{"thread":{"id":"56759","subject":"[Summit topic] Let's have chalk talks (part 1/?)","startedAt":"2021-10-22T07:51:25Z","lastAt":"2021-10-26T15:35:23Z","messageCount":6,"participants":["Johannes Schindelin","Philip Oakley","Ævar Arnfjörð Bjarmason","SZEDER Gábor"],"isPatch":false,"patchVersion":null,"patchTotal":null},"messages":[{"id":"439346","messageId":"nycvar.QRO.7.76.6.2110220950020.62@tvgsbejvaqbjf.bet","threadId":"56759","inReplyTo":null,"subject":"[Summit topic] Let's have chalk talks (part 1/?)","fromName":"Johannes Schindelin","fromEmail":"johannes.schindelin@gmx.de","sentAt":"2021-10-22T07:51:16Z","receivedAt":"2021-10-22T07:51:25Z","isPatch":false,"sender":{"key":"johannes.schindelin@gmx.de","avatar":"https://avatars.githubusercontent.com/u/127790?v=4"},"body":"Apparently I have to send this in chunks, to fool the Bayes filter of our\nbeloved mailing list into doing The Right Thing.\n\nThis session was led by Emily Shaffer. Supporting cast: Ævar Arnfjörð\nBjarmason, brian m. carlson, CB Bailey, and Junio Hamano.\n\nNotes:\n\n 1.  What’s a public chalk talk?\n\n     1.  At Google, once a week, the team meets up with no particular topic in\n         mind, or a couple topics, very informal\n\n     2.  One person’s turn each week to give an informal talk with a white\n         board (not using chalk)\n\n     3.  Topic should be technical and of interest to the presenter\n\n     4.  For example: how does protocol v2 work\n\n     5.  Collaborative, interactive user session\n\n     6.  Helps by learning about things\n\n     7.  Helps by honing skills like presentation skills\n\n     8.  A lot of (good) humility involved. For example, colleagues who have\n         been familiar with the project for a long time admitting they don’t\n         know, or have been wrong about things. Makes others feel more\n         comfortable with their perceived lack of knowledge\n\n     9.  Could be good for everybody on the Git mailing list, might foster less\n         combative communication on the list\n\n     10. Might be a way to attract new people by presenting “old timers” as\n         humble\n\n 2.  Does that appeal to anybody else?\n"},{"id":"439589","messageId":"b485b282-2b26-b9ab-da5d-ea59f5fcf463@iee.email","threadId":"56759","inReplyTo":"nycvar.QRO.7.76.6.2110220950020.62@tvgsbejvaqbjf.bet","subject":"Re: [Summit topic] Let's have chalk talks (part 1/?)","fromName":"Philip Oakley","fromEmail":"philipoakley@iee.email","sentAt":"2021-10-25T20:56:56Z","receivedAt":"2021-10-25T20:57:04Z","isPatch":false,"sender":{"key":"philipoakley@iee.email","avatar":"https://avatars.githubusercontent.com/u/914343?v=4"},"body":"On 22/10/2021 08:51, Johannes Schindelin wrote:\n> Apparently I have to send this in chunks, to fool the Bayes filter of our\n> beloved mailing list into doing The Right Thing.\n>\n> This session was led by Emily Shaffer. Supporting cast: Ævar Arnfjörð\n> Bjarmason, brian m. carlson, CB Bailey, and Junio Hamano.\n>\n> Notes:\n>\n>  1.  What’s a public chalk talk?\n>\n>      1.  At Google, once a week, the team meets up with no particular topic in\n>          mind, or a couple topics, very informal\n>\n>      2.  One person’s turn each week to give an informal talk with a white\n>          board (not using chalk)\n>\n>      3.  Topic should be technical and of interest to the presenter\n[...]\n\nA 'listener' perspective..\n\nAt the summit, the packfile's packing algorithm was mentioned as a\nhistoric 'chalk talk' that's now in the technical documentation.\n\nIn the meantime, while chatting to a colleague about the birthday\nparadox and its relation to Bloom filters, I realised I didn't\nunderstand what our Bloom filters were trying to do and what they recorded.\n\nI had a look at the code and documentation, but there isn't much there\nabout our Bloom filter implementation. A chalk talk could later be used\nin the same manner as the packfile discussion to show what the filters do?\n\nIf there is someone who'd like to talk through what the Bloom filters\nare doing in Git then I'd be all ears.\n\n--\nPhilip\n"},{"id":"439591","messageId":"211025.865ytk3imw.gmgdl@evledraar.gmail.com","threadId":"56759","inReplyTo":"b485b282-2b26-b9ab-da5d-ea59f5fcf463@iee.email","subject":"Re: [Summit topic] Let's have chalk talks (part 1/?)","fromName":"Ævar Arnfjörð Bjarmason","fromEmail":"avarab@gmail.com","sentAt":"2021-10-25T21:20:18Z","receivedAt":"2021-10-25T21:21:49Z","isPatch":false,"sender":{"key":"avarab@gmail.com","avatar":"https://avatars.githubusercontent.com/u/45301?v=4"},"body":"\nOn Mon, Oct 25 2021, Philip Oakley wrote:\n\n> On 22/10/2021 08:51, Johannes Schindelin wrote:\n>> Apparently I have to send this in chunks, to fool the Bayes filter of our\n>> beloved mailing list into doing The Right Thing.\n>>\n>> This session was led by Emily Shaffer. Supporting cast: Ævar Arnfjörð\n>> Bjarmason, brian m. carlson, CB Bailey, and Junio Hamano.\n>>\n>> Notes:\n>>\n>>  1.  What’s a public chalk talk?\n>>\n>>      1.  At Google, once a week, the team meets up with no particular topic in\n>>          mind, or a couple topics, very informal\n>>\n>>      2.  One person’s turn each week to give an informal talk with a white\n>>          board (not using chalk)\n>>\n>>      3.  Topic should be technical and of interest to the presenter\n> [...]\n>\n> A 'listener' perspective..\n>\n> At the summit, the packfile's packing algorithm was mentioned as a\n> historic 'chalk talk' that's now in the technical documentation.\n>\n> In the meantime, while chatting to a colleague about the birthday\n> paradox and its relation to Bloom filters, I realised I didn't\n> understand what our Bloom filters were trying to do and what they recorded.\n>\n> I had a look at the code and documentation, but there isn't much there\n> about our Bloom filter implementation. A chalk talk could later be used\n> in the same manner as the packfile discussion to show what the filters do?\n>\n> If there is someone who'd like to talk through what the Bloom filters\n> are doing in Git then I'd be all ears.\n\nThat sounds like a good idea for a topic, not to take away from any of\nthat discussion, but I believe the bloom filters we have are exclusively\nused for path filtering if they exist in the commit-graph.\n\nI.e. for:\n\n    git rev-list -- $path\n\nSee c525ce95b46 (commit-graph: check all leading directories in changed\npath Bloom filters, 2020-07-01) for one use-case and benchmark numbers.\n\n\n"},{"id":"439604","messageId":"20211025222648.GD2101@szeder.dev","threadId":"56759","inReplyTo":"b485b282-2b26-b9ab-da5d-ea59f5fcf463@iee.email","subject":"Re: [Summit topic] Let's have chalk talks (part 1/?)","fromName":"SZEDER Gábor","fromEmail":"szeder.dev@gmail.com","sentAt":"2021-10-25T22:26:48Z","receivedAt":"2021-10-25T22:26:54Z","isPatch":false,"sender":{"key":"szeder.dev@gmail.com","avatar":"https://avatars.githubusercontent.com/u/116324?v=4"},"body":"On Mon, Oct 25, 2021 at 09:56:56PM +0100, Philip Oakley wrote:\n> In the meantime, while chatting to a colleague about the birthday\n> paradox and its relation to Bloom filters, I realised I didn't\n> understand what our Bloom filters were trying to do and what they recorded.\n> \n> I had a look at the code and documentation, but there isn't much there\n> about our Bloom filter implementation. A chalk talk could later be used\n> in the same manner as the packfile discussion to show what the filters do?\n> \n> If there is someone who'd like to talk through what the Bloom filters\n> are doing in Git then I'd be all ears.\n\nThen perhaps you might find it's worth your while to read through\nthis:\n\n  https://public-inbox.org/git/20200529085038.26008-16-szeder.dev@gmail.com/\n\n"},{"id":"439633","messageId":"539c19db-6137-0bf9-927a-af7282cae2cc@iee.email","threadId":"56759","inReplyTo":"211025.865ytk3imw.gmgdl@evledraar.gmail.com","subject":"Re: [Summit topic] Let's have chalk talks (part 1/?)","fromName":"Philip Oakley","fromEmail":"philipoakley@iee.email","sentAt":"2021-10-26T09:45:54Z","receivedAt":"2021-10-26T09:46:03Z","isPatch":false,"sender":{"key":"philipoakley@iee.email","avatar":"https://avatars.githubusercontent.com/u/914343?v=4"},"body":"On 25/10/2021 22:20, Ævar Arnfjörð Bjarmason wrote:\n> On Mon, Oct 25 2021, Philip Oakley wrote:\n>\n>> On 22/10/2021 08:51, Johannes Schindelin wrote:\n>>> Apparently I have to send this in chunks, to fool the Bayes filter of our\n>>> beloved mailing list into doing The Right Thing.\n>>>\n>>> This session was led by Emily Shaffer. Supporting cast: Ævar Arnfjörð\n>>> Bjarmason, brian m. carlson, CB Bailey, and Junio Hamano.\n>>>\n>>> Notes:\n>>>\n>>>  1.  What’s a public chalk talk?\n>>>\n>>>      1.  At Google, once a week, the team meets up with no particular topic in\n>>>          mind, or a couple topics, very informal\n>>>\n>>>      2.  One person’s turn each week to give an informal talk with a white\n>>>          board (not using chalk)\n>>>\n>>>      3.  Topic should be technical and of interest to the presenter\n>> [...]\n>>\n>> A 'listener' perspective..\n>>\n>> At the summit, the packfile's packing algorithm was mentioned as a\n>> historic 'chalk talk' that's now in the technical documentation.\n>>\n>> In the meantime, while chatting to a colleague about the birthday\n>> paradox and its relation to Bloom filters, I realised I didn't\n>> understand what our Bloom filters were trying to do and what they recorded.\n>>\n>> I had a look at the code and documentation, but there isn't much there\n>> about our Bloom filter implementation. A chalk talk could later be used\n>> in the same manner as the packfile discussion to show what the filters do?\n>>\n>> If there is someone who'd like to talk through what the Bloom filters\n>> are doing in Git then I'd be all ears.\n> That sounds like a good idea for a topic, not to take away from any of\n> that discussion, but I believe the bloom filters we have are exclusively\n> used for path filtering if they exist in the commit-graph.\n\nI sort of thought that something like that was the case\n>\n> I.e. for:\n>\n>     git rev-list -- $path\n>\n> See c525ce95b46 (commit-graph: check all leading directories in changed\n> path Bloom filters, 2020-07-01) for one use-case and benchmark numbers.\n>\n>\nOften it's just managing to get these little linkages between these (to\nthe uninitiated) disparate ideas and system parts. E.g. the\n'exclusively' part, and then the use case.\n\n The \"commit-graph\" has a similar conceptual distance from the \"do they\nmean just the DAG (directed acyclic graph)?\" when it comes to the\ntechnical docs.\n\n Working out where to squeeze in such information can be tricky once the\ndocs have closed in around their existing layouts.\n-- \nPhilip\n"},{"id":"439652","messageId":"8df2fe20-1198-ad99-34ef-f9fa50a67fa2@iee.email","threadId":"56759","inReplyTo":"20211025222648.GD2101@szeder.dev","subject":"Re: [Summit topic] Let's have chalk talks (part 1/?)","fromName":"Philip Oakley","fromEmail":"philipoakley@iee.email","sentAt":"2021-10-26T15:35:20Z","receivedAt":"2021-10-26T15:35:23Z","isPatch":false,"sender":{"key":"philipoakley@iee.email","avatar":"https://avatars.githubusercontent.com/u/914343?v=4"},"body":"On 25/10/2021 23:26, SZEDER Gábor wrote:\n> On Mon, Oct 25, 2021 at 09:56:56PM +0100, Philip Oakley wrote:\n>> In the meantime, while chatting to a colleague about the birthday\n>> paradox and its relation to Bloom filters, I realised I didn't\n>> understand what our Bloom filters were trying to do and what they recorded.\n>>\n>> I had a look at the code and documentation, but there isn't much there\n>> about our Bloom filter implementation. A chalk talk could later be used\n>> in the same manner as the packfile discussion to show what the filters do?\n>>\n>> If there is someone who'd like to talk through what the Bloom filters\n>> are doing in Git then I'd be all ears.\n> Then perhaps you might find it's worth your while to read through\n> this:\n>\n>   https://public-inbox.org/git/20200529085038.26008-16-szeder.dev@gmail.com/\n>\nThanks for the link. It's quite a size of commit message ;-)  Are there\nany plans to extract the 'how it works' parts into a Bloom filter doc?\n\nPhilip\n"}]}