Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

Can you think of any more charitable reason than this? I think it beggars belief that Homebrew isn’t interested in improving performance; it’s something we talk about constantly and is one of our largest and most consistent user asks.

(It’s not just midnight, or terminals, as mentioned, not that either of these is really a “just”. And it’s not just GitHub, as mentioned.)



To be clear I think "it's a lot of work and we don't want to deal with the complexity" is a totally valid reason. I have that reason for not doing a ton of things.

I just don't think "it shouldn't be done" is really the true reason. It's very tempting to say that though because it sounds like a stronger defense than "I don't want to". I think some people that do it don't even realise.


(In this subthread it is just github because we were discussing enabling parallelism for github only. The terminal proposal meant "keep it single thread for CI")


Okay. The problem is still present in the GH case, although to a lesser extent.

These kinds of complications on a "same experience for everyone" flow are also nontrivial: single-threading on CI is hard to explain to users who see very fast local installs and slow CI installs - the latter costs them extra money! - and complicates the story with mirrors, download groups that include non-GH origins, etc.

That's not to say it's impossible; as linked in another thread, it's something that's being actively worked on. But I think it should be understandable why "just parallelize it and you're wrong/stupid/lazy because you haven't already" is neither a contextualized or charitable response.


I agree that the brew devs are not stupid or lazy (as can be seen by the great popularity of the tool they developed among the complaining demographic).

I suspect they are wrong in this specific care based on the "every other package manager solved this" heuristic, but I have no doubt this is not a straightforward issue.

Personally, I dropped the case after offering to send a PR and getting back a "we won't accept this, it's not because we didn't think of it, from where we're standing it's a bad idea". They may or may not be wrong, but X am definitely wrong here, knowing almost nothing about it.




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: