fromEmail":"win@wincent.com","sentAt":"2010-03-19T13:41:11Z","receivedAt":"2010-03-19T13:41:11Z","isPatch":false,"sender":{"key":"greg@hurrell.net","avatar":"https://avatars.githubusercontent.com/u/7074?v=4"},"body":"El 19/03/2010, a las 14:13, Michael Witten escribió:\n\n> On Fri, Mar 19, 2010 at 07:08, Paolo Bonzini wrote:\n>> Maybe you have to define entropy. For human consumers, \"Paolo Bonzini\n>> \" has considerably less \"entropy\" than\n>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b, as does even \"Paolo Bonzini\n>> \". For non-human consumers, a good\n>> mailmap will do.\n> \n> As I've stated before many times, the SHA-1 is not necessary to the proposal.\n> \n> Please go read.\n\nStop telling people to go read your idiotic proposal. It has _already_ been read with great attention, and multiple people have shown immense patience repeatedly explaining to you why the idea is stupid. Your continued trolling is really starting to grate.\n\nThe overwhelming, sustained opposition to your idea should already be enough indication that such a proposal will _never_ be accepted into the Git codebase, so right now you're just wasting people's time.\n\nw\n"},{"id":"137216","messageId":"b4087cc51003190659h3d619fdbgc57ce343ee0527a3@mail.gmail.com","threadId":"23060","inReplyTo":"873DD7AF-6D5C-4F5F-BDC1-1F26480D412F@wincent.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T13:59:40Z","receivedAt":"2010-03-19T13:59:40Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 07:41, Wincent Colaiuta wrote:\n> El 19/03/2010, a las 14:13, Michael Witten escribió:\n>\n>> On Fri, Mar 19, 2010 at 07:08, Paolo Bonzini wrote:\n>>> Maybe you have to define entropy.  For human consumers, \"Paolo Bonzini\n>>> \" has considerably less \"entropy\" than\n>>> 8aacc35ffca0d34fccf8a750e84e3a81bdcb940b, as does even \"Paolo Bonzini\n>>> \".  For non-human consumers, a good\n>>> mailmap will do.\n>>\n>> As I've stated before many times, the SHA-1 is not necessary to the proposal.\n>>\n>> Please go read.\n>\n> Stop telling people to go read your idiotic proposal. It has _already_ been read with great attention, and multiple people have shown immense patience repeatedly explaining to you why the idea is stupid. Your continued trolling is really starting to grate.\n\nI've shown immense patience repeatedly explaining why these\n'explanations' are strawmen or based on misunderstandings and bad\nassumptions.\n\nIt's true that I have been receiving perfectly valid complaints. The\nproblem is that almost all of them have nothing to do with what I've\nbeen saying because people see 'uuid' and a few examples with hex\ndigits and then erroneously construct the rest in their heads.\n\n> The overwhelming, sustained opposition to your idea should already be enough indication that such a proposal will _never_ be accepted into the Git codebase, so right now you're just wasting people's time.\n\nI long ago gave up the notion that it would be included in the git codebase.\n\nInstead, I've been defending the idea, which is a simple but vast\nimprovement over the current system; had it been in place since the\nbeginning, a lot of trouble could have been reduced.\n\nIndeed, the only thing that makes this great idea a bad idea is\nCOMPATIBILITY CONCERNS; that's it.\n"},{"id":"137219","messageId":"4BA3855A.1070005@alum.mit.edu","threadId":"23060","inReplyTo":"b4087cc51003190439x3c9ff269g35d11432bd2a3d60@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Haggerty","fromEmail":"mhagger@alum.mit.edu","sentAt":"2010-03-19T14:08:26Z","receivedAt":"2010-03-19T14:08:26Z","isPatch":false,"sender":{"key":"mhagger@alum.mit.edu","avatar":"https://avatars.githubusercontent.com/u/119718?v=4"},"body":"Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 02:41, Michael Haggerty wrote:\n>> Michael Witten wrote:\n>>> Rather than use a (name,email) pair to identify people, let's use\n>>> a (uuid,name,email) triplet.\n>>> [...]\n>> A UUID doesn't need to be a big hex number. All it has to be is a\n>> \"Universally Unique Identifier\". Like, oh, for example, your\n>>\n>> *** EMAIL ADDRESS ***\n>>\n>> [1]. There is even already a way to fix up mistakes or unavoidable\n>> email address changes, namely the .mailmap file.\n> \n> *facepalm*\n> \n> You've just repeated everything that I've said; go look at the rest of\n> the thread, where I spend plenty of time correcting the same hangups\n> about my choice of the word UUID and my use of hex digits.\n\nNo, my point is to use the *existing* email address as the UUID\n*without* adding another field. Nothing needs to be changed!\n\n> [...] You could use\n> \"Michael Haggerty \" as your uuid, and you could\n> still use it after you change the `email' config variable to something\n> else.\n\nGive me a break. It's not so damn hard to keep an email address over\ntime. And if it changes, I can update the .mailcap file to map my old\nemail address to the new one and *presto* I have a new, equally valid\nUUID that I can continue to commit under.\n\n> I cover all of this numerous times in numerous rebuttals; don't\n> contribute to a thread with more than 60 emails without having read at\n> least some of them.\n\nWrong. I've read the whole idiotic thread. To prove it I'll summarize\nit for you: you argue the same point over and over again while ignoring\nthe legitimate objections of just about every other participant.\n\nAdding a new UUID field is obviously a non-starter, so I suggested a way\nto get the same (very marginal) benefit from the fields that are already\npresent in every git repository.\n\nMichael\n"},{"id":"137220","messageId":"m3tyscjt7h.fsf@localhost.localdomain","threadId":"23060","inReplyTo":"4ba2293f.c5c2f10a.5e9c.5c4a@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2010-03-19T14:08:55Z","receivedAt":"2010-03-19T14:08:55Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"Michael Witten writes:\n\n> Short Version:\n> -------------\n> \n> \n> Rather than use a (name,email) pair to identify people, let's use\n> a (uuid,name,email) triplet.\n> \n> The uuid can be any piece of information that a user of git determines\n> to be reasonably unique across space and time and that is intended to\n> be used by that user virtually forever (at least within a project's\n> history).\n> \n> For instance, the uuid could be an OSF DCE 1.1 UUID or the SHA-1 of\n> some easily remembered, already reasonably unique information.\n\n... or 'canonical-name canonical-email' pair.\n\n> \n> This could really help keep identifications clean, and it is rather\n> straightforward and possibly quite efficient.\n> \n> \n> Long Version:\n> ------------\n[...]\n\n> While git's use of (name,email) pairs to identify each person is\n> extremely practical, it turns out that it's rather `unstable';\n\nThis is non-solution to non-problem.\n\nFirst, the user.name and user.email does not need to be name and email\nfrom some email account. It might be some \"canonical name\" and \n\"canonical email\".\n\nSecond, there are (I think) two main sources of 'unstability' in\n(name,email) pairs, namely A) misconfigured git (when fetching/pushing\nusing git itself), B) wrong name in email etc. (when sending patches\nvia email, 80% of patches in Linux kernel case).\n\nIn the case of misconfigured git (case A) using UUID wouldn't help,\nand only make it worse (you would have to configure the same UUID on\neach machine). What would help here is for git to be more strict and\nperhaps forbid (some of) autogenerated names and emails.\n\nIn the case of sending patches via email, you can use in-body 'From:'\nto provide (name,email) part that is different than account used to\nsend email. In the case of UUID you would need the same: some way to\nprovide UUID in patch (in email). UUID has the disadvantage of being\nrequired also when (name,email) in From: email header is good user ID.\nSo UUID wouldn't help there either.\n\n\nWhat could help in both cases is .mailmap being used (perhaps on\ndemand) in more git commands. See Documentation/mailmap.txt\nor e.g. git-shortlog(1) manpage. It is quite advanced tool for\ncorrecting mistakes (it can correct *both* user name, which is\nmost common usage, but also email address).\n\n-- \nJakub Narebski\nPoland\nShadeHawk on #git\n"},{"id":"137222","messageId":"46a038f91003190713m5d99a38fo898d9d2391103f3@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190659h3d619fdbgc57ce343ee0527a3@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Martin Langhoff","fromEmail":"martin.langhoff@gmail.com","sentAt":"2010-03-19T14:13:14Z","receivedAt":"2010-03-19T14:13:14Z","isPatch":false,"sender":{"key":"martin.langhoff@gmail.com","avatar":"https://gravatar.com/avatar/1e3f311b6c4c15836501901ca58f8c0b0667246488084ba524d8bc9867e22fd9?d=mp&s=160"},"body":"On Fri, Mar 19, 2010 at 9:59 AM, Michael Witten wrote:\n> I've shown immense patience repeatedly explaining\n\nNo, you haven't. _You_ are misunderstanding.\n\nWe have what you want: email + name, and a mapping mechanism (mailmap)\nto cope with variations. It is good enough.\n\n> Indeed, the only thing that makes this great idea a bad idea is\n> COMPATIBILITY CONCERNS; that's it.\n\nGood... at last! But don't put ALL CAPS when you are in the wrong,\nmate. And wasting a lot of people's time.\n\n\n\nm\n-- \n martin.langhoff@gmail.com\n martin@laptop.org -- School Server Architect\n - ask interesting questions\n - don't get distracted with shiny stuff - working code first\n - http://wiki.laptop.org/go/User:Martinlanghoff\n"},{"id":"137226","messageId":"9e4733911003190733h62e2aea0i7573bba641f4d931@mail.gmail.com","threadId":"23060","inReplyTo":"m3tyscjt7h.fsf@localhost.localdomain","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jon Smirl","fromEmail":"jonsmirl@gmail.com","sentAt":"2010-03-19T14:33:27Z","receivedAt":"2010-03-19T14:33:27Z","isPatch":false,"sender":{"key":"jonsmirl@gmail.com","avatar":"https://gravatar.com/avatar/cff3bf5bfdfa6708b905712ff91f0f9b8aaca161659f38c02b787920d5d28b7e?d=mp&s=160"},"body":"On Fri, Mar 19, 2010 at 10:08 AM, Jakub Narebski wrote:\n> Michael Witten writes:\n>\n>> Short Version:\n>> -------------\n>>\n>>\n>> Rather than use a (name,email) pair to identify people, let's use\n>> a (uuid,name,email) triplet.\n>>\n>> The uuid can be any piece of information that a user of git determines\n>> to be reasonably unique across space and time and that is intended to\n>> be used by that user virtually forever (at least within a project's\n>> history).\n>>\n>> For instance, the uuid could be an OSF DCE 1.1 UUID or the SHA-1 of\n>> some easily remembered, already reasonably unique information.\n>\n> ... or 'canonical-name canonical-email' pair.\n>\n>>\n>> This could really help keep identifications clean, and it is rather\n>> straightforward and possibly quite efficient.\n>>\n>>\n>> Long Version:\n>> ------------\n> [...]\n>\n>> While git's use of (name,email) pairs to identify each person is\n>> extremely practical, it turns out that it's rather `unstable';\n>\n> This is non-solution to non-problem.\n>\n> First, the user.name and user.email does not need to be name and email\n> from some email account.  It might be some \"canonical name\" and\n> \"canonical email\".\n>\n> Second, there are (I think) two main sources of 'unstability' in\n> (name,email) pairs, namely A) misconfigured git (when fetching/pushing\n> using git itself), B) wrong name in email etc. (when sending patches\n> via email, 80% of patches in Linux kernel case).\n\nAnother top source is mangling of non-ASCII charsets when they go\nthough the email system. Are the git work flow tools safe for\nalternative charsets? Do the email tools look at the charset header of\nthe email message? Check people's names in the kernel commits and\nyou'll find lots of examples of this type of mangling.\n\nOr people not using UTF-8. There are files in the kernel where\npeople's names are in conflicting codepages. Should git try to look\nfor diffs that aren't UTF-8?\n\n>\n> In the case of misconfigured git (case A) using UUID wouldn't help,\n> and only make it worse (you would have to configure the same UUID on\n> each machine).  What would help here is for git to be more strict and\n> perhaps forbid (some of) autogenerated names and emails.\n>\n> In the case of sending patches via email, you can use in-body 'From:'\n> to provide (name,email) part that is different than account used to\n> send email.  In the case of UUID you would need the same: some way to\n> provide UUID in patch (in email).  UUID has the disadvantage of being\n> required also when (name,email) in From: email header is good user ID.\n> So UUID wouldn't help there either.\n>\n>\n> What could help in both cases is .mailmap being used (perhaps on\n> demand) in more git commands.  See Documentation/mailmap.txt\n> or e.g. git-shortlog(1) manpage.  It is quite advanced tool for\n> correcting mistakes (it can correct *both* user name, which is\n> most common usage, but also email address).\n>\n> --\n> Jakub Narebski\n> Poland\n> ShadeHawk on #git\n> --\n> To unsubscribe from this list: send the line \"unsubscribe git\" in\n> the body of a message to majordomo@vger.kernel.org\n> More majordomo info at  http://vger.kernel.org/majordomo-info.html\n>\n\n\n\n-- \nJon Smirl\njonsmirl@gmail.com\n"},{"id":"137227","messageId":"b4087cc51003190740h680b5dech4edd7a5000f180ee@mail.gmail.com","threadId":"23060","inReplyTo":"m3tyscjt7h.fsf@localhost.localdomain","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T14:40:31Z","receivedAt":"2010-03-19T14:40:31Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 08:08, Jakub Narebski wrote:\n> Michael Witten writes:\n>\n>> Short Version:\n>> -------------\n>>\n>>\n>> Rather than use a (name,email) pair to identify people, let's use\n>> a (uuid,name,email) triplet.\n>>\n>> The uuid can be any piece of information that a user of git determines\n>> to be reasonably unique across space and time and that is intended to\n>> be used by that user virtually forever (at least within a project's\n>> history).\n>>\n>> For instance, the uuid could be an OSF DCE 1.1 UUID or the SHA-1 of\n>> some easily remembered, already reasonably unique information.\n>\n> ... or 'canonical-name canonical-email' pair.\n>\n>>\n>> This could really help keep identifications clean, and it is rather\n>> straightforward and possibly quite efficient.\n>>\n>>\n>> Long Version:\n>> ------------\n> [...]\n>\n>> While git's use of (name,email) pairs to identify each person is\n>> extremely practical, it turns out that it's rather `unstable';\n>\n> This is non-solution to non-problem.\n>\n> First, the user.name and user.email does not need to be name and email\n> from some email account.  It might be some \"canonical name\" and\n> \"canonical email\".\n\nThe vast majority of patches come in through email; the git tools\nexpect the user.name and user.email to reflect physical email account\ninformation.\n\nYou would be correct if it were not for the fact that git currently\nconflates identity and current email system.\n\n> Second, there are (I think) two main sources of 'unstability' in\n> (name,email) pairs, namely A) misconfigured git (when fetching/pushing\n> using git itself), B) wrong name in email etc. (when sending patches\n> via email, 80% of patches in Linux kernel case).\n>\n> In the case of misconfigured git (case A) using UUID wouldn't help,\n> and only make it worse (you would have to configure the same UUID on\n> each machine).  What would help here is for git to be more strict and\n> perhaps forbid (some of) autogenerated names and emails.\n\nThe uuid string would be typed pretty much only during configuration;\nfrom there, it's basically just handled by the git tools. Hence, the\nuuid can indeed suffer from typos, but the name/email pair can suffer\nfrom not only typos but also real life name changing and email account\nswitching.\n\nThere would still be the same problem of variations in uuid for one\nperson, but the problem would very likely be greatly reduced; if a\nperson doesn't use the uuid properly or at all, then we're in the\nexact same situation we were before. Those who do use it, though, will\nbe much better off.\n\nStrictness about names and emails is difficult, and keeping something\nlike the current .mailmap file up-to-date is a centralized process.\nThe uuid field would distribute the responsibility of maintaining\nidentity and make that responsibility easy because the user-chosen\nstring is easy for that user to remember and is typed only very\noccasionally and under very specific circumstances.\n\n> In the case of sending patches via email, you can use in-body 'From:'\n> to provide (name,email) part that is different than account used to\n> send email.\n\nThat's a good solution that I've considered, except for 2 reasons:\n\n * It involves much more opportunities for typos and/or the\n configuration of a non-git tool for a git-specific purpose.\n\n * Many if not most email services will refuse to send messages\n with forged/spoofed email addresses.\n\n> In the case of UUID you would need the same: some way to\n> provide UUID in patch (in email).\n\nYes, but that's automated by tools like git's format-patch. Not using\nsomething like format-patch or some other git interface is an\n'out-of-band' communication and that author has essentially chosen not\nto care about his identity.\n\nThe use of the uuid field and allowing git tools to handle it is just\na way to give a person who does care about his identity to keep it\nconsistent.\n\n> UUID has the disadvantage of being\n> required also when (name,email) in From: email header is good user ID.\n> So UUID wouldn't help there either.\n\nIt's not a good user id because it depends on factors other than identity.\n\n> What could help in both cases is .mailmap being used (perhaps on\n> demand) in more git commands.  See Documentation/mailmap.txt\n> or e.g. git-shortlog(1) manpage.  It is quite advanced tool for\n> correcting mistakes (it can correct *both* user name, which is\n> most common usage, but also email address).\n\nThe disadvantage here is that it centralizes identity management and\nit is more demanding because the name/email pair is quite unstable.\n\nOn the other hand, something like a uuid field would distribute that\nmanagement to the user himself and frees that user from the influences\nof legal name changing and email address switching.\n\nOf course, as already stated, some people may bungle their uuid\nsetting. Then something like .mailmap can be used, but the format\nwould be simpler, the file would not grow nearly as quickly, and with\nsome clever encoding some statistics gathering programs could\n(possibly) run more efficiently.\n"},{"id":"137230","messageId":"4BA38F9F.8050007@drmicha.warpmail.net","threadId":"23060","inReplyTo":"9e4733911003190733h62e2aea0i7573bba641f4d931@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael J Gruber","fromEmail":"git@drmicha.warpmail.net","sentAt":"2010-03-19T14:52:15Z","receivedAt":"2010-03-19T14:52:15Z","isPatch":false,"sender":{"key":"git@grubix.eu","avatar":"https://avatars.githubusercontent.com/u/233215?v=4"},"body":"Jon Smirl venit, vidit, dixit 19.03.2010 15:33:\n> On Fri, Mar 19, 2010 at 10:08 AM, Jakub Narebski wrote:\n>> Michael Witten writes:\n>>\n>>> Short Version:\n>>> -------------\n>>>\n>>>\n>>> Rather than use a (name,email) pair to identify people, let's use\n>>> a (uuid,name,email) triplet.\n>>>\n>>> The uuid can be any piece of information that a user of git determines\n>>> to be reasonably unique across space and time and that is intended to\n>>> be used by that user virtually forever (at least within a project's\n>>> history).\n>>>\n>>> For instance, the uuid could be an OSF DCE 1.1 UUID or the SHA-1 of\n>>> some easily remembered, already reasonably unique information.\n>>\n>> ... or 'canonical-name canonical-email' pair.\n>>\n>>>\n>>> This could really help keep identifications clean, and it is rather\n>>> straightforward and possibly quite efficient.\n>>>\n>>>\n>>> Long Version:\n>>> ------------\n>> [...]\n>>\n>>> While git's use of (name,email) pairs to identify each person is\n>>> extremely practical, it turns out that it's rather `unstable';\n>>\n>> This is non-solution to non-problem.\n>>\n>> First, the user.name and user.email does not need to be name and email\n>> from some email account. It might be some \"canonical name\" and\n>> \"canonical email\".\n>>\n>> Second, there are (I think) two main sources of 'unstability' in\n>> (name,email) pairs, namely A) misconfigured git (when fetching/pushing\n>> using git itself), B) wrong name in email etc. (when sending patches\n>> via email, 80% of patches in Linux kernel case).\n> \n> Another top source is mangling of non-ASCII charsets when they go\n> though the email system. Are the git work flow tools safe for\n> alternative charsets? Do the email tools look at the charset header of\n> the email message? Check people's names in the kernel commits and\n> you'll find lots of examples of this type of mangling.\n>\n\nOr even the quoting of quotes for nick names, appearing as 'nick',\n\"nick\", \\\"nick\\\", nick and what not.\n\n> Or people not using UTF-8. There are files in the kernel where\n> people's names are in conflicting codepages. Should git try to look\n> for diffs that aren't UTF-8?\n\nYou and others are proving a very important point here: This is really\nan lkml proxy fight being taken to the git list, after the futile\nmailmap-ification there.\n\nPeople may disagree on the best approach in general, but this thread\nclearly shows:\n\n- The Git community is happy with mailmap for git.git.\n- The Git community does not see any need for amending the mailmap\nmechanism.\n- How you actually use mailmap (leniently or enforcing) is a per-project\ndecision, just like the patch workflow, the meaning and use of s-o-b\nlines, the requirement for full names and many other things.\n\nBut since the git list is hosted on kernel.org we can't really complain\nabout providing room for an lkml discussion ;)\n\nMichael\n"},{"id":"137232","messageId":"40aa078e1003190756h190af29ao472508aaf624efe3@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190740h680b5dech4edd7a5000f180ee@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Erik Faye-Lund","fromEmail":"kusmabite@googlemail.com","sentAt":"2010-03-19T14:56:52Z","receivedAt":"2010-03-19T14:56:52Z","isPatch":false,"sender":{"key":"kusmabite@gmail.com","avatar":"https://avatars.githubusercontent.com/u/47073?v=4"},"body":"On Fri, Mar 19, 2010 at 3:40 PM, Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 08:08, Jakub Narebski wrote:\n>> First, the user.name and user.email does not need to be name and email\n>> from some email account.  It might be some \"canonical name\" and\n>> \"canonical email\".\n>\n> The vast majority of patches come in through email; the git tools\n> expect the user.name and user.email to reflect physical email account\n> information.\n\nWhat git tools would that be? The only one I know of that does\nanything near assuming that is git send-email, and it only uses\nuser.email if neither sendemail.from is configured nor --from option\nis specified. And even when it does, it prompts the user so it can be\nchanged if called from a terminal. So I wouldn't say that it assumes\nanything about the \"physicalness\" of user.email, it just uses it's as\nthe most sane default unless anything else has been specified.\n\n-- \nErik \"kusma\" Faye-Lund\n"},{"id":"137231","messageId":"3f4fd2641003190757y39050691y3dc0ca08bd5196fb@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190516h42202e34k598a163c246cb9f2@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Reece Dunn","fromEmail":"msclrhd@googlemail.com","sentAt":"2010-03-19T14:57:59Z","receivedAt":"2010-03-19T14:57:59Z","isPatch":false,"sender":{"key":"msclrhd@googlemail.com","avatar":null},"body":"On 19 March 2010 12:16, Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 06:09, Reece Dunn wrote:\n>> On 19 March 2010 11:54, Mike Hommey wrote:\n>>> On Fri, Mar 19, 2010 at 04:45:38AM -0700, david@lang.hm wrote:\n>>>> here is where you are missing the point.\n>>>>\n>>>> no, there is not 'much less chance' of it getting messed up.\n>>>>\n>>>> you seem to assume that people would never need to set the UUID on\n>>>> multiple machines.\n>>>>\n>>>> if they don't need to set it on multiple machines, then the\n>>>> e-mail/userid is going to be reliable anyway\n>>>>\n>>>> if they do need to set it on multiple machines and can't be bothered\n>>>> to keep their e-mail consistant, why would they bother keeping this\n>>>> additional thing considtant? Linus is pointing out that people don't\n>>>> care now about their e-mail and name, and will care even less about\n>>>> some abstract UUID\n>>>>\n>>>> people who care will already make their e-mail consistant.\n>>>\n>>> While I don't agree with the need for that uuid thing, I'd like to\n>>> pinpoint that people who care can't necessarily make their e-mail\n>>> consistant. For example, Linus used to use an @osdl.org address, and\n>>> he now uses an @linux-foundation.org address. It's still the same Linus,\n>>> but the (name, email) pair has legitimately changed.\n>>\n>> So create an aliases list that maps one (name,email) to another that\n>> is from the same person. There is no need for an additional item (a\n>> uuid) to solve this problem. It also means that searching on any\n>> (name,email) pair will find the others, so you only need to\n>> remember/find one of the identities for the person you are interested\n>> in finding the commits for.\n>>\n>> AFAICS, mailmap is about correcting mistakes (primarily in the\n>> reported name for a given email address). In this case, mailmap and\n>> this aliases-map will work in conjunction with each other to give what\n>> the original poster wanted. However, I haven't seen any of his replies\n>> that answer this (or sufficiently address why mailmap does not solve\n>> his problem).\n>\n> See:\n>\n>    http://marc.info/?l=git&m=126900051102958&w=2\n>\n> The idea is to distribute the responsibility for maintaining a\n> consistent identity AND to make that responsibility EASY.\n>\n> The extra uuid `field' can only suffer from typos, while the\n> name/email pair can suffer from typos, changing email accounts, and\n> changing real life names. If the uuid `field' does get bungled by a\n> typo or is not used, then we're no worse off than we were before.\n\nWhat specific problem(s) are you trying to solve?\n\nThe main issue is identifying who made what changes to a repository\n(e.g. by a script, or database/statistics algorithms). The mailmap\nfile allows for corrections to a canonical (name,email) pair for a\nspecified repository.\n\nFor identifying the same person working across multiple projects,\nideally they should keep the canonical (name,email) pair consistent\nacross all projects, with mailmap files in the respective projects to\nkeep the canonical form correct.\n\nThis canonical (name,email) pair is then a unique identifier for that\nperson and then effectively becomes a uuid. There is no need to add an\nextra uuid field that needs *more* work fixing up errors and making\nconsistent.\n\nIf you change email address or name, *and* care enough about it being\nconsistent, there is no reason why you cannot update the mailmap file\nto use the new canonical (name,email) pair.\n\nOh, and you are expressing it wrong (if I understand you correctly)...\n\nWhat you are after is a string U (the uuid) that is used to identify a\nperson irrespective of their name and email. At the moment\n U = (name,email)\nis used to achieve that, with mailmap to normalise the variations.\n\nWhat you are trying to express is:\n U <=> (name,email)\nwhere U can be any unique string. This is different from using a\n(name,email,uuid) triple to identify someone.\n\nSo, lets say that I choose U=abc to identify myself uniquely, so that:\n \"abc\" <=> \"Reece Dunn \"\n \"abc\" <=> \"Reece Dunn \"\n \"abc\" <=> \"Reece Dunn \"\n \"abc\" <=> \"Reece H. Dunn \"\n \"abc\" <=> \"Reece H Dunn \"\n\nI would still need to define all these variations when and as they\noccur in a repository to fixup any typos and email address changes\nthat occur, so why not just pick U = \"Reece H. Dunn\n\" as the canonical form instead of \"abc\" or some\nother string?\n\nAs has been said, mailmap supports name variations (\"Reece Dunn\",\n\"Reece H Dunn\", \"Reece H. Dunn\") and email variations\n(msclrhd@hotmail.com, msclrhd@gmail.com, msclrhd@googlemail.com), so\nhow does a string that I need to set on the git client in addition to\nname and email help me define a canonical form *in the git\nrepository*?\n\nSo, I'll ask again: what problems are you trying to solve that cannot\nbe solved by mailmap?\n\n- Reece\n"},{"id":"137233","messageId":"b4087cc51003190805h1eb91af9wf0c8e987f9a75026@mail.gmail.com","threadId":"23060","inReplyTo":"40aa078e1003190756h190af29ao472508aaf624efe3@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T15:05:53Z","receivedAt":"2010-03-19T15:05:53Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 08:56, Erik Faye-Lund wrote:\n> On Fri, Mar 19, 2010 at 3:40 PM, Michael Witten wrote:\n>> On Fri, Mar 19, 2010 at 08:08, Jakub Narebski wrote:\n>>> First, the user.name and user.email does not need to be name and email\n>>> from some email account.  It might be some \"canonical name\" and\n>>> \"canonical email\".\n>>\n>> The vast majority of patches come in through email; the git tools\n>> expect the user.name and user.email to reflect physical email account\n>> information.\n>\n> What git tools would that be?\n\nAnything involving emailed patches.\n\n> The only one I know of that does\n> anything near assuming that is git send-email, and it only uses\n> user.email if neither sendemail.from is configured nor --from option\n> is specified. And even when it does, it prompts the user so it can be\n> changed if called from a terminal. So I wouldn't say that it assumes\n> anything about the \"physicalness\" of user.email, it just uses it's as\n> the most sane default unless anything else has been specified.\n\nIt's useless to spoof the From field because many email services won't\nsend it, a point I already covered in the email you quoted.\n\nWhen a patch is finally emailed, it's the From field that is used for\nAuthor attribution.\n\nYou see? Your identity has been tied to whatever email service you\nhappen to use at any given time rather than to something with more\nlong term stability.\n"},{"id":"137235","messageId":"3f4fd2641003190812p7625f2a6l4678d9a01fcdbd6a@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190740h680b5dech4edd7a5000f180ee@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Reece Dunn","fromEmail":"msclrhd@googlemail.com","sentAt":"2010-03-19T15:12:12Z","receivedAt":"2010-03-19T15:12:12Z","isPatch":false,"sender":{"key":"msclrhd@googlemail.com","avatar":null},"body":"On 19 March 2010 14:40, Michael Witten wrote:\n> Strictness about names and emails is difficult, and keeping something\n> like the current .mailmap file up-to-date is a centralized process.\n> The uuid field would distribute the responsibility of maintaining\n> identity and make that responsibility easy because the user-chosen\n> string is easy for that user to remember and is typed only very\n> occasionally and under very specific circumstances.\n\nI don't get this - it is the other way around.\n\nFor the mailmap file, you check that file into the git repository\nitself. Therefore, by implication, mailmap *is* distributed. It is\ntherefore kept locally and accessed locally. It also does not suffer\nfrom configuration issues, as you don't need to re-enter it if you\nchange your computer.\n\nFor a uuid to work the way you intend it, there would need to be some\nuniversal central server that would be queried to look up and resolve\nthe uuid so you can get consistent user identification information for\nevery git command by every person/script from every git repository.\nThis is never going to fly for all the reasons distributed VCSs were\ncreated in the first place.\n\nUnless by distributed you mean in the .git/config file, which is\nalways local and never distributed to others. However, the uuid data\nin the repository will be distributed in the repositories, so how is\nthis any better than what git has now?\n\n- Reece\n"},{"id":"137234","messageId":"b4087cc51003190812v73c23cb1jdb37341183831421@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190805h1eb91af9wf0c8e987f9a75026@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T15:12:31Z","receivedAt":"2010-03-19T15:12:31Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 09:05, Michael Witten wrote:\n>\n> It's useless to spoof the From field because many email services won't\n> send it, a point I already covered in the email you quoted.\n>\n> When a patch is finally emailed, it's the From field that is used for\n> Author attribution.\n>\n> You see? Your identity has been tied to whatever email service you\n> happen to use at any given time rather than to something with more\n> long term stability.\n\nA lot of trouble could probably be avoided if the Authorship\ninformation could be sent as something separate from the From field. I\ndon't think it would be quite as powerful as having a uuid, but it\nwould be less invasive and probably practically as effective.\n"},{"id":"137236","messageId":"40aa078e1003190825j57ba4a0axbb4d2bf7cbdb32d0@mail.gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190805h1eb91af9wf0c8e987f9a75026@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Erik Faye-Lund","fromEmail":"kusmabite@googlemail.com","sentAt":"2010-03-19T15:25:26Z","receivedAt":"2010-03-19T15:25:26Z","isPatch":false,"sender":{"key":"kusmabite@gmail.com","avatar":"https://avatars.githubusercontent.com/u/47073?v=4"},"body":"On Fri, Mar 19, 2010 at 4:05 PM, Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 08:56, Erik Faye-Lund wrote:\n>> On Fri, Mar 19, 2010 at 3:40 PM, Michael Witten wrote:\n>>> On Fri, Mar 19, 2010 at 08:08, Jakub Narebski wrote:\n>>>> First, the user.name and user.email does not need to be name and email\n>>>> from some email account.  It might be some \"canonical name\" and\n>>>> \"canonical email\".\n>>>\n>>> The vast majority of patches come in through email; the git tools\n>>> expect the user.name and user.email to reflect physical email account\n>>> information.\n>>\n>> What git tools would that be?\n>\n> Anything involving emailed patches.\n\nWhich are...?\n\n>\n>> The only one I know of that does\n>> anything near assuming that is git send-email, and it only uses\n>> user.email if neither sendemail.from is configured nor --from option\n>> is specified. And even when it does, it prompts the user so it can be\n>> changed if called from a terminal. So I wouldn't say that it assumes\n>> anything about the \"physicalness\" of user.email, it just uses it's as\n>> the most sane default unless anything else has been specified.\n>\n> It's useless to spoof the From field because many email services won't\n> send it, a point I already covered in the email you quoted.\n>\n> When a patch is finally emailed, it's the From field that is used for\n> Author attribution.\n\nThe From-field isn't assumed to be a physical-address, but the\nFrom-header is. If the From-field and the From-header are identical,\nthe From-field doesn't get emitted. This is the same mechanism that is\nused when people forward patches from other authors, and there's no\nattempts to validate the From-field, only the From-header.\n\nSo no, the author-email shouldn't need to be a physical address as far\nas send-email is concerned.\n\n-- \nErik \"kusma\" Faye-Lund\n"},{"id":"137237","messageId":"4BA397B0.2000908@drmicha.warpmail.net","threadId":"23060","inReplyTo":"3f4fd2641003190757y39050691y3dc0ca08bd5196fb@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael J Gruber","fromEmail":"git@drmicha.warpmail.net","sentAt":"2010-03-19T15:26:40Z","receivedAt":"2010-03-19T15:26:40Z","isPatch":false,"sender":{"key":"git@grubix.eu","avatar":"https://avatars.githubusercontent.com/u/233215?v=4"},"body":"Reece Dunn venit, vidit, dixit 19.03.2010 15:57:\n\n[snip]\n> \n> So, I'll ask again: what problems are you trying to solve that\n> cannot be solved by mailmap?\n> \n> - Reece\n\n[Attention, conspiracy theories below!]\n\nThe problem seems to be that some people are interested in statistics,\nso some are interested in consistent author information, but this\nrequires others (the authors) to maintain this information, at least on\nlarge projects where this information cannot be kept consistent by a few\npeople. So, some people are looking for a way to enforce this on the\nothers... Of course, one could also rephrase this is as \"help authors\nmaintain their authorship information in a consistent way\" ;)\n\nMichael\n"},{"id":"137238","messageId":"alpine.DEB.2.00.1003190903540.3821@asgard.lang.hm","threadId":"23060","inReplyTo":"4BA397B0.2000908@drmicha.warpmail.net","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"","fromEmail":"david@lang.hm","sentAt":"2010-03-19T16:05:46Z","receivedAt":"2010-03-19T16:05:46Z","isPatch":false,"sender":{"key":"david@lang.hm","avatar":null},"body":"On Fri, 19 Mar 2010, Michael J Gruber wrote:\n\n> Reece Dunn venit, vidit, dixit 19.03.2010 15:57:\n>\n> [snip]\n>>\n>> So, I'll ask again: what problems are you trying to solve that\n>> cannot be solved by mailmap?\n>>\n>> - Reece\n>\n> [Attention, conspiracy theories below!]\n>\n> The problem seems to be that some people are interested in statistics,\n> so some are interested in consistent author information, but this\n> requires others (the authors) to maintain this information, at least on\n> large projects where this information cannot be kept consistent by a few\n> people. So, some people are looking for a way to enforce this on the\n> others... Of course, one could also rephrase this is as \"help authors\n> maintain their authorship information in a consistent way\" ;)\n\nbut a UUID doesn't help you.\n\nif you can force people to have a consistant UUID, you can force them to \nhave a consistant e-mail address (and submit mapping updates if it \nchanges)\n\nif you can't force people to maintain a consistant e-mail, why do you \nthink they would maintain a consistant UUID?\n\nDavid Lang\n"},{"id":"137240","messageId":"alpine.DEB.2.00.1003190959410.10411@asgard.lang.hm","threadId":"23060","inReplyTo":"4BA3855A.1070005@alum.mit.edu","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"","fromEmail":"david@lang.hm","sentAt":"2010-03-19T17:02:15Z","receivedAt":"2010-03-19T17:02:15Z","isPatch":false,"sender":{"key":"david@lang.hm","avatar":null},"body":"On Fri, 19 Mar 2010, Michael Haggerty wrote:\n\n> Michael Witten wrote:\n>> On Fri, Mar 19, 2010 at 02:41, Michael Haggerty wrote:\n>>> Michael Witten wrote:\n>>>> Rather than use a (name,email) pair to identify people, let's use\n>>>> a (uuid,name,email) triplet.\n>>>> [...]\n>>> A UUID doesn't need to be a big hex number. All it has to be is a\n>>> \"Universally Unique Identifier\". Like, oh, for example, your\n>>>\n>>> *** EMAIL ADDRESS ***\n>>>\n>>> [1]. There is even already a way to fix up mistakes or unavoidable\n>>> email address changes, namely the .mailmap file.\n>>\n>> *facepalm*\n>>\n>> You've just repeated everything that I've said; go look at the rest of\n>> the thread, where I spend plenty of time correcting the same hangups\n>> about my choice of the word UUID and my use of hex digits.\n>\n> No, my point is to use the *existing* email address as the UUID\n> *without* adding another field. Nothing needs to be changed!\n\nif you are now proposing using the e-mail address, that already \nexists and is supported by the tools, it sounds like you are just \nwithdrawing your proposal (other than possibly proposing that the e-mail \nfield gets renamed to UUID????)\n\nDavid Lang\n"},{"id":"137241","messageId":"b4087cc51003191006s6aa781c1k5c7fd476cc0bceb1@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.DEB.2.00.1003190959410.10411@asgard.lang.hm","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T17:06:34Z","receivedAt":"2010-03-19T17:06:34Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 11:02, wrote:\n>\n> if you are now proposing using the e-mail address, that already exists and\n> is supported by the tools, it sounds like you are just withdrawing your\n> proposal (other than possibly proposing that the e-mail field gets renamed\n> to UUID????)\n\nYou're responding to a different Michael.\n"},{"id":"137244","messageId":"b4087cc51003191016u3d4af923n613f3feca0e6e222@mail.gmail.com","threadId":"23060","inReplyTo":"alpine.DEB.2.00.1003190903540.3821@asgard.lang.hm","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-19T17:16:15Z","receivedAt":"2010-03-19T17:16:15Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Fri, Mar 19, 2010 at 10:05, wrote:\n>\n> if you can force people to have a consistant UUID, you can force them to\n> have a consistant e-mail address (and submit mapping updates if it changes)\n>\n> if you can't force people to maintain a consistant e-mail, why do you think\n> they would maintain a consistant UUID?\n\nFirstly, please note that a UUID is defined in this context as any\nstring that the user deems for himself to be uniquely identifying of\nhimself; a UUID allows a user to determine his canonical\nrepresentation from the very start.\n\nThere's no forcing; there can't be. This is meant to help users manage\ntheir own identities.\n\nA UUID is basically only subject to change due to:\n\n * typos when configuring\n\nA name/email pair (as in the user.name and user.email variables) is\nsubject to change due to:\n\n * typos when configuring\n * legal name changes\n * email account switching\n\nNaturally, older commits and wrong UUIDs would need mappings, but\nthat's no different than the current situation except for the fact\nthat UUIDs would not change as frequently.\n\nThat aside, an alternative solution that is not as powerful but that\nis less invasive would be to allow users to transmit authorship\ninformation as part of the patch payload separate from the usual email\nheaders (or something like this). Erik Faye-Lund suggests this is\nalready easily done, but I'm not so sure.\n"},{"id":"137263","messageId":"7vvdcsje3t.fsf@alter.siamese.dyndns.org","threadId":"23060","inReplyTo":"alpine.LFD.2.00.1003181256310.18017@i5.linux-foundation.org","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Junio C Hamano","fromEmail":"gitster@pobox.com","sentAt":"2010-03-19T19:39:02Z","receivedAt":"2010-03-19T19:39:02Z","isPatch":false,"sender":{"key":"gitster@pobox.com","avatar":"https://avatars.githubusercontent.com/u/54884?v=4"},"body":"Linus Torvalds writes:\n\n> (And yes, it does say that git should probably have errored out way more \n> aggressively about badly set up host/domain names in the \"guess at email \n> address\" code. My bad. Maybe it's still worth fixing for the future)\n\nWe made a small step in that direction in 49ff9a7 (commit: show\ninteresting ident information in summary, 2010-01-13). I think what it\ndoes is sufficiently loud (but not annoying).\n"},{"id":"137287","messageId":"201003200121.02560.jnareb@gmail.com","threadId":"23060","inReplyTo":"b4087cc51003190740h680b5dech4edd7a5000f180ee@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2010-03-20T00:21:00Z","receivedAt":"2010-03-20T00:21:00Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On Fri, 19 Mar 2010, Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 08:08, Jakub Narebski wrote:\n\n>> This is non-solution to non-problem.\n>>\n>> First, the user.name and user.email does not need to be name and email\n>> from some email account.  It might be some \"canonical name\" and\n>> \"canonical email\".\n> \n> The vast majority of patches come in through email; the git tools\n> expect the user.name and user.email to reflect physical email account\n> information.\n> \n> You would be correct if it were not for the fact that git currently\n> conflates identity and current email system.\n\nIt is not true. From the git-config(1) manpage, the description (meaning)\nof user.name and user.email is:\n\n user.email::\n Your email address to be recorded in any newly created commits.\n Can be overridden by the 'GIT_AUTHOR_EMAIL', 'GIT_COMMITTER_EMAIL', and\n 'EMAIL' environment variables. See linkgit:git-commit-tree[1].\n\n user.name::\n Your full name to be recorded in any newly created commits.\n Can be overridden by the 'GIT_AUTHOR_NAME' and 'GIT_COMMITTER_NAME'\n environment variables. See linkgit:git-commit-tree[1].\n \nAs you can see there is nothing about email, and physicsl email account.\n\nIt is true that git-send-email asks about the \"From\" email address to\nsend email from with user.name + user.email as default value...\nunless either sendemail.from or --from option is used. \n[See also below].\n\n>> Second, there are (I think) two main sources of 'unstability' in\n>> (name,email) pairs, namely A) misconfigured git (when fetching/pushing\n>> using git itself), B) wrong name in email etc. (when sending patches\n>> via email, 80% of patches in Linux kernel case).\n>>\n>> In the case of misconfigured git (case A) using UUID wouldn't help,\n>> and only make it worse (you would have to configure the same UUID on\n>> each machine).  What would help here is for git to be more strict and\n>> perhaps forbid (some of) autogenerated names and emails.\n> \n> The uuid string would be typed pretty much only during configuration;\n> from there, it's basically just handled by the git tools. Hence, the\n> uuid can indeed suffer from typos, but the name/email pair can suffer\n> from not only typos but also real life name changing and email account\n> switching.\n\nYou do not need (in theory at least) to change user.name nor user.email\nwith real life name changing (like marriage or adoption) and email \naccount switching.\n\n[...]\n>> In the case of sending patches via email, you can use in-body 'From:'\n>> to provide (name,email) part that is different than account used to\n>> send email.\n> \n> That's a good solution that I've considered, except for 2 reasons:\n> \n> * It involves much more opportunities for typos and/or the\n> configuration of a non-git tool for a git-specific purpose.\n> \n> * Many if not most email services will refuse to send messages\n> with forged/spoofed email addresses.\n\nActually git-send-email would automatically add in-body \"From:\" header\nif it is different from the \"From:\" address for email, and git-am would\nautomatically prefer in-body \"From:\" over sender (in-header \"From:\")\nfor authorship information.\n\nSender can be different from author of the patch, there is no problem\nwith that.\n\nWhat git can improve here (and perhaps already does it) is handling of\nnon-ASCII characters in name (e.g. when commit message does not contain\nnon US-ASCII letters, but user.name does). Perhaps it got corrected\n(improved) already.\n\n\nP.S. Backward compatibility (older git-am) would probably require\nUUID in the form of canonical name+email, and use of in-body \"From:\"\nheader to pass this UUID when sending patches.\n\n>> In the case of UUID you would need the same: some way to\n>> provide UUID in patch (in email).\n> \n> Yes, but that's automated by tools like git's format-patch. Not using\n> something like format-patch or some other git interface is an\n> 'out-of-band' communication and that author has essentially chosen not\n> to care about his identity.\n> \n> The use of the uuid field and allowing git tools to handle it is just\n> a way to give a person who does care about his identity to keep it\n> consistent.\n\ngit-send-email *already* automatically deals with sender != author.\n\n[...]\n>> What could help in both cases is .mailmap being used (perhaps on\n>> demand) in more git commands.  See Documentation/mailmap.txt\n>> or e.g. git-shortlog(1) manpage.  It is quite advanced tool for\n>> correcting mistakes (it can correct *both* user name, which is\n>> most common usage, but also email address).\n> \n> The disadvantage here is that it centralizes identity management and\n> it is more demanding because the name/email pair is quite unstable.\n\nHow in-tree .mailmap file (in-tree like .gitignore and .gitattributes)\nis *centralized identity management*? It is as distributed as git\nrepositories are.\n\nOn the other hand user.uuid is not distributed; for security reasons\nconfig is not transferred.\n\n[...]\n> [...], and with\n> some clever encoding some statistics gathering programs could\n> (possibly) run more efficiently.\n\nWell, I guess it is statistics that dominates, not id part. Such\ntools shoud simply take .mailmap into account (unless they rely on\ngit for that.).\n\n-- \nJakub Narebski\nPoland\n"},{"id":"137544","messageId":"20100322120649.GA28003@sirena.org.uk","threadId":"23060","inReplyTo":"b4087cc51003190509y6ce7ad9dy992254cfe6ac2fb3@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Mark Brown","fromEmail":"broonie@sirena.org.uk","sentAt":"2010-03-22T12:06:49Z","receivedAt":"2010-03-22T12:06:49Z","isPatch":false,"sender":{"key":"broonie@sirena.org.uk","avatar":"https://gravatar.com/avatar/9e798c729a4a709279df497d9608ad68422755c9670e92434a5436f7e607cf86?d=mp&s=160"},"body":"On Fri, Mar 19, 2010 at 06:09:34AM -0600, Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 05:54, Mike Hommey wrote:\n\n> > While I don't agree with the need for that uuid thing, I'd like to\n> > pinpoint that people who care can't necessarily make their e-mail\n> > consistant. For example, Linus used to use an @osdl.org address, and\n> > he now uses an @linux-foundation.org address. It's still the same Linus,\n> > but the (name, email) pair has legitimately changed.\n\n> Indeed.\n\n> This is because the name/email pair (as in the 'name' and 'email'\n> config variables) CONFLATES the idea of identity and current email\n> account.\n\nYou're assuming they aren't conflated - for example, when people do work\nboth personally and for their employer they often use distinct e-mail\naddresses to identify how the work was funded.\n"},{"id":"137555","messageId":"4ba780dc.5744f10a.5517.4c2d@mx.google.com","threadId":"23060","inReplyTo":"20100322120649.GA28003@sirena.org.uk","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-22T14:38:20Z","receivedAt":"2010-03-22T14:38:20Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Mon, Mar 22, 2010 at 06:06, Mark Brown wrote:\n>\n> You're assuming they aren't conflated - for example, when people do work\n> both personally and for their employer they often use distinct e-mail\n> addresses to identify how the work was funded.\n\nIndeed.\n\nThe model I propose handles this case much better, as I explain here:\n\n http://marc.info/?l=git&m=126900051102958&w=2\n\nSpecifically:\n\n > if they do need to set it on multiple machines and\n > can't be bothered to keep their e-mail consistant,\n > why would they bother keeping this additional thing\n > considtant? Linus is pointing out that people don't\n > care now about their e-mail and name, and will care\n > even less about some abstract UUID\n \n The user doesn't have a damn choice!\n\n [These first few paragraphs aren't completley correct;\n there's an explanation below them. It's mainly just\n setting up for the important part below.]\n \n The email can't be kept consistent over time because\n the tools expect it to be and/or use the actual\n physical email used to send/receive stuff. It's\n information that CONFLATES identity with whatever\n tool/system you're using.\n \n For instance, Michael Haggerty cannot reasonably use\n \n [user]\n name = Michael Haggerty\n email = mhagger@MIT.EDU\n \n because he likely no longer has that email account\n to use. He is forced to change it and therefore\n forced to make his identity confused.\n\n [The above isn't quite true; my mistake. Michael\n could actually keep \"mhagger@MIT.EDU\" but inform\n tools like \"git send-email\" to send patches from\n another email address; this way, send-email will\n emit the necessary information to carry that\n authorship identity (\"mhagger@MIT.EDU\") along\n with the patch.\n \n However, it's still the case that Michael Haggerty\n is essentially stuck with \"mhagger@MIT.EDU\" for\n his identification---a problem that my proposal\n essentially fixes, as described now:]\n \n I'm proposing ALLOWING him to say:\n \n [user]\n uuid = Michael Haggerty \n name = Michael Haggerty\n email = mhagger@ALUM.mit.edu\n \n Heck, let's say he works at Red Hat as well; he\n might make some commits under this config AT WORK:\n \n [user]\n uuid = Michael Haggerty \n name = Michael Haggerty\n email = mhagger@redhat.com\n \n Then, he can make, say, commits to the Linux kernel\n repo for both work and hobby related issues and\n still be recognized as the same person.\n \n That is, he can have some commits [publicly] under:\n \n Michael Haggerty \n \n and other commits [publicly] under:\n \n Michael Haggerty \n \n and still link them all together as the [SAME PERSON]\n with just the uuid:\n \n Michael Haggerty \n\nThe idea is to help users manage their own identities more effectively.\n\nIt's clearly advantageous to be able to apply different public identities\n(personal vs. work identity for instance) to different commits, tags, etc.,\nbut it's also advantageous to be able to link those different identities\ntogether.\n\nAt the moment, the different identities can only be linked together by\nediting and transmitting a .mailmap file to be used by git tools. My\nproposal distributes this kind of work UPFRONT by having individuals\nchoose UPFRONT some reasonably unique identification string to use\nas the link between public identities.\n\nSincerely,\nMichael Witten\n"},{"id":"137720","messageId":"4BAA5EF7.6000906@redhat.com","threadId":"23060","inReplyTo":"b4087cc51003191006s6aa781c1k5c7fd476cc0bceb1@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Avi Kivity","fromEmail":"avi@redhat.com","sentAt":"2010-03-24T18:50:31Z","receivedAt":"2010-03-24T18:50:31Z","isPatch":false,"sender":{"key":"avi@redhat.com","avatar":null},"body":"On 03/19/2010 07:06 PM, Michael Witten wrote:\n> On Fri, Mar 19, 2010 at 11:02, wrote:\n> \n>> if you are now proposing using the e-mail address, that already exists and\n>> is supported by the tools, it sounds like you are just withdrawing your\n>> proposal (other than possibly proposing that the e-mail field gets renamed\n>> to UUID????)\n>> \n> You're responding to a different Michael.\n> \n\nI guess he should have checked the UUID.\n\n-- \nDo not meddle in the internals of kernels, for they are subtle and quick to panic.\n"},{"id":"137725","messageId":"40aa078e1003241218s753c1986r3dde7281c513de07@mail.gmail.com","threadId":"23060","inReplyTo":"4ba780dc.5744f10a.5517.4c2d@mx.google.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Erik Faye-Lund","fromEmail":"kusmabite@googlemail.com","sentAt":"2010-03-24T19:18:35Z","receivedAt":"2010-03-24T19:18:35Z","isPatch":false,"sender":{"key":"kusmabite@gmail.com","avatar":"https://avatars.githubusercontent.com/u/47073?v=4"},"body":"On Mon, Mar 22, 2010 at 3:38 PM, Michael Witten wrote:\n>     However, it's still the case that Michael Haggerty\n>     is essentially stuck with \"mhagger@MIT.EDU\" for\n>     his identification---a problem that my proposal\n>     essentially fixes, as described now:]\n>\n>    I'm proposing ALLOWING him to say:\n>\n>        [user]\n>            uuid  = Michael Haggerty \n>            name  = Michael Haggerty\n>            email = mhagger@ALUM.mit.edu\n>\n\n...which is the exact same situation as above, where he's \"stuck\"\nusing \"mhagger@MIT.EDU\" for identification. I don't see how this\nchanges anything (except allowing to distribute an updated\ncontact-email... But let's face it, git-repos aren't Facebook)\n\n-- \nErik \"kusma\" Faye-Lund\n"},{"id":"137727","messageId":"b4087cc51003241223l4a7feed9m3d1d0e83c38a7e94@mail.gmail.com","threadId":"23060","inReplyTo":"40aa078e1003241218s753c1986r3dde7281c513de07@mail.gmail.com","subject":"Re: What's in a name? Let's use a (uuid,name,email) triplet","fromName":"Michael Witten","fromEmail":"mfwitten@gmail.com","sentAt":"2010-03-24T19:23:38Z","receivedAt":"2010-03-24T19:23:38Z","isPatch":false,"sender":{"key":"mfwitten@gmail.com","avatar":"https://avatars.githubusercontent.com/u/597101?v=4"},"body":"On Wed, Mar 24, 2010 at 13:18, Erik Faye-Lund wrote:\n> I don't see how this\n> changes anything\n\nI don't see how you can't see it.\n\nOh well.\n"}]}