Debian discusses Discourse
Debian discusses Discourse
Posted Apr 21, 2020 12:42 UTC (Tue) by pizza (subscriber, #46)In reply to: Debian discusses Discourse by jezuch
Parent article: Debian discusses Discourse
There's a pretty substantial gap between "configuring [the] stupid email client just the way the reviewer likes it" and "semi-effectively operating an email client"
The objection folks seem to have is that email is used _at all_, because *waah, it's too much effort*. Meanwhile they will hand out their email address or phone number to every app or website they encounter, and click on the link that comes back to "confirm their account". Congratulations, that's the total effort required to subscribe to a mailing list, and unlike the "messages from our selected partners" the apps or websites send in buckets, the mailing list will have clear subject lines -- and far more often than not, be considerably less voliminous.
But sure, let's scrap email for github-style PRs. And guess what? All three of the F/OSS projects I actively work on have github and/or gitlab mirrors. One has seen a total of 4 PRs opened, and 3 of those ended up abandoned because the submitter didn't respond. The fourth was eventually resubmitted via the "proper" channels, but still didn't make it in because the submitter didn't follow through with requested changes. Meanwhile, the other two projects have yet to see a single PR opened. Indeed, one project (for which I am the sole developer) has yet to see anyone click on the "fork" button, and, over the course of _eight years_, has received only one patch of any substance and maybe three or four trivial few-liners.
The overwhelming reason project X doesn't get drive-by contributions is that there isn't anyone (beyond the 1-4 people already involved) that gives enough of a f*ck to even look at the code, much less make changes to it. If local changes do get made, IME the biggest impediment to contributing upstream are actually corporate policies and processes that default to "don't share anything, ever."
As evidenced by the steady trickle of folks contacting me over email with questions, highly-technically-challennged support requests, or the rare "offering to test", "semi-effectively operating an email client" is a pretty low barrier indeed, I find it pretty hilarious that clueless end-users can manage this just fine but this mythical pool of hyper-talented developers that would be all over my code can't.
Indeed, it seems that the best way to generate contributions is to make your software crappier, because folks simply don't notice or care about the stuff that JustWorks(tm).
Posted Apr 21, 2020 15:49 UTC (Tue)
by mathstuf (subscriber, #69389)
[Link] (2 responses)
Plus, for any project that only gets a handful of external contributions over a long stretch, why would I want to set up a mailing list for it? The hosting service is largely irrelevant there (maybe CI solutions differentiate, but CI for such things is also usually the empty set), but if it's at least on *some* forge, those drive-bys are much more possible.
And any contribution mechanism will have abandoned patches. I don't know that one can derive rationale behind them without doing actual studies, especially with such a low N. As a contributor, I was definitely put off for a while when I had sent a patchset and it got, seemingly, completely ignored. Next time I go to rebase it (when I got a chance). Hey, it's been merged (with some minor fixups). A notice to the list would have been nice (not only for me, but else wanting those patches). Forges at least let me know when that happened.
And I'm not saying forges are always better. Lists work…ok for the kernel (as an occasional contributor), but feedback and better status tracking would definitely be appreciated.
Posted Apr 21, 2020 17:21 UTC (Tue)
by Cyberax (✭ supporter ✭, #52523)
[Link] (1 responses)
Those young whippersnappers...
Posted Apr 21, 2020 17:27 UTC (Tue)
by mathstuf (subscriber, #69389)
[Link]
Posted Apr 22, 2020 5:45 UTC (Wed)
by jezuch (subscriber, #52988)
[Link]
Autism works in mysterious ways...
Anyway, I was responding to the argument that since the project's subject matter is complex then it's fine if the submission process is also complex - and especially to ascribing an ideology to that. It's way simpler to be honest and just call it a hazing ritual :) It's fine to have reasons you work this way (it's better to try to improve that) but ideologies are bad.
Debian discusses Discourse
Debian discusses Discourse
It's actually quite funny. If you host your code on Github under an Open Source license then you can get free CI from Travis or Github Actions.
Debian discusses Discourse
Debian discusses Discourse