| 2024-12-29 00:05:02 +0100 | cowboy8625 | (~cowboy@2605-4A80-7405-640-B51A-FA7D-9084-E360-dynamic.midco.net) |
| 2024-12-29 00:09:06 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2024-12-29 00:10:10 +0100 | __monty__ | (~toonn@user/toonn) (Quit: leaving) |
| 2024-12-29 00:11:07 +0100 | Xe | (~Xe@perl/impostor/xe) Xe |
| 2024-12-29 00:12:54 +0100 | sawilagar | (~sawilagar@user/sawilagar) (Ping timeout: 260 seconds) |
| 2024-12-29 00:13:40 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 244 seconds) |
| 2024-12-29 00:23:44 +0100 | <haskellbridge> | <thirdofmay18081814goya> hm, anyone has some insight about why the indexed monad's "bind" has that type? |
| 2024-12-29 00:24:35 +0100 | <haskellbridge> | <thirdofmay18081814goya> "(a -> m j k b) -> m i j a -> m i k b", the two last types makes sense, but the first type "(a -> m j k b)" I'm not too sure about |
| 2024-12-29 00:25:44 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2024-12-29 00:26:13 +0100 | housemate | (~housemate@pa49-185-30-217.pa.vic.optusnet.com.au) (Ping timeout: 248 seconds) |
| 2024-12-29 00:26:21 +0100 | <haskellbridge> | <thirdofmay18081814goya> I'm reading "m i j a" as, the monad indexed by "i", such that "j" depends on "i" |
| 2024-12-29 00:26:37 +0100 | tromp | (~textual@92-110-219-57.cable.dynamic.v4.ziggo.nl) (Quit: My iMac has gone to sleep. ZZZzzz…) |
| 2024-12-29 00:30:17 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 265 seconds) |
| 2024-12-29 00:30:30 +0100 | <geekosaur> | looks like a straightforward generalization of (a -> m b) -> m a -> m b, with an assumption that the action produces both a new value type and a new index type to go with it, and the result type encodes the initial and final index types while discarding the intermediate one (j) because it becomes k when the action is run |
| 2024-12-29 00:32:01 +0100 | <geekosaur> | the initial state encodes i->j, then the action gives you j->k, meaning the result is i->k |
| 2024-12-29 00:32:41 +0100 | <haskellbridge> | <thirdofmay18081814goya> hm I'll think about the idea of discarding the intermediate index type, ty |
| 2024-12-29 00:33:10 +0100 | <haskellbridge> | <thirdofmay18081814goya> right ok |
| 2024-12-29 00:33:13 +0100 | <haskellbridge> | <thirdofmay18081814goya> that makes more sense, ok ty |
| 2024-12-29 00:33:21 +0100 | <geekosaur> | there isn't much to think about, any more than it's worth thinking about 2+3+4 giving you an intermediate 5 |
| 2024-12-29 00:33:42 +0100 | euleritian | (~euleritia@dynamic-176-006-135-074.176.6.pool.telefonica.de) |
| 2024-12-29 00:35:40 +0100 | <geekosaur> | AIUI the whole point is that the indexed monad tracks an initial and a final index. individual actions within it produce new "final" indexes which aren't the final index of the entire computation because the next action maps from the first "final" index to a new one |
| 2024-12-29 00:39:35 +0100 | <haskellbridge> | <thirdofmay18081814goya> ok right, I wasn't quite getting that in "m i j a", both "i" and "j" are indextypes |
| 2024-12-29 00:43:50 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2024-12-29 00:45:37 +0100 | foul_owl | (~kerry@193.42.0.124) (Ping timeout: 252 seconds) |
| 2024-12-29 00:46:55 +0100 | <haskellbridge> | <thirdofmay18081814goya> they are from the same indexing type* |
| 2024-12-29 00:47:36 +0100 | ash3en | (~Thunderbi@2a03:7846:b6eb:101:93ac:a90a:da67:f207) (Quit: ash3en) |
| 2024-12-29 00:47:51 +0100 | <haskellbridge> | <thirdofmay18081814goya> was reading that wrong |
| 2024-12-29 00:48:26 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 252 seconds) |
| 2024-12-29 00:52:35 +0100 | <haskellbridge> | <thirdofmay18081814goya> next question is... why two indexes? |
| 2024-12-29 00:53:11 +0100 | <haskellbridge> | <thirdofmay18081814goya> the usage of one index is understandable, e.g. the type of lists with a given length |
| 2024-12-29 00:54:10 +0100 | <haskellbridge> | <thirdofmay18081814goya> hm but maybe you'll two indexes if you're mapping from lists of length n to lists of length m? |
| 2024-12-29 00:54:22 +0100 | <haskellbridge> | <thirdofmay18081814goya> you'll want two indexed* |
| 2024-12-29 00:55:18 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) merijn |
| 2024-12-29 00:56:00 +0100 | gentauro | (~gentauro@user/gentauro) (Read error: Connection reset by peer) |
| 2024-12-29 00:59:55 +0100 | merijn | (~merijn@128-137-045-062.dynamic.caiway.nl) (Ping timeout: 252 seconds) |
| 2024-12-29 00:59:55 +0100 | <geekosaur> | think a and b having different Ord instances |
| 2024-12-29 01:00:21 +0100 | <geekosaur> | so it tracks the original and final Ord instances |
| 2024-12-29 01:00:36 +0100 | foul_owl | (~kerry@174-21-81-201.tukw.qwest.net) foul_owl |
| 2024-12-29 01:01:27 +0100 | lol_ | jcarpenter2 |
| 2024-12-29 01:01:59 +0100 | gentauro | (~gentauro@user/gentauro) gentauro |
| 2024-12-29 01:02:18 +0100 | <geekosaur> | (since one of the uses of indexed monads is to allow things like an indexed monad instance for Set, which normally isn't possible because changing the key type alters the structure to reflect a different Ord instance. AIUI at least) |
| 2024-12-29 01:02:54 +0100 | <geekosaur> | (meaning i, j, k are the key types since those specify the Ord instances, again AIUI) |