[Numpy-discussion] Software Freedom Conservancy calls for move from Github
Hi, I just came across this: https://sfconservancy.org/GiveUpGitHub/ I guess this is something we should review and consider - although it would obviously have serious costs. Cheers, Matthew ___ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
[Numpy-discussion] Re: Software Freedom Conservancy calls for move from Github
On Mon, Jul 4, 2022 at 7:48 AM Matthew Brett wrote: > Hi, > > I just came across this: > > https://sfconservancy.org/GiveUpGitHub/ > > I guess this is something we should review and consider - although it > would obviously have serious costs. > > Cheers, > > I didn't see anything in the article that made me want to switch. I would need to see actual cases of abuse. I also recall that Linus used Bitkeeper because it was the best tool at the time and made that argument. The trouble arose when Andrew Tridgell reversed engineered the protocol, which Larry McVoy complained violated the terms of use. So on and so forth. I haven't noted problems along that line with GitHub. It doesn't bother me that it is proprietary as long as they are responsive and I do appreciate the work going on to make it better. Someone has to pay for that and we are getting a free ride. It might be nice if we could back up the issues and PRs so history wouldn't be lost if we did need to move at some point, but that would be a good idea on any site. Chuck ___ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
[Numpy-discussion] Re: Software Freedom Conservancy calls for move from Github
Thanks Matthew! I will say one thing, I agree that there are major costs, but the longer I work in this space the more I appreciate the benefits there might be to *not* being on GitHub. I recently (finally) read Nadia Eghbal's Working in Public, where she points out that *adding* friction to the process of users commandeering maintainers' attention might be a good thing — commons tend to be depleted when there is no cost to exploiting them, and open source communities are an attention commons where maintainers' attention is constantly being used unsustainably. That's been bouncing around in my head for a few months. This post just adds fuel to that idea. Juan. On Mon, 4 Jul 2022, at 11:46 PM, Matthew Brett wrote: > Hi, > > I just came across this: > > https://sfconservancy.org/GiveUpGitHub/ > > I guess this is something we should review and consider - although it > would obviously have serious costs. > > Cheers, > > Matthew > ___ > NumPy-Discussion mailing list -- numpy-discussion@python.org > To unsubscribe send an email to numpy-discussion-le...@python.org > https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ > Member address: j...@fastmail.com ___ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
[Numpy-discussion] Re: Software Freedom Conservancy calls for move from Github
Hi, On Mon, Jul 4, 2022 at 3:21 PM Charles R Harris wrote: > > > > On Mon, Jul 4, 2022 at 7:48 AM Matthew Brett wrote: >> >> Hi, >> >> I just came across this: >> >> https://sfconservancy.org/GiveUpGitHub/ >> >> I guess this is something we should review and consider - although it >> would obviously have serious costs. >> >> Cheers, >> > > I didn't see anything in the article that made me want to switch. I would > need to see actual cases of abuse. I also recall that Linus used Bitkeeper > because it was the best tool at the time and made that argument. The trouble > arose when Andrew Tridgell reversed engineered the protocol, which Larry > McVoy complained violated the terms of use. So on and so forth. I haven't > noted problems along that line with GitHub. It doesn't bother me that it is > proprietary as long as they are responsive and I do appreciate the work going > on to make it better. Someone has to pay for that and we are getting a free > ride. It might be nice if we could back up the issues and PRs so history > wouldn't be lost if we did need to move at some point, but that would be a > good idea on any site. Yes, I felt the same - that the arguments there were not completely compelling to me - yet - but I was wondering whether we should take the opportunity to take stock, and make sure we have prepared, in case we do have to move, or if there are a subset of developers who want to work elsewhere, collaborating with the main repository on Github. Cheers, Matthew ___ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
[Numpy-discussion] Re: Software Freedom Conservancy calls for move from Github
On Mon, Jul 4, 2022 at 4:23 PM Juan Nunez-Iglesias wrote: > Thanks Matthew! > > I will say one thing, I agree that there are major costs, but the longer I > work in this space the more I appreciate the benefits there might be to > *not* being on GitHub. I recently (finally) read Nadia Eghbal's Working in > Public, where she points out that *adding* friction to the process of users > commandeering maintainers' attention might be a good thing — commons tend > to be depleted when there is no cost to exploiting them, and open source > communities are an attention commons where maintainers' attention is > constantly being used unsustainably. > > That's been bouncing around in my head for a few months. This post just > adds fuel to that idea. > This is a great point, and it came to mind for me as well. The amount of noise on a repo as popular as numpy can be overwhelming at times. There's no easy solution though, because we'd want to add some friction to the low-value pings/issues/PRs, but without adding significantly more friction for the valuable ones. And that includes maintainers of one project contributing to other projects. Imagine if every other project is on a different hosting site, and you'd have to create and maintain credentials separately for each project in order to file a bug or open a PR. Note that GitLab doesn't even let you have full-featured code search if you don't log in - that'd be pretty bad. > > Juan. > > On Mon, 4 Jul 2022, at 11:46 PM, Matthew Brett wrote: > > Hi, > > > > I just came across this: > > > > https://sfconservancy.org/GiveUpGitHub/ I agree with Chuck's assessment. I would go a step further: the only valid argument is the second one (ethics/ICE) - that is important and perhaps worth discussing here, or in a future developers meeting. The rest is quite ignorant stuff that you typically hear from GPL zealots. Who cares that the GitHub UI is not open source? I don't, I just want it to work well. GitHub is a business, and in the end they need to make money somewhere. The model of providing free services to open source projects and making you pay if you want private repos with all the bells and whistles is perfectly valid and fine with me. We're getting a lot of value out of GitHub, from issue tracker and code hosting to large amounts of free and well-designed CI services. And if you've ever interacted with GitHub staff, you have probably found that they truly care about open source and want to address pain points maintainers are experiencing. > It might be nice if we could back up the issues and PRs so history wouldn't be lost if we did need to move at some point, but that would be a good idea on any site. This would be great. It's not hard to do as a one-off, but doing it on an ongoing basis and keeping all of the history is pretty tricky. Having an easy way to have a full mirror on another hosting site would be super useful. Ralf ___ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
[Numpy-discussion] video of Ryan Cooper's presentation for the NumPy Newcomers Hour
A video recording of the latest Newcomers Hour with Ryan C. Cooper has been posted on the NumPy YouTube channel: https://youtu.be/lHDEr2eCjAM. On Sun, Jun 26, 2022 at 2:53 PM Inessa Pawson wrote: > The next NumPy Newcomers Hour will be held this Thursday, June 30th at 4 > pm UTC. > > Ryan C. Cooper, an assistant professor-in-residence at the University of > Connecticut (Mansfield, Connecticut, USA), will share how he uses NumPy in > his Engineering classes, from individual student research to semester-long > courses. We will talk about lessons learned and key strategies to motivate > and engage new Python users. > > Join us via Zoom: https://us02web.zoom.us/j/87192457898 > -- Every good wish, Inessa *Inessa Pawson* Contributor Experience Lead | NumPy email: ine...@albuscode.org ___ NumPy-Discussion mailing list -- numpy-discussion@python.org To unsubscribe send an email to numpy-discussion-le...@python.org https://mail.python.org/mailman3/lists/numpy-discussion.python.org/ Member address: arch...@mail-archive.com
