2026/03/11

Newest at the top

2026-03-11 17:49:30 +0100CloneOfNone_(~CloneOfNo@user/CloneOfNone) (Ping timeout: 268 seconds)
2026-03-11 17:48:13 +0100CloneOfNone(~CloneOfNo@user/CloneOfNone) CloneOfNone
2026-03-11 17:42:31 +0100 <tomsmeding> ircbrowse is not a fast server and you're DOSing the service for others; when you crawl a site, be courteous and put delays between your requests
2026-03-11 17:40:46 +0100 <tomsmeding> can the guy in Bulgaria (?) with IP 195.178.110.33 stop spamming ircbrowse
2026-03-11 17:40:42 +0100humasect(~humasect@184.151.37.182) (Read error: Connection reset by peer)
2026-03-11 17:28:31 +0100PHO`(~pho@akari.cielonegro.org) PHO`
2026-03-11 17:28:24 +0100troydm(~troydm@user/troydm) (Quit: What is Hope? That all of your wishes and all of your dreams come true? To turn back time because things were not supposed to happen like that (C) Rau Le Creuset)
2026-03-11 17:23:36 +0100PHO`(~pho@akari.cielonegro.org) (Ping timeout: 268 seconds)
2026-03-11 17:22:22 +0100humasect(~humasect@184.151.37.182) humasect
2026-03-11 17:19:34 +0100prdak(~Thunderbi@user/prdak) prdak
2026-03-11 17:17:24 +0100prdak(~Thunderbi@user/prdak) (Ping timeout: 245 seconds)
2026-03-11 17:16:35 +0100prdak1(~Thunderbi@user/prdak) (Read error: Connection reset by peer)
2026-03-11 17:15:52 +0100prdak1(~Thunderbi@user/prdak) prdak
2026-03-11 17:10:33 +0100CiaoSen(~Jura@2a02:8071:64e1:da0:5a47:caff:fe78:33db) (Ping timeout: 246 seconds)
2026-03-11 17:09:06 +0100dolio(~dolio@130.44.140.168) dolio
2026-03-11 17:05:29 +0100madresch(~Thunderbi@user/madresch) madresch
2026-03-11 17:03:37 +0100dolio(~dolio@130.44.140.168) (Quit: ZNC 1.10.1 - https://znc.in)
2026-03-11 17:02:26 +0100tromp(~textual@2001:1c00:3487:1b00:2807:b44c:c102:bda9)
2026-03-11 17:01:59 +0100Enrico63(~Enrico63@host-82-61-84-117.retail.telecomitalia.it) Enrico63
2026-03-11 17:00:50 +0100Alex_delenda_est(~al_test@5.139.232.240)
2026-03-11 16:58:42 +0100prdak1prdak
2026-03-11 16:56:21 +0100prdak1(~Thunderbi@user/prdak) prdak
2026-03-11 16:56:15 +0100prdak(~Thunderbi@user/prdak) (Read error: Connection reset by peer)
2026-03-11 16:48:26 +0100skinkitten(~skinkitte@user/skinkitten) (Quit: Client closed)
2026-03-11 16:38:13 +0100 <tomsmeding> there being multiple levels of representation accuracy, and probably also multiple distinct designs at the same level
2026-03-11 16:37:46 +0100 <tomsmeding> but in that case, "has haskell a denotational semantics" should just be answered with "for what purpose"
2026-03-11 16:36:56 +0100 <tomsmeding> right
2026-03-11 16:36:52 +0100 <tomsmeding> while in a language specification, you definitely want to know it if the language deviates from call-by-value
2026-03-11 16:36:52 +0100 <dminuoso> Nothing stops you from imposing an order of some kind to your denotation.
2026-03-11 16:36:39 +0100 <dminuoso> Well, that depends on you I think.
2026-03-11 16:36:29 +0100 <tomsmeding> AFAIK denotational semantics also tends not to say anything about operational concerns such as order of evaluation
2026-03-11 16:36:20 +0100prdak1prdak
2026-03-11 16:36:19 +0100prdak(~Thunderbi@user/prdak) (Ping timeout: 264 seconds)
2026-03-11 16:35:42 +0100 <dminuoso> It seems rather like a useful vehicle to do some research, rather than defining a language.
2026-03-11 16:35:39 +0100 <tomsmeding> s/trs$/ts/
2026-03-11 16:34:01 +0100 <tomsmeding> in terms of "relatively" simple categorical constructrs
2026-03-11 16:33:41 +0100 <tomsmeding> like, one that actually models interesting structure of the language
2026-03-11 16:33:20 +0100 <tomsmeding> perhaps "useful denotational semantics"
2026-03-11 16:33:14 +0100 <tomsmeding> isn't it possible to attach denotational semantics to anything?
2026-03-11 16:32:52 +0100 <dminuoso> I think a more accurate phrasing would be "It is possible to attach denotational semantics to a subset of Haskell", rather than saying that we *have* them (which might suggest that the language was signed ontop of it)
2026-03-11 16:32:32 +0100prdak1(~Thunderbi@user/prdak) prdak
2026-03-11 16:31:45 +0100 <tomsmeding> I was somehow only thinking of LambdaCase etc., which are not
2026-03-11 16:31:30 +0100 <tomsmeding> yes you're right, for denotational semantics all that's important
2026-03-11 16:31:00 +0100 <tomsmeding> ... I guess that types are actually relevant, even in operational semantics, depending on how you model type class resolution
2026-03-11 16:30:15 +0100 <dminuoso> tomsmeding: What about things like all the type wizardry (TypeInType, TyFams, etc)?
2026-03-11 16:29:29 +0100 <tomsmeding> in operational semantics you can exclude GADTs because you don't need types at runtime, but in denotational semantics the point is to retain types
2026-03-11 16:28:49 +0100 <tomsmeding> hm, fair point
2026-03-11 16:28:20 +0100 <dminuoso> Dunno, I can think of a few extensions that have meaning like GADTs
2026-03-11 16:27:40 +0100 <tomsmeding> the rest is going to be either syntactic sugar or complex, ad-hoc stuff that doesn't enlighten anyone about anything
2026-03-11 16:27:02 +0100 <tomsmeding> I suspect that all that people actually care about is a semantics of Haskell98 without the FFI