GHC 9.12.5-rc4 is now available
MangoIV: Binary distributions, source distributions, and documentation are available at downloads.haskell.org and via GHCup. the link of download.haskell.org is wrong, it points to rc3
Posts from every feed QUEST has found across the web — blogs, news, podcasts, channels and more.
MangoIV: Binary distributions, source distributions, and documentation are available at downloads.haskell.org and via GHCup. the link of download.haskell.org is wrong, it points to rc3
This works well if the main cost of a new worktree is dependencies. But if you have a big project, then often compiling your own code is most of the cost. It would be super if cabal could also have something similar. Relocatable build tree · Issue #12137 · haskell/cabal · GitHub is a good step in that direction.
Nixpkgs’ Haskell support looks a bit like stack in that they support a fixed package set but if your package happens to build with that package set you’ll be able to download all your dependencies from their cache, no need to set up anything yourself. Your local package can just be built using cabal.
Yep. You do need a separate build cache service (Buck2 doesn’t have one built in) but there’s a lot we could do to automate parts of the workflow to make it more streamlined. We could distribute Buck2 and bazel-remote through ghcup; we could have cabal-buck2 start the build cache itself; we could have cabal-buck2 clone the buck2/ directory if it doesn’t exist. Then it might just be a ghcup and a cabal install cabal-buck2, and you’re good to go.
This is a fascinating discussion with so many differing views, some compatible, some less so. I was surprised at some ad hominems (as in “platforming horrible people”, for example), and cheered up by @chreekat’s recognition of the ongoing conflicts of interest being peaceful and civil. Most speakers here are insightful, kind, and often wise. To preface the actual comment, I’ve no stakes in the game because Haskell is one of several languages I love, I’m barely European, forget American, and I’m…
Please rename Haskell Foundation Find like-minded people and create/build/discuss/enjoy life together. Not every bubble is a bad bubble. (As a complete tangent, this is something very like the philosophy of Epicureanism, which I’m sharing because I came across it by chance some years ago. It’s an idea that helped me even if I don’t practice it!)
I greatly appreciate the response, thank you. I now feel like I need to go back over my reply with a fine toothed comb to make sure I’m not implying something more than I intended

For better or worse this is also not just a theoretical debate. Communities whose member I see myself as, who foster inclusion and diversity, including the necessity of strongly rejecting those seeking to destroy it are spilling over from people of diverse backgrounds who want to contribute and make that community thrive, not for its “technical merits” but for its people, for its values. It’s such an incredible joy to arrive at their events, filled with people who don’t have to fear to show who…
LaurentRDC: It is incompetence, on my part at the very least. I am ultimately responsible for the integrity of the process. There was no malice, no intent to try something to see if it would work. Unless you’re chronically online, it’s very easy to miss these parts of the process. I’ve certainly interacted with him on github and other fora, but never twitter or anywhere else where the shitposting echochambers were present, so it was news to me as well. In fact, I’m even pro having someone like…
Please rename Haskell Foundation the statement one or more people on the board were incompetent in this specific instance of discussion is the best possible interpretation of events It is incompetence, on my part at the very least. I am ultimately responsible for the integrity of the process. There was no malice, no intent to try something to see if it would work. When I saw the posts you surfaced in the other thread – which were plainly visible, not at all hard to find – I truly, genuinely,…
You know, as an American I despaired for a long time that our two big political parties no longer seemed to adequately champion my interests. As a nerd I had dumb ideas about getting into politics to fix things™ but that seemed pretty hopeless, too. I’m obviously not cut out to be a politician. Eventually — quite lately, I’m ashamed to admit — I realized there’s a path that’s actually more achievable and more likely to have results: form my own groups. Don’t hang out with people who bother me…
mpilgrem: can be no more representative of the community than the pool of people who respond to its open calls for applications to join the board This is obvious, yes, but it’s not the problem that we ought to solve. If we assume that our community is in fact diverse, then the problem is rather why is there no diversity on the board, to the degree where board members don’t realize that they are putting people into positions of power over a community whose members they hatefully attack. There…
Ah yes: email! The system administrator’s favorite system. Nothing ever goes wrong with email! I will be cleaning up a few things and tightening up our bounce/spam/forwarding machinery. This should have no noticeable effect, but see also the preceding paragraph. [Edit] Actually, it should have a few noticeable effects: less rate limiting from big peers like Google, and more emails getting to where they were meant to get.
simonpj: still less to imply that the motives of the board members are suspect or that they are incompetent. Why? It is quite clear that the statement one or more people on the board were incompetent in this specific instance of discussion is the best possible interpretation of events. I don’t think listing out less positive interpretations is helpful, even if there are strong visible signs that it this best interpretation is not the most likely one. To be incredibly clear: The problem I am…
I can certainly understand and sympathise with the lack of applicants but… What active steps can the HF take to get these people involved and what active steps can the HF take to make the community more inclusive in general? You’ve identified a problem, what is being done to address it? I don’t think there’s been any public discussion or communication from the foundation itself to this problem until now, has there? This is not a ‘build it and they’ll come’ thing. IMO active action is needed.
As above, this a personal contribution. The Foundation’s board can be no more representative of the community than the pool of people who respond to its open calls for applications to join the board. Rightly or wrongly, the Foundation does not publish the identity of people who apply but are not successful, and I will respect that. However, I think I can say this. I have experienced a few of those calls since I joined the board in April 2024. As Secretary, I handle some of the administation of…
One of the advantages cabal/stack have over solutions like buck2 or Bazel is that it’s very much pick-up-and-go. In contrast, above, the first step is to set up a build cache. Is there a way to make cabal-buck2 use a local directory like ~/.cache/cabal-buck2 as a default build cache? Or is the process of managing build caches fundamentally separate?
As usual, very well put @TeofilC : What active steps can the HF take to get these people involved and what active steps can the HF take to make the community more inclusive in general? Very good reminder that the burden to improve things here is on the HF itself. I try not to write this much in here or elsewhere. I have a tendency to be both verbose and inflammatory, which tends to be counter-productive. I will, in spite of it all, give it a try. Here we go. Whilst I can understand (and yet not…
Hi all, MLabs is partnering with the Haskell Foundation to offer ecosystem packages, as part of our ecosystem partnership program. Contracting MLabs via one of these packages results in direct contributions to both the Haskell ecosystem, and the Haskell Foundation. Thanks MLabs! haskell.foundation Haskell Offerings by MLabs Clients hiring MLabs through this partnership are also supporting the Haskell Foundation and the Haskell ecosystem.
simonpj: So the solution is not to undermine the Foundation, still less to imply that the motives of the board members are suspect or that they are incompetent. Rather it is to act as a “critical friend”. Affirm the Foundation and seek to improve it. Offer to help with specific tasks. Make constructive suggestions. Nominate yourself to become a board member. I agree. My feeling is that everyone in the HF was acting in good faith and intending to work towards the good of the Haskell community.…
Thanks for that reminder. It seems the way to do away with this confusion would be to rename it the /unit db/ and rename ghc-pkg to ghc-unit(s) . Units are arguably also not part of the Haskell language, but GHC needs to know about them to be able to link.
although I forget what that difference was. Packages are a Cabal concept. Packages have several components (libraries, sub-libraries, executables, test-suites, etc.). Components can be configured differently (think Cabal flags, platform-dependent conditionals, selected dependencies…). Each configured component is a unit identified by its unit-id (package name + library name + hash). Some units (the ones made from library components) can be installed in the now badly named package db (the thing…
Also, as a consequence, supposedly cabal and GHC have a slightly different idea of what a package is, although I forget what that difference was. On the other hand, I would much prefer, if the compiler didn’t have to know what packages are.
Something we often want to do is create a new worktree to work on a bug, PR, or small feature, especially if you’re forking off tasks for an AI to work on. Creating the worktree is easy, but to build and test the changes, you’re starting from a cold build. I want to run through quickly how to set things up so that you can spin up a new worktree and use cabal-buck2 to pull build artifacts from a cache, so the build in the fresh worktree is ready instantly. Get a build cache running. I use…
Perhaps I can add a few thoughts about the Haskell Foundation. The Haskell open-source ecosystem finds itself in an “uncanny valley”. One stable place (A) is the enthusiast project, where project community is relatively small and consists entirely of volunteers, people know each other, there are few governance issues, and no one truly relies on the software for mission-critical applications so they are tolerant of technical shotcomings. Another stable place (B) is where the project has a large…
That’s right, packages do have their separate specification in this document: https://www.haskell.org/cabal/proposal/pkg-spec.pdf which introduced them. That document is now only interesting for historical reasons, and has long since been superseded by the documentation in the cabal user guide. Packages not being a part of the Haskell language report might have been a mistake, since several language features arguably depend on the semantics of packages: the backpack module system, package…
GHCup metadata is not available yet, I will update you when it is available as usual in the pre-release channel of ghcup.
The GHC developers are pleased to announce the availability of the fourth release candidate for GHC 9.12.5, ghc-9.12.5-rc4. The fourth release candidate was necessary due to a large amount of critical issues that required backporting. This will be the last release candidate before release if no critical regressions are discovered in this RC – we’re so far past due that we don’t backport anymore bugfixes for this release. Binary distributions, source distributions, and documentation are…
I think the explanation is that the original language report did not define packages at all. There is now an extension in GHC, that enables disambiguation of module names by package name.
Package qualified imports is a GHC specific extension, not part of any language standard, IIRC.