Whole-known-network
<p><span class="h-card" translate="no"><a href="https://mastodon.social/@whitequark" class="u-url mention">@<span>whitequark</span></a></span> for the users? who can also be the contributors</p><p>the easier it is to contribute, the less the original dev burns out, right?</p>
<p><span class="h-card" translate="no"><a href="https://astrodon.social/@zrb" class="u-url mention">@<span>zrb</span></a></span> long-term health _for whom_?</p>
<p><span class="h-card" translate="no"><a href="https://mastodon.social/@whitequark" class="u-url mention">@<span>whitequark</span></a></span> <span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> I’m not really concerned with popularity either, but the long term health of a project hinges on the fluidity of its contributions IMO</p><p>of course every project is different</p>
<p><span class="h-card" translate="no"><a href="https://types.pl/@jj" class="u-url mention">@<span>jj</span></a></span> <span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> plus i suppose that is also a collective vote against gcc's nightmare policies</p>
<p><span class="h-card" translate="no"><a href="https://types.pl/@jj" class="u-url mention">@<span>jj</span></a></span> <span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> there are caveats, mainly that<br />a) LLVM is an incredibly complex piece of software of limited general interest, so only people already well versed into compilers get past that first reviewed commit hurdle, and<br />b) you are not free from "post-commit review" and you are quickly aware of this fact, so not many are inclined to try and push low-effort chnages in first place</p><p>i don't think this will works universally, but it still says a lot that this works at LLVM's scale</p>
<p><span class="h-card" translate="no"><a href="https://astrodon.social/@zrb" class="u-url mention">@<span>zrb</span></a></span> <span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> eh, i think that goes in the other direction?</p><p>even though i've never regretted doing so, there are absolutely reasons why i wouldn't do that anyway; a really simple one is "i'm not interested in collaborating on this project"</p><p>you presuppose that projects exist to become popular. meanwhile i wrote a bunch of popular stuff and i think popularity is suuuuper overrated</p><p>someone decided to make something their problem and not mine? fuck yeah!</p>
<p><span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> <span class="h-card" translate="no"><a href="https://mastodon.social/@whitequark" class="u-url mention">@<span>whitequark</span></a></span> projects that jealously guard their contributor list are projects that get aggressively forked</p><p>or rewritten entirely and die into obscurity, depending on how obstinate the core developers are</p><p>**I'm looking at you, <a href="https://astrodon.social/tags/flake8" class="mention hashtag" rel="tag">#<span>flake8</span></a>. You had no one to blame but yourselves for <a href="https://astrodon.social/tags/ruff" class="mention hashtag" rel="tag">#<span>ruff</span></a> taking your spot.**</p>
<p><span class="h-card" translate="no"><a href="https://chaos.social/@buzzyrobin" class="u-url mention">@<span>buzzyrobin</span></a></span> <span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> hi5</p>
<p><span class="h-card" translate="no"><a href="https://chaos.social/@jacqueline" class="u-url mention">@<span>jacqueline</span></a></span> i've never regretted giving commit access to someone who i felt has a clue in my entire life, interestingly. the llvm policy of "you get commit access after you get one commit accepted" has made an impression upon me</p><p>(only a single person has ever lost their llvm commit bit adversarially and nobody talks about that)</p>