29:null 2a:["$","title",null,{"children":"Analyzing Cpython release patterns | Jonathan St-Onge | Observable"}] 2f:T918,In _Error and attack tolerance of collective problem solving: The DARPA Shredder Challenge_, Stefanovitch et al. (2014) showed that crowds are vulnerable to targeted attacks when solving collective problems. By examining the completion of the puzzle, they identify different patterns of activity before and after an attack. In this final project, we investigate how toxic episodes might have a similar (negative) impact on code release patterns. We interpret open source projects as sets of _nested_ puzzles that distributed teams must ''solve''. In this case, solving means that open source communities must, at one level, deliver modules that work, which then must be integrated into a feature-complete, functional library. Of course, a library may never be truly complete, or completely bug-free. Instead we focus on library releases, as these are discrete points in the library history where communities believe there are enough new features, or patches, that have been added. As such, we treat each major and minor release as its own challenge similar to the DARPA Shredder challenge, themselves comprised of small puzzles. As a first data analysis, we will assess if library releases are impacted by what we refer to as a global toxic event, that is, a toxic event with a notable engagement from the community at large. As in Stefanovich et al.'s article, the issue here is to assess whether such events occur, as well as identifying the potential immediate and long-term impacts of toxicity on the community. To identify this global event, we find inspiration in social media studies (Bagrow et al. 2011; De Domenico et al. 2020). In this case, we interpret open source projects as a large-scale social system, in which we try to identify social bursts in collective attention. Stefanovitch et al. looked at global patterns, but in our case we are also interested to zoom in on local events. To do so, we seek to assess the impact of toxicity on commit patterns at individual- and team-level. This is a more challenging research question, as commit time series are inherently _bursty_ (Karsai et al. 2012, 2017}. We embark on a journey to characterize what is a typical commit behavior from that perspective, and if burst dynamics is impacted by committers flagged as toxic from their social behaviors on mailing list.30:[{"id":"6acb48a1525681e98dff4ea39ce679648415cec10a85170a2f3d9ce6622480cf6cebea118ee55792813f77968120d36c2a6f44914415c2e6ff27fc0e8957970d","href":"https://static.observableusercontent.com/files/6acb48a1525681e98dff4ea39ce679648415cec10a85170a2f3d9ce6622480cf6cebea118ee55792813f77968120d36c2a6f44914415c2e6ff27fc0e8957970d","name":"1-overview-commits.png","type":"image/png","lastModified":1650383544564.977,"size":241983},{"id":"961f188d52d490166d841d088a111f067aba4993e0ad86e78d877fc2f069bf7492a68013f4e8e4dab8f8ec371b0086a56cbc403d61a6d4fce0c82deb17032e8d","href":"https://static.observableusercontent.com/files/961f188d52d490166d841d088a111f067aba4993e0ad86e78d877fc2f069bf7492a68013f4e8e4dab8f8ec371b0086a56cbc403d61a6d4fce0c82deb17032e8d","name":"2-bursty-episodes.png","type":"image/png","lastModified":1650384167696.021,"size":54728},{"id":"d48278a0b255deebd7b3096b524d68e549c5443d5cef679f60cea03fad1f9e8145ddb99c3babff7a1b218642abc33a30bb644cf151d43d8d6d4e6fda02942245","href":"https://static.observableusercontent.com/files/d48278a0b255deebd7b3096b524d68e549c5443d5cef679f60cea03fad1f9e8145ddb99c3babff7a1b218642abc33a30bb644cf151d43d8d6d4e6fda02942245","name":"11-completionD.png","type":"image/png","lastModified":1650384771459.035,"size":41206},{"id":"0ed37e070e1a2c60f6e207e2fbdb7a10b553d74744edfdba4cfd72ca2fd208df7419f542b45147834524d44aaa12d55af2883b95808d454100816f0fa04e7f4d","href":"https://static.observableusercontent.com/files/0ed37e070e1a2c60f6e207e2fbdb7a10b553d74744edfdba4cfd72ca2fd208df7419f542b45147834524d44aaa12d55af2883b95808d454100816f0fa04e7f4d","name":"11-completionC.png","type":"image/png","lastModified":1650384771407.761,"size":30529},{"id":"70ef075d44b750e713c96dd77fc1dd5057245da451f047e07314ae7e25aff595401b4e756ce32047a739e9c075ca813b2691905dd8f17f5b73bb5ecd284a29d9","href":"https://static.observableusercontent.com/files/70ef075d44b750e713c96dd77fc1dd5057245da451f047e07314ae7e25aff595401b4e756ce32047a739e9c075ca813b2691905dd8f17f5b73bb5ecd284a29d9","name":"11-completionA.png","type":"image/png","lastModified":1650384771411.203,"size":22427},{"id":"2973d451eaa85dea2e865b745c53d0518f8bea5de3b19e1ac40f97597d25e2e8b071ec6632edcf850254d30bac66a8f63830469875a674978ce53d7dcf95fd43","href":"https://static.observableusercontent.com/files/2973d451eaa85dea2e865b745c53d0518f8bea5de3b19e1ac40f97597d25e2e8b071ec6632edcf850254d30bac66a8f63830469875a674978ce53d7dcf95fd43","name":"11-completionB.png","type":"image/png","lastModified":1650384771395.065,"size":25974}] 2d:{"id":"8a1293b9ad120051","slug":"analyzing-cpython-burst","title":"Analyzing Cpython release patterns","thumbnail":"db7211cbc6e934f0dd2fa4bac3ddd76a1d117d5535276429944638fd0da5da24","access_level":"public","create_time":"$D2022-04-19T13:40:55.232Z","update_time":"$D2022-04-19T17:05:54.152Z","view_time":null,"first_public_time":"$D2024-03-04T18:01:01.430Z","has_unpublished_edits":null,"fork":false,"stars":1,"starred":false,"owner":{"type":"team","id":"4fabdbefe825d266","login":"jstonge","name":"Jonathan St-Onge","avatar_url":"https://avatars-next.observableusercontent.com/48507f1ab99162ceb5bc1fccad23b163ba525a2a9a76b045557676a38f2a7014","tier":"starter_2024"},"version":130,"publish_version":null,"role":null,"authored":false,"authors":[{"type":"individual","id":"7cb18c5b30f73a44","login":"jstonge","name":"Jonathan St-Onge","avatar_url":"https://avatars-next.observableusercontent.com/48507f1ab99162ceb5bc1fccad23b163ba525a2a9a76b045557676a38f2a7014","tier":null}],"members":[{"type":"individual","id":"7cb18c5b30f73a44","login":"jstonge","name":"Jonathan St-Onge","avatar_url":"https://avatars-next.observableusercontent.com/48507f1ab99162ceb5bc1fccad23b163ba525a2a9a76b045557676a38f2a7014","tier":null,"role":"editor"},{"type":"team","id":"4fabdbefe825d266","login":"jstonge","name":"Jonathan St-Onge","avatar_url":"https://avatars-next.observableusercontent.com/48507f1ab99162ceb5bc1fccad23b163ba525a2a9a76b045557676a38f2a7014","tier":"starter_2024","role":"viewer"}],"comments":[],"custom_thumbnail":false,"trashed":false,"lapsed":null,"subscribed":null,"forks":0,"collection":{"id":"9b7668b044605017","slug":"community-vibrancy","title":"Quantifying F/OSS community vibrancy","owner":{"id":"4fabdbefe825d266","login":"jstonge","name":"Jonathan St-Onge"}},"type":"notebook","body":{"title":"Analyzing Cpython release patterns","cells":[{"id":0,"value":"# Analyzing Cpython release patterns\n## Do rudeness influence burstiness in cpython library release patterns?","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":53,"value":"$2f","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":44,"value":"----","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":23,"value":"### [1. What is the relationship between open source practices and software releases](https://observablehq.com/@jstonge/release-patterns?collection=@jstonge/community-vibrancy)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":21,"value":"

","pinned":false,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":84,"value":"In this notebook, we provide an overview of the `Cpython` release history. We visualize the number of daily commits, highlighting episodes of rudeness across the library history. We also investigate how each release globally unfolds over time, looking at the relationship between the `absolute number of lines` and `time to completion`.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":63,"value":"### [2. Characterizing open source software bursty patterns for the `Cpython` library](https://observablehq.com/@jstonge/bursty-patterns?collection=@jstonge/community-vibrancy)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":69,"value":"

","pinned":false,"hidden":false,"mode":"html","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":75,"value":"There are stories of programmers who frantically created legendary software in no time. Like many other social and natural phenomena, this frenziness is studied as _bursty dynamics_. In this notebook, we examine different facets of burstiness, characterizing what commit behaviors look like for each contributor to the release. \n\nAs we will see, there are both bursts as well as more steady commit behaviors. It depends on the type of code being committed, the role of the developpers, and project maturity.","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"},{"id":113,"value":"### [Inspiration](https://observablehq.com/@jstonge/bursty-patterns?collection=@jstonge/community-vibrancy)","pinned":false,"hidden":false,"mode":"md","author":"$undefined","summary":"$undefined","output":"","database":"$undefined"}],"files":"$W30","stdlib":"1","titling":"auto"}} 2e:"$2d:body"