TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Developing GHC for a Living

167 pointsby jacobwgover 5 years ago

6 comments

_bxg1over 5 years ago
&gt; I’m an idealist, and I believe that a perfect programming language with perfect tooling (or, at least, Pareto-optimal language and tooling) can exist. This is in contrast to people who just see languages as tools to get the job done. For them, a language is probably as good as the libraries available for their tasks.<p>I used to think that way. But it only took me half a decade of professional development (and reading&#x2F;discussing in circles like HN) to see that it&#x27;s a fallacy. A programming language cannot be perfect because it&#x27;s made for humans, who are not perfect (and runs on computers, which arguably are not perfect).<p>Accepting the fact that there&#x27;s no singular, perfect way to express a given idea has been freeing, actually. I used to constantly chase the dragon, getting hung up on refactoring and refactoring and refactoring, trying to attain that perfect description. I still get caught up in it sometimes. A little bit of that spirit makes for better code. But you have to be able to pull yourself away.
评论 #22195204 未加载
评论 #22194611 未加载
评论 #22193656 未加载
评论 #22194715 未加载
评论 #22196375 未加载
评论 #22194796 未加载
评论 #22194574 未加载
评论 #22197456 未加载
评论 #22199734 未加载
hopiaover 5 years ago
In terms of GHC&#x27;s evolution, this looks really promising to me: <a href="http:&#x2F;&#x2F;www.well-typed.com&#x2F;blog&#x2F;2019&#x2F;10&#x2F;nonmoving-gc-merge&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.well-typed.com&#x2F;blog&#x2F;2019&#x2F;10&#x2F;nonmoving-gc-merge&#x2F;</a>
评论 #22193449 未加载
kossaeover 5 years ago
For reference, as I wasn&#x27;t aware myself, GHC stands for Glasgow Haskell Compiler.
评论 #22195545 未加载
评论 #22196523 未加载
yingw787over 5 years ago
I would say I have a beginners understanding of Haskell (proof: <a href="https:&#x2F;&#x2F;bytes.yingw787.com&#x2F;posts&#x2F;2020&#x2F;01&#x2F;30&#x2F;a_review_of_haskell&#x2F;" rel="nofollow">https:&#x2F;&#x2F;bytes.yingw787.com&#x2F;posts&#x2F;2020&#x2F;01&#x2F;30&#x2F;a_review_of_hask...</a>) and IMHE Haskell is an academic research language for good reason.<p>For most software, the SDLC is a stream, not tightly scoped like a library or utility. I haven’t seen migration or upgrade plans, docs aren’t really there, links rot extensively, and the strangeness budget is a blank check. It’s not for most developers and it’s not because most developers are “lazy” or “dumb”. I would have loved to use Haskell for my personal projects, but it didn’t satisfy my requirements and I’m sticking to Python.<p>For starters, in Haskell Stack, there’s no uninstall option. You use shell to rm a package. <a href="https:&#x2F;&#x2F;github.com&#x2F;commercialhaskell&#x2F;stack&#x2F;issues&#x2F;361" rel="nofollow">https:&#x2F;&#x2F;github.com&#x2F;commercialhaskell&#x2F;stack&#x2F;issues&#x2F;361</a><p>That’s nope territory for most developers.
评论 #22197382 未加载
goofballlogicover 5 years ago
What happened to Idris? As someone who knows absolutely nothing about Haskell, I thought Idris was essentially Haskell+dependent types?
评论 #22195883 未加载
Brave-Steakover 5 years ago
How hard is it to start working on a compiler like this?