2026/04/27

Newest at the top

2026-04-27 19:23:14 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 245 seconds)
2026-04-27 19:21:56 +0000target_i(~target_i@user/target-i/x-6023099) target_i
2026-04-27 19:19:01 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn
2026-04-27 19:17:14 +0000pavonia(~user@user/siracusa) siracusa
2026-04-27 19:17:11 +0000 <int-e> "enumerate", to me, means covering all possible cases (assignments of variables)
2026-04-27 19:16:29 +0000 <int-e> __monty__: that's not a contradiction; your search space (from splitting into cases) can explode long before you exhaust all possibilities (which may even be infinite)
2026-04-27 19:15:15 +0000 <__monty__> I may just not be good at using them, causing my observations of their working to skew towards lengthy enumerations : )
2026-04-27 19:14:14 +0000 <int-e> that's my objection
2026-04-27 19:14:06 +0000 <int-e> it's just the least important and interesting aspect of these solvers
2026-04-27 19:13:33 +0000 <__monty__> I didn't say they were exhaustive. But they do "enumerate" through relevant classes of values, no?
2026-04-27 19:13:29 +0000 <int-e> (but keeping learned clauses and information about how fruitful variables are for finding forced inferences (i.e., pruning the search tree early))
2026-04-27 19:12:36 +0000 <int-e> Even modern SAT solvers (which operate with boolean variables, so everything is discrete) are primarily trying to find forced inferences that allow better pruning of the search space, and a good order in which to try variables. They restart a lot (wiping the whole search tree))
2026-04-27 19:08:33 +0000 <int-e> __monty__: Not really. The point of these tools is generally to work symbolically. But there's usually a lot of case splitting, which adds an enumerative flavor, and can degernate into enumerating all possible cases.
2026-04-27 19:08:05 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 245 seconds)
2026-04-27 19:05:29 +0000r1bilski_(~r1bilski@user-31-175-22-58.play-internet.pl) (Ping timeout: 252 seconds)
2026-04-27 19:04:56 +0000 <__monty__> Enumerating is pretty much what model checkers do, no? And SMT solvers kinda.
2026-04-27 19:03:48 +0000layline-awaylayline_
2026-04-27 19:03:42 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn
2026-04-27 19:02:15 +0000layline_layline-away
2026-04-27 19:01:56 +0000layline-awaylayline_
2026-04-27 19:00:18 +0000redshuffle(~quassel@45.43.70.75)
2026-04-27 19:00:11 +0000redshuffle(~quassel@45.43.70.75) (Remote host closed the connection)
2026-04-27 18:57:07 +0000peterbecich(~Thunderbi@71.84.33.135) peterbecich
2026-04-27 18:52:12 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 255 seconds)
2026-04-27 18:50:03 +0000tzh(~tzh@c-76-115-131-146.hsd1.or.comcast.net) (Remote host closed the connection)
2026-04-27 18:49:23 +0000jmcantrell_jmcantrell
2026-04-27 18:48:40 +0000layline_layline-away
2026-04-27 18:47:46 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn
2026-04-27 18:46:30 +0000dtman34(~dtman34@c-73-242-68-179.hsd1.mn.comcast.net) (Ping timeout: 244 seconds)
2026-04-27 18:45:10 +0000sord937(~sord937@gateway/tor-sasl/sord937) (Quit: sord937)
2026-04-27 18:44:35 +0000jmcantrell_(~weechat@user/jmcantrell) jmcantrell
2026-04-27 18:41:17 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 252 seconds)
2026-04-27 18:36:21 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn
2026-04-27 18:32:51 +0000 <EvanR> that's just an intermediate form of proving things by cases
2026-04-27 18:29:10 +0000doyougnu(~user@38.175.72.111) (Client Quit)
2026-04-27 18:29:00 +0000doyougnu(~user@38.175.72.111) doyougnu
2026-04-27 18:28:36 +0000doyougnu(~user@38.175.72.111) (Client Quit)
2026-04-27 18:26:22 +0000doyougnu(~user@38.175.72.111) doyougnu
2026-04-27 18:25:35 +0000doyougnu(~user@38.175.72.111) (Remote host closed the connection)
2026-04-27 18:25:25 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) (Ping timeout: 244 seconds)
2026-04-27 18:23:06 +0000doyougnu(~user@38.175.72.111) doyougnu
2026-04-27 18:20:34 +0000merijn(~merijn@host-cl.cgnat-g.v4.dfn.nl) merijn
2026-04-27 18:19:50 +0000doyougnu-(~doyougnu@38.175.72.111) doyougnu-
2026-04-27 18:18:33 +0000doyougnu(~doyougnu@38.175.72.111) (Killed (NickServ (GHOST command used by doyougnu`!~user@38.175.72.111)))
2026-04-27 18:17:07 +0000 <davean> "Whats the smallest valid test type?" "Ok, now why isn't that a proof already?"
2026-04-27 18:14:32 +0000 <davean> You have to start by articulating why the version is the same but smaller ... and now you've listed the cases ...
2026-04-27 18:14:06 +0000 <davean> though the "making a small enough version" ... gets you very close to a proof ...
2026-04-27 18:13:39 +0000 <davean> And often via paramatricity one can make a small enough version too
2026-04-27 18:12:51 +0000 <monochrom> Yeah.
2026-04-27 18:12:39 +0000 <davean> Lots of things *are* small enough