{"thread":{"id":"10137","subject":"Git User's Survey 2007 unfinished summary (long)","startedAt":"2007-10-04T09:12:55Z","lastAt":"2007-10-05T07:57:30Z","messageCount":8,"participants":["Jakub Narebski","Benoit SIGOURE","Johannes Schindelin","Shawn O. Pearce","David Tweed","Andreas Ericsson"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"54805","messageId":"8fe92b430710040212u332bb4e5yfa6c10b8236c6301@mail.gmail.com","threadId":"10137","inReplyTo":null,"subject":"Git User's Survey 2007 unfinished summary (long)","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2007-10-04T09:12:55Z","receivedAt":"2007-10-04T09:12:55Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"This is partial summary of Git User's Survey 2007,\nending at state from 28 September 2007.\n\nThe survey can be found here:\n http://www.survey.net.nz/survey.php?94e135ff41e871a1ea5bcda3ee1856d9\n http://tinyurl.com/26774s\n\n----\nThere were 683 individual responses\n\nAbout you\n~~~~~~~~~\n\n01. What country are you in?\n\n Answer | Count\n ------------------------------------------\n Algeria | 1\n Argentina | 3\n Australia | 25\n Austria | 9\n Belgium | 5\n Brazil | 20\n Bulgaria | 1\n Canada | 44\n Chile | 2\n China | 4\n Colombia | 2\n Czech Republic | 10\n Denmark | 7\n Ecuador | 1\n Estonia | 1\n European Union | 1\n Finland | 23\n France | 36\n Germany | 64\n Greece | 3\n Hungary | 2\n India | 13\n Ireland | 2\n Israel | 6\n Italy | 14\n Japan | 4\n Jersey | 1\n Latvia | 1\n Malaysia | 1\n Mexico | 1\n Netherlands | 15\n New Zealand | 5\n Norway | 14\n Philippines | 3\n Poland | 6\n Portugal | 2\n Puerto Rico | 1\n Romania | 1\n Russian Federation | 6\n Samoa | 1\n Serbia | 1\n Singapore | 2\n Slovak Republic | 1\n Slovenia | 2\n South Africa | 4\n Spain | 11\n Sri Lanka | 1\n Sweden | 14\n Switzerland | 15\n UK / US | 1\n United Kingdom | 40\n United States of America | 218\n Venezuela | 1\n Vietnam | 1\n ------------------------------------------\n Base | 673 / 683\n Total (sum) | 673\n\nEngland and Scotland counts as United Kingdom here.\nTable is sorted in alphabetical order.\n\nAs one can easily see, slightly less than third of GIT users\nare in the USA (those who answered this survey).\n\n\n02. What is your preferred non-programming language?\n\nThis is multiple answers question, although most people\ngave only one preferred language.\n\n Answer | Count\n ------------------------------------------\n Afrikaans | 1\n Bulgarian | 1\n Castellano | 2\n Catalan | 1\n Chinese | 2\n Czech | 10\n Danish | 6\n Dutch | 12\n English | 416\n Finnish | 16\n French | 33\n Galician | 1\n German | 58\n Greek | 2\n Hebrew | 1\n HibernoEnglish | 1\n Hungarian | 3\n Italian | 9\n Japanese | 1\n LSF (French sign language) | 1\n Norwegian | 4\n Polish | 5\n Portuguese | 11\n Romanian | 1\n Russian | 13\n Serbian | 1\n Slovenian | 2\n Spanish | 13\n Swedish | 13\n Swiss | 1\n Ukrainian | 1\n Vietnamese | 1\n ------------------------------------------\n invalid (computer language) | 37\n not understood | 4\n ------------------------------------------\n Base | 662 / 683\n Total (sum) | 684\n\nThe question itself is not well formulated, as one can see from the\nnumber of answers with computer language, and \"not understood\"\nanswers. I am not native English speaker, but there were suggestions\nto use \"natural language\" instead of \"non-programming language\".\n\nAround two third of git users prefer English language, at least for\ndealing with computers.\n\n\n03. How old are you?\n\n Answer | Count\n ------------------------------------------\n < 18 | 11\n 18-21 | 75\n 22-25 | 174\n 26-30 | 203\n 31-40 | 146\n 41-50 | 45\n 51-75 | 13\n 76+ | 0\n ------------------------------------------\n Base | 667 / 683\n Total (sum) | 667\n\nYoungest git user who answered this survey is 14 years old,\noldest is 74 years old. This is quite a span, I'd say, The age of 25\ngot most count (51 answers).\n\n\n04. Which programming languages you are proficient with?\n\n Answer | Count\n ------------------------------------------\n C | 582\n shell | 449\n Perl | 244\n Python | 316\n Tcl/Tk | 26\n ------------------------------------------\n Base | 648 / 683\n Total (sum) | 1617\n\nThe choices include programming languages used by git. This is\nmultiple choice question (you can be proficient in more than one\nprogramming language).\n\nIt look like there is only around 3/4 people proficient in Perl as\ncompared to Python; it looks like Python is more popular. C is most\npopular; shell is more popular than Perl or Python. The fewest people\nare proficient in Tcl/Tk. I'm sorry, git-gui and gitk guys; it looks\nlike not many developers... around 4% of git users.\n\n\nGetting started with GIT\n~~~~~~~~~~~~~~~~~~~~~~~~\n\n05. How did you hear about GIT?\n\n Answer | Count\n ------------------------------------------\n LKML | 109\n LWN (Linux Weekly News) | 39\n KernelTrap | 15\n KernelTraffic | 1\n kernel.org | 9\n freedesktop.org | 5\n Linus presentation at Google | 48\n Slashdot | 28\n blog | 19\n community site[*1*] | 12\n news site | 34\n\n searching Internet[*2*] | 6\n other SCM / SCM research[*3*] | 20\n Internet | 32\n IRC | 6\n\n Linux kernel uses it | 73\n some project uses git | 47\n\n developer by name[*4*] | 21\n friend | 39\n word of mouth | 15\n work / coworker | 22\n\n initial GIT announcement | 12\n BitKeeper news | 24\n ------------------------------------------\n don't remember | 13\n other / uncategorized | 44\n ------------------------------------------\n Base | 658 / 683\n Total (sum) | 693\n\n[*1*] Community site are sites like Digg, Reddit and \"planet\" sites.\n[*2*] This includes answer of \"Google\"\n[*3*] This includes some other SCM mailing list, VCS comparison,\n and searching for an SCM.\n[*4*] Linus Torvalds, Carl Worth, Keith Packard, Randal Schwartz,...\n\nThis was free-form question, and tabularizing answers was quite a work.\nIt is taken as multiple choice question (for example link to Linus\npresentation at Google found at Slashdot).\n\nOther / uncategorized includes for example GoogleTalk IM, 3 answers\nIIRC.\n\nNote that Linus Torvalds presentation at Google / YouTube got it's\nown category, and generated quite a bit of git users.\n\n\n06. Did you find GIT easy to learn?\n\n Answer | Count\n ------------------------------------------\n very easy | 38\n easy | 136\n reasonably | 318\n hard | 131\n very hard | 33\n ------------------------------------------\n Base | 656 / 683\n Total (sum) | 656\n\nNice gaussian curve. Most users find it reasonably easy to use.\nOn the other hand git is not considered easy...\n\n\n07. What helped you most in learning to use it?\n\n TO DO\n 646 / 683 non-empty responses\n\n\n08. What did you find hardest?\n\n TO DO\n 596 / 683 non-empty responses\n\n\n09. When did you start using git? From which version?\n\n Answer | Count\n ------------------------------------------\n (no version string) | 165\n 0.99x | 26\n 0.x | 12\n 1.0x | 31\n 1.1x | 9\n 1.2x | 12\n 1.3x | 22\n 1.4x | 147\n 1.5x | 198\n ------------------------------------------\n Base | 626 / 683\n Total (sum) | 626\n\nNOTE! This table shows _only_ answers in which there was given git\nversion explicitely. Some people gave date, some people wrote how long\nthey have used git. Those answers needs also processing; they are\nskipped here.\n\nIt looks like the git users community is divided into part of users\nwho started using it from beginning, or almost from beginning, and\nusers which started using git post v1.3.0 (post e.g. making separate\nremotes the default layout of branches).\n\n\nOther SCMs\n~~~~~~~~~~\n\n10. What other SCMs did/do you use?\n\nThis question is not well thought, as it gathers together (in the\nfree-form which is not easy to tabularize with large number of\nresponses we got) SCMs which one used and no longer uses, SCMs which\nare used in parallel with git, and SCMs which are used instead of git\n(which are chosen as main SCM for a project). Nevertheless it shows\nwith which VCS, and its conceps, are users familiar with.\n\n Answer | Count\n ------------------------------------------\n AccuRev | 3\n Aegis | 1\n Bazaar | 19\n Bazaar-NG | 50\n BitKeeper | 27\n CCC | 1\n CMS (Digital) | 1\n CMS (VAX) | 1\n CMS (VMS) | 1\n CVCS | 1\n CVS | 454\n ClearCase | 43\n CodeMgr | 1\n Continuus | 1\n Darcs | 78\n DesignSync | 1\n GNU Arch | 57\n Mercurial | 92\n Monotone | 31\n Omniworks | 1\n OpenCM | 1\n PRCS | 1\n PVCS | 12\n Perforce | 50\n Quilt | 2\n RCS | 61\n SCCS | 18\n SCM | 1\n SCSS | 1\n SVK | 19\n Serena Version Manager | 1\n SourceForge | 1\n Sourcerer's Apprentice | 1\n StarTeam | 4\n Subversion | 524\n Sun NSE | 2\n Sun TeamWare | 4\n VCS | 1\n VMS | 1\n VSS | 26\n 'cp -a' | 1\n akpm patch scripts | 1\n custom in-house tools | 1\n diff patch | 2\n none | 9\n notes-on-paper-made-by-hand | 1\n really horrible stuff | 1\n scripts for 'shadow trees' | 1\n tarballs | 1\n tlib | 1\n undisclosed | 1\n ------------------------------------------\n Base | 654 / 683\n Total (sum) | 1615\n\nThe above table is in alphabetical order. It was generated from\nfree-form answers, tabularized as multiple choice answer.\n\nNote that this question does not distinguish between SCMs/VCSs which\nwere used prior to Git and used no longer, SCMs which are used beside\n(in parallel) to Git perhaps interacting with Git, and SCMs which are\nused instead of Git. Also note that this is _Git User's_ survey, so it\nthose number for example do not represent number of e.g. users of\nMercurial as compared to e.g. users of Subversion.\n\nBelow there is table of SCM used, sorted by the number of responses.\nNote that annotations (like \"a little CVS\") were not weighted here.\nOnly SCMs which has count more that 10 are shown. One person can (and\nusually did) chose more than one SCM.\n\n Answer | Count\n ------------------------------------------\n Subversion | 524\n CVS | 454\n Mercurial | 92\n Darcs | 78\n RCS | 61\n GNU Arch | 57\n Bazaar-NG | 50\n Perforce | 50\n ClearCase | 43\n Monotone | 31\n BitKeeper | 27\n VSS (MS Visual SourceSafe) | 26\n Bazaar | 19\n SVK | 19\n SCCS | 18\n PVCS | 12\n\n tla+baz+bzr | 129\n ------------------------------------------\n Base | 654 / 683\n\nAs you can see two most popular SCMs are Subversion ('svn') and CVS,\nwith Subversion being a bit more popular. Among distributed SCMs\nwith most count are Mercurial ('hg') and Arch and its descendants\n('tla', 'baz', 'bzr'). From proprietary (non-OSS) revision control\nsystems Perforce ('p4'), ClearCase (and ClearCase UCM), BitKeeper ('bk')\nand Visual SourceSafe (aka. that awful M$ one ;-) got most count.\n\nNote that the count for given version control system does not reflect\n_preferences_ of git users. One can be forced to use specified SCM.\n\nSee also question 35: \"How does GIT compare to other SCM?\".\n\n\n11. Why did you choose GIT?\n\n TO DO\n 643 / 683 non-empty responses\n\n\n12. Why did you choose other SCMs?\n\n TO DO\n 606 / 683 non-empty responses\n\n\n13. What would you require from GIT to enable you to change,\n if you use other SCM for your project?\n\n TO DO\n 474 / 683 non-empty responses\n\n\n14. Did you import your repository from foreign SCM? What SCM?\n\n Answer | Count\n ------------------------------------------\n N/A | 15\n No | 169\n Yes | 372\n ------------------------------------------\n Base | 556 / 683\n Total (sum) | 556\n\nThis is anly partial analysis, as it deals only with first part of\nquestion (which, because of second part, has free-form structure and\nneeded processing).\n\nOne can see that around half of git users have imported (at least one\nproject) from foreign SCM.\n\n\n15. What tool did you use for import?\n\n Answer | Count\n ------------------------------------------\n by hand | 7\n custom script | 21\n fast-import script | 3\n\n Tailor | 28\n -------------------------------------------\n git-cvsimport | 81\n parsecvs | 12\n fromcvs | 2\n cvs2git | 1\n cvstogit | 1\n\n git-svn | 150\n git-svnimport | 66\n\n git-archimport | 15\n bk2git (customized) | 1\n darcs2git | 4\n git-p4 | 4\n git-p4import | 1\n git-ucmimport | 1\n hg-to-git | 2\n hg2git | 2\n hgpullsvn | 1\n hgsvn | 1\n moin2git | 1\n ------------------------------------------\n unspecified | 18\n N/A | 114\n ------------------------------------------\n Base | 467 / 683\n Total (sum) | 538\n\n\n16. Do your GIT repository interact with other SCM? Which SCM?\n\n Answer | Count\n ------------------------------------------\n N/A | 35\n No | 228\n Yes | 228\n ------------------------------------------\n Base | 491 / 683\n Total (sum) | 491\n\nThis is anly partial analysis, as it deals only with first part of\nquestion (which, because of second part, has free-form structure and\nneeded processing).\n\nOne can see that around third of git users interacts (for at least one\nproject) with foreign SCM, as compared to half of git users which have\nimported other SCM.\n\n\n17. What tool did/do you use to interact?\n\n Answer | Count\n ------------------------------------------\n by hand | 10\n custom script | 16\n\n Tailor | 4\n convert-repo | 1\n\n fromcvs | 1\n git-cvsexportcommit | 8\n git-cvsimport | 19\n git-cvsserver | 2\n\n git-svn | 164\n git-svnimport | 2\n\n git-p4 | 4\n git-p4import | 1\n -----------------------------------------\n unspecified | 2\n N/A | 153\n ------------------------------------------\n Base | 385 / 683\n Total (sum) | 388\n\nThe only tool which really allows to interact (two-way) with other SCM\nis git-svn (164 / 232).\n\n\nHow you use GIT\n~~~~~~~~~~~~~~~\n\n18. Do you use GIT for work, unpaid projects, or both?\n\n Answer | Count\n ------------------------------------------\n work | 56\n unpaid projects | 212\n both | 377\n\n work + both | 433\n ------------------------------------------\n Base | 645 / 683\n Total (sum) | 645\n\nAround two third of people use git at work, or for work.\n\nSee also question 55: \"Would commerical (paid) support from a support\nvendor be of interest to you/your organization?\"\n\n\n19. How do you obtain GIT?\n\n Answer | Count\n ------------------------------------------\n binary package | 283\n source tarball | 210\n pull from main repository | 153\n ------------------------------------------\n Base | 646 / 683\n Total (sum) | 646\n\nAround half more people use binary packages than source tarball (or\nsource package, I think). More than half of people compile its own git\n(pull is also followed by compilation).\n\nIn earlier partial survey summary, from 27-08-2004 (a month earlier)\nthere were twice as many people installing git from binary packages.\n\n\n20. What hardware platforms do you use GIT on?\n (on Unices: result of \"uname -i\")\n\nNote: 'uname -i' does not work on all Unices. I'm sorry for this\nmistake.\n\n Answer | Count\n ------------------------------------------\n Intel | 73\n Athlon | 2\n i386 | 171\n i586 | 9\n i686 | 60\n IA-32 | 6\n IA-64 | 8\n AMD | 35\n amd64 | 48\n x86 | 156\n x86-64 | 112\n\n Apple | 8\n iMac | 1\n MacBook | 7\n PowerBook | 10\n PPC | 52\n ppc64 | 7\n\n ARM | 6\n Alpha | 2\n PA-RISC | 1\n parisc64 | 1\n MIPS | 1\n mips64 | 1\n mipsel | 2\n SPARC | 8\n sparc64 | 6\n SUNW | 6\n Sun-Fire | 4\n sun4u | 3\n sun4v | 1\n k8 | 1\n ------------------------------------------\n unknown | 67\n ------------------------------------------\n Base | 637 / 683\n Total (sum) | 875\n\nThe above table is in some arbitrary order. It was generated from\nfree-form answers, tabularized as multiple choice answer, as one\nperson can use git on more than one architecture.\n\nThose results probably needs further processing to reduce number of\nchoices, by gathering architectures.\n\n\"Unknown\" usually means that something else instead of architecture\nwas given (like operating system), or architecture was too generic,\nlike \"PC\".\n\n\n21. What OS (please include the version) do you use GIT on?\n\n Answer | Count\n ------------------------------------------\n AIX | 1\n FreeBSD | 16\n OpenBSD | 3\n HP-UX | 1\n Linux | 582\n MS Windows (Cygwin) | 22\n MS Windows (msys) | 1\n MS Windows (unsp.) | 35\n MacOS X / Darwin | 94\n Solaris | 11\n SunOS | 5\n UNIX (unsp.) | 1\n too many to list | 1\n ------------------------------------------\n MS Windows (together) | 58\n FreeBSD / OpenBSD | 19\n Unices (together) | 19\n ------------------------------------------\n Base | 645 / 683\n Total (sum) | 775\n\nThe above was generated from free-form answers, tabularized as\nmultiple choice answer, as one person can use git on more than one\noperating system.\n\nThis is only rough analysis, because it does not include operating\nsystem version, or for Linux distribution(s) used.\n\n\n22. What projects do you track (or download) using GIT (or git web interface)?\n\n TO TABULARIZE\n 560 / 683 non-empty responses\n\n\n23. How many people do you collaborate with using GIT?\n\n TO TABULARIZE\n 575 / 683 non-empty responses\n\nWhat ranges should be used here: 1, 2-9, 10-99, 100-999, 1000+ or\nperhaps 1, 2-5, 6-15, 16-25, 26-50, 50-100, 100-1000, 1000+, or yet\nanother?\n\n\n24. How big are the repositories that you work on?\n\n TO DO\n 525 / 683 non-empty responses\n\nSome git repositories have more than 50k files, more than 150k\ncommits, more than 100mb largest file (although not single directory\nhas all those).\n\n\n25. How many different projects do you manage using GIT?\n\n TO TABULARIZE\n 577 / 683 non-empty responses\n\n\n26. Which porcelains do you use?\n\nMultiple answers (one can use more than one porcelain).\n\n Answer (multiple choice) | Count\n ------------------------------------------\n core-git | 558\n cogito (deprecated) | 56\n\n Patch management interface: : 57\n ...........................................\n StGIT | 41\n Guilt (formerly gq) | 13\n pg (deprecated) | 3\n\n own scripts | 95\n other | 14\n ------------------------------------------\n Base | 593 / 683\n Total (sum) | 780\n\nThose 14 \"other\" answers make me wish to have provided \"if other,\nwhat it was?\" (sub)question; actually not only for this question.\n\n\nIt is understandable that Cogito still has some users, even though it is\ndeprecated, and [I think] all of its functionality can be found in\ngit-core porcelain. It was meant as SCM / porcelain layer when git-core\ndidn't have it and consisted almost only of plumbing commands.\n\nQuite a bit of people use patch management interface: StGIT, Guilt,\neven deprecated and abandoned pg (Patchy Git). StGIT has more users\nthan Guilt (formerly gq), although that might be caused by the fact\nthat StGIT was here longer... Some say that it is because of Guilt\nhaving non-standalone documentation; it needs reading hgbook, as Guilt\nconcepts are based on mq (Mercurial [patch] queues) extension.\n\nLarge number of users use their own scripts, more than any\nnon-standard porcelain. One wonders if this is a result of good\ngit scriptability.\n\n14 users choose other... and there is no \"what other\" question,\nunfortunately...\n\nIf you are reading this: what were those other porcelains?\n\n\n27. Which git GUI do you use?\n\nMultiple answers (one can use more than one GUI). Note that for the\nfirst week and a bit of survey \"CLI\" answer had no explanation that it\nmeans command line interface, so results might be bit skewed.\n\n Answer (multiple choice) | Count\n ------------------------------------------\n CLI (command line) | 398\n gitk | 347\n git-gui | 123\n qgit | 82\n gitview | 15\n giggle | 48\n tig | 41\n instaweb | 16\n (h)gct | 3\n qct | 3\n KGit | 6\n git.el | 31\n other | 15\n\n giggle + gitview | 63\n ------------------------------------------\n Base | 572 / 683\n Total (sum) | 1128\n\nAs one can see almost as many people use gitk as CLI. Most used GUI\nare gitk and git-gui, most probably because they are distributed with\ngit, and because they are portable. QGit is also quite popular,\nalthough GTK+ viewers, namely giggle and gitview are also quite\npopular (note that there might be instances of users using both giggle\nand gitview). I am a bit suprised about popularity of Giggle, I'd say.\n\nTig (text-mode interface for git) and git.el (GIT mode for Emacs) are\nalso quite popular.\n\nI wonder what are those 15 other GUI. Why oh why there were no \"What\nis this 'other GUI'?\" question.\n\n\n28. Which (main) git web interface do you use for your projects?\n\n Answer | Count\n ------------------------------------------\n gitweb | 382\n cgit | 7\n wit (Ruby) | 5\n git-php | 1\n other | 27\n ------------------------------------------\n Base | 422 / 683\n Total (sum) | 422\n\nAround two third (closer to 4/7) of git users use some kind of web\ninterface for their projects.\n\nMost use gitweb (which is distributed with git), 7 use cgit, 5 wit\n(Ruby), most probably projects sharing site with XMMS2, 1 git-php\n(I wonder who...), and there are 27 \"other\" answers, which I am most\ncurious about. What are they?\n\nSome answered \"other\" as \"N/A\" (meaning they do not use web interface)\ninstead, what it is not obvious, not answering this question. The\nexplanation that this is possible was added later during duration of\nsurvey.\n\n\n29. How do you publish/propagate your changes?\n\nMultiple choices, as one can use different methods for different\nprojects, for example pushing to public repo for a project maintained,\nand sending patches via email to participate in third person project\ndevelopment.\n\n Answer | Count\n ------------------------------------------\n push | 456\n pull | 220\n format-patch + email | 172\n pull request | 65\n bundle | 13\n\n other | 70\n ------------------------------------------\n Base | 589 / 683\n Total (sum) | 996\n\nHere \"other\" means just not listed workflow, although it would be also\ninteresting to know details.\n\nMost popular is push, I guess to public \"publishing\" repository\n(and/or probably to mirror repositories). It is twice as popular\nas next method, gathering more than 3/4 of users.\n\nPull was supposed to mean logging to public server and pulling\nchanges from private repo, not pulling someone other changes.\nIt is second most popular method.\n\nSending patches via email is two to three times as popular\nas sending pull request.\n\n\n30. Does git.git repository include code produced by you?\n\n Answer | Count\n ------------------------------------------\n Yes | 99\n No | 512\n ------------------------------------------\n Base | 611 / 683\n Total (sum) | 611\n\nAs it can be seen, only (or perhaps it is that many?) around\na 6th to 7th of git users participate in its development by\nproviding code.\n\n\nInternationalization\n~~~~~~~~~~~~~~~~~~~~\n\n31. Is translating GIT required for wider adoption?\n\n Answer | Count\n ------------------------------------------\n Yes | 49\n No | 383\n Somewhat | 158\n ------------------------------------------\n Base | 590 / 683\n Total (sum) | 590\n\nMore than half of responders doesn't think that translating GIT\nis required for wider adoption.\n\n\n32. What do you need translated?\n\n TO TABULARIZE\n 172 / 683 non-empty responses\n\n\n33. For what language do you need translation for?\n\nIn alphabetic order, free-form question, treated as multiple choice.\n\n Answer | Count\n ------------------------------------------\n Afrikaans | 1\n Belorussian | 1\n Chinese | 5\n Dutch | 2\n English[*1*] | 3\n Filipino | 1\n Finnish | 3\n French | 21\n German | 17\n Greek | 1\n Hebrew | 1\n Hindi | 1\n Hungarian | 1\n Italian | 4\n Japanese | 5\n Mandarin | 1\n Norwegian | 2\n Polish | 2\n Portuguese | 6\n Russian | 6\n Serbian | 2\n Slovenian | 1\n Spanish | 9\n Swedish | 2\n Tagalog | 1\n Ukrainian | 1\n ------------------------------------------\n Base | 143 / 683\n Total (sum) [*2*] | 100\n\n[*1*] Git messages and documentation _is_ in English\n[*2*] Some answers were: \"do not translate\".\n\nGerman and French are most popular. Spanish, Portuguese, Russian,\nChinese + Mandarin and Japanese have count 5 or more.\n\n\nWhat you think of GIT\n~~~~~~~~~~~~~~~~~~~~~\n\n34. Overall, how happy are you with GIT?\n\n Answer | Count\n ------------------------------------------\n unhappy | 13\n not so happy | 36\n happy | 179\n very happy | 302\n completely ecstatic | 112\n ------------------------------------------\n Base | 642 / 683\n Total (sum) | 642\n\nNice, git users are mostly very happy with git.\n13 grumpy users, hmmm...\n\n\n35. How does GIT compare to other SCM tools you have used?\n\n Answer | Count\n ------------------------------------------\n Better | 505\n Equal (comparable) | 96\n Worse | 30\n\n N/A | 52\n ------------------------------------------\n Base | 631 / 683\n Total (sum) | 631\n\nClearly Git is superior SCM! (In the minds of Git users at least) ;-)\nSeriously, one should take into consideration that those results are\nbiased, because it is _Git User's_ Survey, and people usually choose\nSCM because they think it is best choice.\n\nNo answer (null answer) might mean that responder does not use and did\nnot use other SCMs to compare, or at least think that he/she does not\nhave sufficient basis for a comparison.\n\nNOTE: I have to check if number of non-empty responses is calculated\nproperly. I think it is, but I will make sure.\n\n\n36. What do you like about using GIT?\n\n TO DO\n 578 / 683 non-empty responses\n\nNo full analysis yet, and no count of answers, but it seems that the\nfollowing are encountered most often (or at least those I remember\nbest):\n * clean design\n * performance (speed)\n * size of repository\n * reliability, robustness, data integrity\n * flexibility, scriptable\n * command set, features (git-bisect)\n * history rewriting (amend, rebase, interactive rebase, reset)\n * history viewers, searching and browsing history\n * distributed nature, offline work, full history locally,\n \"local commits\"\n * easy and in-place branching\n * easy merging, includes renames detection\n * push/pull via ssh, pull via plain http -- no server needed\n * easy to install: few dependencies, portable, lightweight\n\n\n37. What would you most like to see improved about GIT?\n (features, bugs, plug-ins, documentation, ...)\n\n TO DO\n 512 / 683 non-empty responses\n\nSo many suggestions...\n\nFirst, some of suggestions are actually implemented already, for\nexample git development Changelog (present in the form of RelNotes),\nshallow clone support, submodules support. This means that new\nfeatures are not very well announced (which was one of comments here,\ntoo).\n\nSome of features and improvements mentioned:\n * generic behavior:\n - less chatter, clean distinction between success or fail\n (it it was about git-pull, it is addressed already)\n - better, more verbose error messages, with link to documentation\n - diagnostic output in case of problems;\n something like git-explain / git-state\n - more universal undo / continue\n - command line consistency, option consistency: getopt / gitopt\n * documentation:\n - unified, organized, searchable documentation\n - The Git Book (c.f. cvs/svn/hgbook);\n tutorial leading to advanced concepts like rebase, filter-branch,\n grafts, submodules, gitattributes\n - Git Cookbook / Git Recipies;\n best practices document, usage scenarios, workflows used, HOWTO\n - Git for Dummies (for people who haven't used SCM at all)\n - more hooks and hook examples\n - mid-level (script / plugin writer leve) docs\n - git1line docs a la sed1line.txt\n - \"git --help\" returning one page of short command summary,\n not manpage; \"git help --all --summary\" for all command with\n oneline description\n * other SCMs:\n - cogito migration guide / tutorial (!)\n - other SCM to git (concept, commands) cheat sheet\n - git-bzr (and other SCMs) two-way integration\n - git-svnserve: git functioning as Subversion server\n * git-svn:\n - automatic handling svn:externals using submodules and vice versa\n - svnmerge and svk merge markers tracking/marking of merges\n * more --interactive:\n - git add --interactive in git-gui: allow to divide hunks\n - git rebase -interactive: graphical interactive rebase in git-gui\n - ncurses-based remote editing\n * tools:\n - better Emacs support; Vim plugin; IDE plugins (Eclipse, KDevelop,\n IntelliJ IDEA,...)\n - MS Windows explorer shell integration; filemanagers integration\n (Nautilus, Konqueror)\n - side by side diffs in gitweb, a la KDiff3/Meld/ediff etc.\n * code:\n - port scripts to C (builtinification)\n - git library (libification)\n - gitbox: single static, pre-packaged binary\n * other:\n - bisect dunno / skip / next\n - partial checkout\n - light working copies; multiple working copies per repo\n - git-notes, to annotate commit messages\n - push over sftp\n - option to track empty directories\n - option to track permissions and metainfo: ACL, EA, forks\n - rebase and blame merge strategies\n - merge / rebase into dirty tree\n - resumable git-clone; faster and less CPU hungry git-clone\n - checkout/merge/diff/hunk header handlers in distribution for\n ChangeLog, XML, odf, jar and xul, po files\n\nThis is only [large] excerpt, not a list of all requested features.\nAs one can see tabularizing (dividing into categories) this data will\nbe not an easy task.\n\n\n38. If you want to see GIT more widely used, what do you think\n we could do to make this happen?\n\n TO DO\n 459 / 683 non-empty responses\n\nSo many suggestions...\nOne of the more striking (and funny):\n\n * Big fat posters world-wide with Catherine Zeta-Jones stating how\n happy she is since she switched to git ;-)\n\n She can actually write C-code.\n\n * Offer Git stickers to put on laptops and such\n\n\nChanges in GIT (since year ago, or since you started using it)\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n39. Did you participate in previous Git User's Survey?\n\n Answer | Count\n ------------------------------------------\n Yes | 51\n No | 583\n ------------------------------------------\n Base | 634 / 683\n Total (sum) | 634\n\n51 people did out of 634 who answered this question, out of 115 (?)\nwho did participate in the previous survey. Around half. Bit curious.\n\n\n40. What improvements you wanted got implemented?\n\n TO TABULARIZE\n 129 / 683 non-empty responses\n\n\n41. What improvements you wanted didn't get implemented?\n\n TO TABULARIZE\n 104 / 683 non-empty responses\n\n\n42. How do you compare current version with version from year ago?\n\nShould be: from year ago, or since you started using this.\nNo responses (303 / 683) migh be caused by the fact that one\ndo not use git for a year, so he/she doesn't know what git\nlooked like year ago.\n\n Answer | Count\n ------------------------------------------\n Better | 319\n No changes | 60\n Worse | 1\n ------------------------------------------\n Base | 380 / 683\n Total (sum) | 380\n\nMost think that git has improved. One user thinks that it is worse\nthan year ago; I wonder why...\n\n\n43. Which of the new features do you use?\n\nMultiple choice; the list does not include all new features.\nSome new features are not visible at first glance, and one\nuses them without conscious choice.\n\n Answer | Count\n ------------------------------------------\n git-gui | 103\n blame improvements | 74\n detached HEAD | 71\n stash | 68\n mergetool | 67\n interactive rebase | 66\n reflog | 54\n submodules | 52\n shallow clone | 31\n commit template | 24\n eol conversion | 22\n gitattributes | 21\n bundle | 17\n worktree | 17\n ------------------------------------------\n Base | 259 / 683\n Total (sum) | 687\n\nThis table is sorted by count.\n\nDetached HEAD support and stash command were long requested;\nno wonder they are popular.\n\n\nDocumentation\n~~~~~~~~~~~~~\n\n Usefulness of | Yes / Some / No | Base\n ----------------------------------------------\n GIT wiki | 191 / 69 / 198 | 458\n on-line help | 377 / 172 / 28 | 577\n distributed help | 425 / 154 / 22 | 601\n ----------------------------------------------\n individual responses : 683\n\n\n44. Do you use the GIT wiki?\n\n Answer | Count\n ------------------------------------------\n Yes | 316\n No | 300\n ------------------------------------------\n Base | 616 / 683\n Total (sum) | 616\n\n\n45. Do you find GIT wiki useful?\n\n Answer | Count\n ------------------------------------------\n Yes | 191\n No | 69\n Somewhat | 198\n ------------------------------------------\n Base | 458 / 683\n Total (sum) | 458\n\n\n46. Do you contribute to GIT wiki?\n\n Answer | Count\n ------------------------------------------\n Yes | 17\n No | 460\n Corrections and removing spam | 45\n\n Some contribution | 62\n ------------------------------------------\n Base | 522 / 683\n Total (sum) | 522\n\n\n47. Do you find GIT's on-line help (homepage, documentation) useful?\n\n Answer | Count\n ------------------------------------------\n Yes | 377\n No | 28\n Somewhat | 172\n ------------------------------------------\n Base | 577 / 683\n Total (sum) | 577\n\n\n48. Do you find help distributed with GIT useful\n (manpages, manual, tutorial, HOWTO, release notes)?\n\n Answer | Count\n ------------------------------------------\n Yes | 425\n No | 22\n Somewhat | 154\n ------------------------------------------\n Base | 601 / 683\n Total (sum) | 601\n\n\n49. Did/Do you contribute to GIT documentation?\n\n Answer | Count\n ------------------------------------------\n Yes | 43\n No | 551\n ------------------------------------------\n Base | 594 / 683\n Total (sum) | 594\n\n\n50. What could be improved on the GIT homepage?\n\n TO DO\n 171 / 683 non-empty responses\n\n\n51. What topics would you like to have on GIT wiki?\n\n TO DO\n 134 / 683 non-empty responses\n\n\n52. What could be improved in GIT documentation?\n\n TO DO\n 190 / 683 non-empty responses\n\n\nGetting help, staying in touch\n~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~\n\n Do you use | Yes / No | Base\n ----------------------------------------------\n GIT wiki | 316 / 300 | 616\n mailing list | 204 / 406 | 610\n IRC channel | 182 / 376 | 558\n ----------------------------------------------\n\n Usefulness of | Yes / Some / No | Base\n ----------------------------------------------\n GIT wiki | 191 / 69 / 198 | 458\n mailing list | 146 / 79 / 34 | 259\n IRC channel | 127 / 59 / 26 | 212\n ----------------------------------------------\n individual responses : 683\n\nAs one can see mailing list is the primary medium for git,\nand IRC secondary one.\n\n\n53. Have you tried to get GIT help from other people?\n\n Answer | Count\n ------------------------------------------\n Yes | 357\n No | 261\n ------------------------------------------\n Base | 618 / 683\n Total (sum) | 618\n\nIt might be that Git is not documented enough; it might be that\nversion control is not an easy task.\n\n\n54. If yes, did you get these problems resolved quickly\n and to your liking?\n\n Answer | Count\n ------------------------------------------\n Yes | 326\n No | 53\n ------------------------------------------\n Base | 379 / 683\n Total (sum) | 379\n\nThat is I think very good (around 86%) percentage.\n\n\n55. Would commerical (paid) support from a support vendor\n be of interest to you/your organization?\n\n Answer | Count\n ------------------------------------------\n Not Applicable | 203\n Yes | 69\n No | 322\n\n Yes + No | 391\n ------------------------------------------\n Base | 594 / 683\n Total (sum) | 594\n\nOnly 69 (12%) answers yes, 322 no, 203 not applicable (which meant\nto encompass people who do not use git at and for work).\n\n433 = 56 + 377 people participating in this survey use git for work,\nor for both work and unpaid projects (as compared to 391 who answered\nthis question not with N/A).\n\nNote that Git is GPLv2, and it will probably stay that forever, so you\nare _free_ to start a commercial support scheme for Git, but others\nare free not to choose it. This question is to get to know if there is\nsufficient demand for commercial Git support for it to be viable.\n\n\n56. Do you read the mailing list?\n\n Answer | Count\n ------------------------------------------\n Yes | 204\n No | 406\n ------------------------------------------\n Base | 610 / 683\n Total (sum) | 610\n\nThe fact that only one third of git users are reading mailing list\n(which is nevertheless quite large percentage) means that features\n_have_ to be documented somewhere besides mailing list archive and\nlogs (commit messages).\n\n\n57. If yes, do you find the mailing list useful?\n\n Answer | Count\n ------------------------------------------\n Yes | 146\n No | 34\n Somewhat | 79\n ------------------------------------------\n Base | 259 / 683\n Total (sum) | 259\n\n\n58. Do you find traffic levels on GIT mailing list OK?\n\n Answer | Count\n ------------------------------------------\n Yes | 193\n No | 50\n ------------------------------------------\n Base | 243 / 683\n Total (sum) | 243\n\n\n59. Do you use the IRC channel (#git on irc.freenode.net)?\n\n Answer | Count\n ------------------------------------------\n Yes | 182\n No | 376\n ------------------------------------------\n Base | 558 / 683\n Total (sum) | 558\n\n\n60. If yes, do you find IRC channel useful?\n\n Answer | Count\n ------------------------------------------\n Yes | 127\n No | 26\n Somewhat | 59\n ------------------------------------------\n Base | 212 / 683\n Total (sum) | 212\n\n\n61. Did you have problems getting GIT help on mailing list\n or on IRC channel? What were it? What could be improved?\n\n TO TABULARIZE\n 99 / 683 non-empty responses\n\n\nOpen forum\n~~~~~~~~~~\n\n62. What other comments or suggestions do you have, that are not\n covered by the questions above?\n\n TO DO\n 141 / 683 non-empty responses\n\n-- \nJakub Narebski\n"},{"id":"54809","messageId":"CDD8E66B-44E0-4258-A3CC-4E48BC58FD2F@lrde.epita.fr","threadId":"10137","inReplyTo":"8fe92b430710040212u332bb4e5yfa6c10b8236c6301@mail.gmail.com","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"Benoit SIGOURE","fromEmail":"tsuna@lrde.epita.fr","sentAt":"2007-10-04T12:04:10Z","receivedAt":"2007-10-04T12:04:10Z","isPatch":false,"sender":{"key":"tsunanet@gmail.com","avatar":"https://avatars.githubusercontent.com/u/128281?v=4"},"body":"Hi Jakub,\nnice work you've done there. Well done!\n\nOn Oct 4, 2007, at 11:12 AM, Jakub Narebski wrote:\n\n> 55. Would commerical (paid) support from a support vendor\n> be of interest to you/your organization?\n>\n> Answer | Count\n> ------------------------------------------\n> Not Applicable | 203\n> Yes | 69\n> No | 322\n>\n> Yes + No | 391\n> ------------------------------------------\n> Base | 594 / 683\n> Total (sum) | 594\n>\n> Only 69 (12%) answers yes, 322 no, 203 not applicable (which meant\n> to encompass people who do not use git at and for work).\n>\n> 433 = 56 + 377 people participating in this survey use git for work,\n> or for both work and unpaid projects (as compared to 391 who answered\n> this question not with N/A).\n>\n> Note that Git is GPLv2, and it will probably stay that forever, so you\n> are _free_ to start a commercial support scheme for Git, but others\n> are free not to choose it. This question is to get to know if there is\n> sufficient demand for commercial Git support for it to be viable.\n\nOnce again (AFAIR this was already raised during one of the previous \nsummary) what's the link between GPLv2 and commercial support? You \nseem to imply that because Git won't move to GPLv3, it's a good thing \nfor potential paid support, or something. I don't quite see how \nGPLvX comes into play with commercial support. I'm not a license \nexpert though.\n\n-- \nBenoit Sigoure aka Tsuna\nEPITA Research and Development Laboratory\n\n\n"},{"id":"54830","messageId":"8fe92b430710040759r5777264aj5ecb09e0252fef9a@mail.gmail.com","threadId":"10137","inReplyTo":"CDD8E66B-44E0-4258-A3CC-4E48BC58FD2F@lrde.epita.fr","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"Jakub Narebski","fromEmail":"jnareb@gmail.com","sentAt":"2007-10-04T14:59:26Z","receivedAt":"2007-10-04T14:59:26Z","isPatch":false,"sender":{"key":"jnareb@gmail.com","avatar":"https://avatars.githubusercontent.com/u/2706?v=4"},"body":"On 10/4/07, Benoit SIGOURE wrote:\n\n> Hi Jakub,\n> nice work you've done there. Well done!\n\nHalf done. But thanks.\n\nBy the way, I'd put final data from the survey as an attachement at\n http://git.or.cz/gitwiki/GitSurvey2007\nCurrently there is data I used to generate this summary, in CSV format.\nI have used horrible Perl script to tabularize free-form questions, removing\ncomments etc., and to generate tables in this summary.\n\n> On Oct 4, 2007, at 11:12 AM, Jakub Narebski wrote:\n>\n\n> > Note that Git is GPLv2, and it will probably stay that forever, so you\n> > are _free_ to start a commercial support scheme for Git, but others\n> > are free not to choose it. This question is to get to know if there is\n> > sufficient demand for commercial Git support for it to be viable.\n>\n> Once again (AFAIR this was already raised during one of the previous\n> summary) what's the link between GPLv2 and commercial support? You\n> seem to imply that because Git won't move to GPLv3, it's a good thing\n> for potential paid support, or something. I don't quite see how\n> GPLvX comes into play with commercial support. I'm not a license\n> expert though.\n\nThe only link between GPL and commercial support is that GPL does not\nprohibit commercial support (like noncommercial-free licenses for example),\nand that having commercial support doesn't mean that license would change\nto proprietary (it cannot).\n\n-- \nJakub Narebski\n"},{"id":"54846","messageId":"Pine.LNX.4.64.0710041712120.4174@racer.site","threadId":"10137","inReplyTo":"8fe92b430710040212u332bb4e5yfa6c10b8236c6301@mail.gmail.com","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2007-10-04T16:16:34Z","receivedAt":"2007-10-04T16:16:34Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Hi,\n\nOn Thu, 4 Oct 2007, Jakub Narebski wrote:\n\n> This is partial summary of Git User's Survey 2007,\n> ending at state from 28 September 2007.\n\nThanks.\n\n> 26. Which porcelains do you use?\n> \n> Multiple answers (one can use more than one porcelain).\n> \n> Answer (multiple choice) | Count\n> ------------------------------------------\n> core-git | 558\n> cogito (deprecated) | 56\n> \n> Patch management interface: : 57\n> ...........................................\n> StGIT | 41\n> Guilt (formerly gq) | 13\n> pg (deprecated) | 3\n> \n> own scripts | 95\n> other | 14\n> ------------------------------------------\n> Base | 593 / 683\n> Total (sum) | 780\n> \n> Those 14 \"other\" answers make me wish to have provided \"if other,\n> what it was?\" (sub)question; actually not only for this question.\n\ngit-gui, of course. I consider it porcelain, because it uses core-git as \nbackend.\n\nIn the same vein, I should consider gitk porcelain now, since it has \nrebase capabilities. I will not, and I am not very happy that this viewer \ngot a non-view-only capability, instead of git-gui, where that feature \nshould have belonged (as suggested by at least one answer to a later \nquestion in the survey -- not by me).\n\nCiao,\nDscho\n"},{"id":"54894","messageId":"20071005012726.GR2137@spearce.org","threadId":"10137","inReplyTo":"Pine.LNX.4.64.0710041712120.4174@racer.site","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2007-10-05T01:27:27Z","receivedAt":"2007-10-05T01:27:27Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Johannes Schindelin wrote:\n> On Thu, 4 Oct 2007, Jakub Narebski wrote:\n> > 26. Which porcelains do you use?\n> > \n> > Multiple answers (one can use more than one porcelain).\n> > \n> > Answer (multiple choice) | Count\n...\n> > other | 14\n...\n> > Those 14 \"other\" answers make me wish to have provided \"if other,\n> > what it was?\" (sub)question; actually not only for this question.\n> \n> git-gui, of course. I consider it porcelain, because it uses core-git as \n> backend.\n\nI tried to get git-gui into this list as a choice as I really do\nconsider it porcelain but Jakub thought it wasn't and wanted to\nhave a specific category for GUIs. Whatever. Its probably all\ngit-gui and qgit users that picked other here. Probably in about\nthe same ratio too, so 8 git-gui's and 6 qgit's.\n\nFor the most part git-gui tries to only ever invoke plumbing. I\nbreak that rule in only three places:\n\n - git-merge: I'm lazy and didn't have time yet to rewrite this\n properly using Tcl. I already do about half of the\n work anyway (e.g. merge base testing, fast-forward\n detection, message formatting).\n\n - git-fetch: Now that this is in C I'm going to call it plumbing\n even if nobody else does. Especially for say HTTP\n as git-fetch process does all of the HTTP requests\n directly. I won't reimplement it in Tcl, it would\n be slower and suck more. So git-gui won't be calling\n git-fetch-pack anytime soon.\n\n - git push: Same as the above reason for git-fetch.\n\nIMHO, porcelain is anything that only invokes plumbing. Seats\nhowever can sit above porcelain to make the position slightly\nmore comfortable. :-)\n\n> In the same vein, I should consider gitk porcelain now, since it has \n> rebase capabilities. I will not, and I am not very happy that this viewer \n> got a non-view-only capability, instead of git-gui, where that feature \n> should have belonged (as suggested by at least one answer to a later \n> question in the survey -- not by me).\n\nI agree. I actually never use the modification features of gitk.\nThey are so hidden that most users don't even know they are there.\nOf course that's just as hidden as the hunk selection in git-gui\nis so I shouldn't be complaining.\n\nI'm seriously looking at implementing those modification features\ninto git-gui and will probably start work on some of that during\nthe trip. I got plenty of time in a sealed tube at 30,000 feet\nto kill. More time than I got battery packs for the laptop. :-\\\n\nI think the first thing to implement is cherry-pick and revert as\nthose are easy and co-workers have been asking about them. That way\nyou can then rebase using cherry-pick and the Mark I wetware loop.\nThen we need a cool graph thing that you can drag the nodes around\non to create a visual `rebase -i`.\n\n-- \nShawn.\n"},{"id":"54896","messageId":"20071005014229.GS2137@spearce.org","threadId":"10137","inReplyTo":"8fe92b430710040759r5777264aj5ecb09e0252fef9a@mail.gmail.com","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"Shawn O. Pearce","fromEmail":"spearce@spearce.org","sentAt":"2007-10-05T01:42:29Z","receivedAt":"2007-10-05T01:42:29Z","isPatch":false,"sender":{"key":"spearce@spearce.org","avatar":"https://avatars.githubusercontent.com/u/34844?v=4"},"body":"Jakub Narebski wrote:\n> On 10/4/07, Benoit SIGOURE wrote:\n> > On Oct 4, 2007, at 11:12 AM, Jakub Narebski wrote:\n> > > Note that Git is GPLv2, and it will probably stay that forever, so you\n> > > are _free_ to start a commercial support scheme for Git, but others\n> > > are free not to choose it. This question is to get to know if there is\n> > > sufficient demand for commercial Git support for it to be viable.\n> >\n> > Once again (AFAIR this was already raised during one of the previous\n> > summary) what's the link between GPLv2 and commercial support? You\n> > seem to imply that because Git won't move to GPLv3, it's a good thing\n> > for potential paid support, or something. I don't quite see how\n> > GPLvX comes into play with commercial support. I'm not a license\n> > expert though.\n> \n> The only link between GPL and commercial support is that GPL does not\n> prohibit commercial support (like noncommercial-free licenses for example),\n> and that having commercial support doesn't mean that license would change\n> to proprietary (it cannot).\n\nRight. There has been some discussion in the past about forming\n\"The Git Company\".\n\nWhen this survey question was first posed there was some concern that\nGit might move to a commerical license of some sort and perhaps not\nbe GPLvX anymore. That concern is a non-issue; the copyrights for\nGit are held by over 300 people, many of whom are kernel hackers and\nstrong believers in the value of GPLv2. I'm not a kernel hacker,\nbut I also believe strongly in the value of the GPLv2 license.\nYou won't see me agreeing to move code I wrote to a non-GPL license\nanytime soon. Most (if not all!) of Git's authors feel the same way.\n\nThere's several reasons why forming \"The Git Company\" might help\nthe overall Git cause, and this question was a feeler to see if\nthe community was interested in acquiring support through it. Many\nother open source projects seem to get some benefit from having a\ncompany loosely affiliated with them, not the least of which are\nthings like:\n\n - some of the developers can focus more time on the project and\n still keep food on the table;\n\n - there are people focused on advertising/marketing the project\n and its benfits to potential end-users;\n\n - companies that feel warm-and-fuzzy by having a phone number they\n can call for help are more likely to want to use the project\n for critical services;\n\n - companies that want training or short-term consulting services\n know who they can contact for expertise.\n\nand the list goes on. The problem with said company is it costs\nmoney to keep the lights on and employees fed; money which obviously\ncannot be extorted from users through arcane licensing agreements.\n:-)\n\n-- \nShawn.\n"},{"id":"54899","messageId":"e1dab3980710041848i3aba0072lf166233f8d967e30@mail.gmail.com","threadId":"10137","inReplyTo":"20071005012726.GR2137@spearce.org","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"David Tweed","fromEmail":"david.tweed@gmail.com","sentAt":"2007-10-05T01:48:51Z","receivedAt":"2007-10-05T01:48:51Z","isPatch":false,"sender":{"key":"david.tweed@gmail.com","avatar":null},"body":"On 10/5/07, Shawn O. Pearce wrote:\n> Johannes Schindelin wrote:\n> > On Thu, 4 Oct 2007, Jakub Narebski wrote:\n> > > 26. Which porcelains do you use?\n> > >\n> > > Multiple answers (one can use more than one porcelain).\n> > >\n> > > Answer (multiple choice) | Count\n> ...\n> > > other | 14\n\nFWIW, I answered other there because I use my chronoversion\nhacky-scripts (in addition to doing some other fine-level work with at\nthe low level git command line). Obviously it's not a general purpose\nporcelain and it's \"deliberately not using\" normal development\npractice because I do lots of loosely related speculative research\nrather than working on a well-defined \"product\". I was initially\nconfused by what the index does (particularly since in those days it\nwas often referred to as a cache) and so wrote the whole routines\nusing low-level commands; if I was to do it again I'd have my script\njust call \"git add\" and then using the vanilla \"git commit\".\n\n-- \ncheers, dave tweed__________________________\ndavid.tweed@gmail.com\nRm 124, School of Systems Engineering, University of Reading.\n\"we had no idea that when we added templates we were adding a Turing-\ncomplete compile-time language.\" -- C++ standardisation committee\n"},{"id":"54912","messageId":"4705EE6A.6020006@op5.se","threadId":"10137","inReplyTo":"20071005014229.GS2137@spearce.org","subject":"Re: Git User's Survey 2007 unfinished summary (long)","fromName":"Andreas Ericsson","fromEmail":"ae@op5.se","sentAt":"2007-10-05T07:57:30Z","receivedAt":"2007-10-05T07:57:30Z","isPatch":false,"sender":{"key":"ae@op5.se","avatar":"https://gravatar.com/avatar/426e89595c75a8f5252dd0c989e5fabe5bcac616e68557427ad9aef6b0ca342a?d=mp&s=160"},"body":"Shawn O. Pearce wrote:\n> Jakub Narebski wrote:\n>> On 10/4/07, Benoit SIGOURE wrote:\n>>> On Oct 4, 2007, at 11:12 AM, Jakub Narebski wrote:\n>>>> Note that Git is GPLv2, and it will probably stay that forever, so you\n>>>> are _free_ to start a commercial support scheme for Git, but others\n>>>> are free not to choose it. This question is to get to know if there is\n>>>> sufficient demand for commercial Git support for it to be viable.\n>>> Once again (AFAIR this was already raised during one of the previous\n>>> summary) what's the link between GPLv2 and commercial support? You\n>>> seem to imply that because Git won't move to GPLv3, it's a good thing\n>>> for potential paid support, or something. I don't quite see how\n>>> GPLvX comes into play with commercial support. I'm not a license\n>>> expert though.\n>> The only link between GPL and commercial support is that GPL does not\n>> prohibit commercial support (like noncommercial-free licenses for example),\n>> and that having commercial support doesn't mean that license would change\n>> to proprietary (it cannot).\n> \n> Right. There has been some discussion in the past about forming\n> \"The Git Company\".\n> \n> When this survey question was first posed there was some concern that\n> Git might move to a commerical license of some sort and perhaps not\n> be GPLvX anymore. That concern is a non-issue; the copyrights for\n> Git are held by over 300 people, many of whom are kernel hackers and\n> strong believers in the value of GPLv2. I'm not a kernel hacker,\n> but I also believe strongly in the value of the GPLv2 license.\n> You won't see me agreeing to move code I wrote to a non-GPL license\n> anytime soon. Most (if not all!) of Git's authors feel the same way.\n> \n> There's several reasons why forming \"The Git Company\" might help\n> the overall Git cause, and this question was a feeler to see if\n> the community was interested in acquiring support through it. Many\n> other open source projects seem to get some benefit from having a\n> company loosely affiliated with them, not the least of which are\n> things like:\n> \n> - some of the developers can focus more time on the project and\n> still keep food on the table;\n> \n> - there are people focused on advertising/marketing the project\n> and its benfits to potential end-users;\n> \n> - companies that feel warm-and-fuzzy by having a phone number they\n> can call for help are more likely to want to use the project\n> for critical services;\n> \n> - companies that want training or short-term consulting services\n> know who they can contact for expertise.\n> \n> and the list goes on. The problem with said company is it costs\n> money to keep the lights on and employees fed; money which obviously\n> cannot be extorted from users through arcane licensing agreements.\n\nActually it can. I work for precisely such a company (although using\nNagios, cacti and syslog-ng as the base of our products rather than\ngit). The GPL doesn't state how one is allowed to charge money for\nthe products, but since larger networks with more users generate more\nsupport calls, we also use a license payment model, where larger\ncustomers pay more and smaller pay less.\n\nThe difference between proprietary software is that we have to trust\nur customers to *want* to pay the licenses, as it would be ridiculously\neasy for them to replace our versions of the programs with the pristine\noss kind, or even with our patches, as we aren't allowed to keep them\nprivate. However, doing so voids the support-agreement, as we don't take\nsupport for code we haven't audited and tested. In other words, we *must*\nprovide first-class support and coding aid to our customers, or they\nwon't want to pay anymore. In practice, it all works out rather well, and\neveryone gets something they want.\n\n* I get paid to work with something I like.\n* Customers get support, well-tested upgrades and nifty extensions.\n* Project maintainers get patches, money, hardware and appreciation.\n\nThat last bit is important though. It's not terribly expensive for us to\nbuy a couple of books or a laptop and send it as a christmas gift to some\nproject maintainer, but doing so shows appreciation and also buys us the\nattention of the developers for when we want our feature-patches accepted ;-)\n\n-- \nAndreas Ericsson andreas.ericsson@op5.se\nOP5 AB www.op5.se\nTel: +46 8-230225 Fax: +46 8-230231\n"}]}