Message-ID: <69f540a7@news.ausics.net> From: not@telling.you.invalid (Computer Nerd Kev) Subject: Re: AI Is Killing Some Legacy Hardware Support Newsgroups: comp.os.linux.misc References: <10sbn6f$2kkkk$8@dont-email.me> <69eb2a5f@news.ausics.net> <10skbd3$1cssv$3@dont-email.me> <10skfk3$93hh$1@news1.tnib.de> <10skglm$1ej2r$1@dont-email.me> <69ee94fe@news.ausics.net> <10snd8b$29c0d$2@dont-email.me> <20260427121054.00003218@gmail.com> <69efe7b1@news.ausics.net> <10sphtg$ka69$1@news1.tnib.de> <69f13be9@news.ausics.net> <10ss6bf$qcto$1@news1.tnib.de> <10ssmgu$3quob$1@dont-email.me> <69f4705e$0$11451$426a74cc@news.free.fr> User-Agent: tin/2.6.5-20251224 ("Glenury") (Linux/2.4.31 (i586)) NNTP-Posting-Host: news.ausics.net Date: 2 May 2026 10:09:11 +1000 Organization: Ausics - https://newsgroups.ausics.net Lines: 64 X-Complaints: abuse@ausics.net Path: csiph.com!news.bbs.nz!news.ausics.net!not-for-mail Xref: csiph.com comp.os.linux.misc:86075 Stephane CARPENTIER wrote: > Le 29-04-2026, Rich a ecrit : >> >> I suspect this may reveal how many (most?) kernel maintainers are now >> corporate employees rather than volunteers, > > Not so long ago I read that at least 80% of kernel contributors were > paid by societies. I don't know neither how accurate that is nor where > the number were coming from. LWN regularly post statistics that include tracking the employers of contributors. In their most recent analysis, for kernel v6.19, they attribute employer "(none)" to 3.9% of "changesets" and 3.7% of "lines changed". https://lwn.net/Articles/1057302/ I've wondered before how they get the employer data, and reading comments this time I found out: "My second kernel patch was in here! Posted Feb 10, 2026 3:52 UTC (Tue) by vasi (subscriber, #83946) Proud to contribute to employer: Unknown My second kernel patch was in here! Posted Feb 10, 2026 6:29 UTC (Tue) by hrw (subscriber, #44826) Sooner or later you will get an email asking how to count you. Company or individual. So, check your inbox and spam." https://lwn.net/Articles/1057837/ So when a patch gets merged by a new contributor I take it LWN (or whoever provides their data) sends an email to their address asking who they work for. Those who don't respond are counted as "(unknown)", currently 9-10% of contributors. Therefore the number of contributors employed by companies is over 80%, and probably over 90% assuming non-responders are equally distributed. The Linux Foundation also has their own "Organizations leaderboard". I'm finding their Javascripty webpage quite infuriating and it stops showing percentages when you click on "All organizations", but their figures seem different. Though they still appear to count commercial code contributions well above 80% (I count The Linux Foundation as commercial, since I believe most of their funding is from companies). https://insights.linuxfoundation.org/project/korg/contributors Some of those stats look pretty dubious to me. The Linux Foundation is at the top of "authors of commits" but they're not in the top 20 in the LWN lists. LWN.net is listed there themselves as the #36 organisation for "authors of commits" over the last year. A Linux news website contributing more kernel patches than Amazon or Fedora Linux? Surely that's wrong. I'm not sure if the Linux Foundation just has web designers compiling these stats with lazy methods, or they've deliberately skewed the calculations to make themselves rank higher. -- __ __ #_ < |\| |< _#