| 23 Jul 2021 |
 Dmitrii Okunev (at work) | And ironically, generics are not something personally I'm waiting for :( | 11:37:57 |
| Dmitrii Okunev (at work) | * And ironically, generics are not something I'm waiting for. | 11:38:01 |
| Dmitrii Okunev (at work) | * And ironically, generics are not something personally I'm waiting for :( | 11:38:09 |
 Moritz | I don't really care about generics either… I am just fearful of the things to come. | 11:39:10 |
| Moritz | I have one use case where they would be remotely useful | 11:39:26 |
| Dmitrii Okunev (at work) | * Basically the absence of immutable variables (except for string) is the main reason why I'm thinking about converting into Rust. | 11:45:26 |
|  alex.yov joined the room. | 12:17:18 |
 ezaquarii | I'd really appreciate easy way of defining custom error types, without implementing full interface. Generics probably could help here, even if it's limited like in case of a map. | 16:38:51 |
| ezaquarii | Interestingly, i don't see much issue with immutability as I don't see ppl doing bad stuff. Most are like Pythonistas: grown up and responsible (one of the arguments Python uses against strict protections). | 16:42:05 |
| Moritz | i had to put 128 big.Int into a function each instead of just making immutable variables… it's just a hazzle | 16:54:09 |
| Moritz | company policy | 16:57:00 |
| Moritz | "permission values must not be changable" so into functions they go | 16:58:45 |
| Moritz | why did we need 128 permission bits? because legacy software | 16:59:21 |
|  Squirrel joined the room. | 19:01:24 |
| Squirrel changed their display name from finnbear to Squirrel. | 19:02:33 |
| Squirrel set a profile picture. | 19:04:19 |
 Discordian | In reply to @vladyslav:matrix.webdevelop.pro
a lot
I personally don't see a big deal in that idm people using them ofc, but abusing them would be unfortunate IMO | 19:12:11 |
| Discordian | Biggest use-case I'd have for generics I can think of is just math. | 19:13:41 |
 diamondburned | I'll be using generics mostly for container data types | 19:33:30 |
| diamondburned | would be nice to have a glib.List[T] type | 19:33:47 |
| Moritz | being able to tell a function "every kind of number is cool" would be more than enough for me | 19:37:21 |
| 24 Jul 2021 |
|  markyjackson joined the room. | 08:53:49 |
| markyjackson set a profile picture. | 09:03:04 |
 condaatje | A (debatable) claim: there is no situation in golang where a generic would be superior to an interface | 20:25:15 |
| Discordian | Math library | 20:33:50 |
| Moritz | In reply to @moritz:moritz.sh
being able to tell a function "every kind of number is cool" would be more than enough for me condaatje: you can currently only do this by accepting the empty interface and doing a typeswitch, and even then you'd have to still cast the result in the target function. Or at least that is to the limit of my knowledge. Open for counterexamples | 20:46:32 |
| diamondburned | generics have some overlaps with interfaces, but the rest are not the same | 21:17:56 |
| diamondburned | generics are more for accepting certain types (>1), while interfaces are for accepting any type with the wanted methods | 21:18:30 |
|  mvdan joined the room. | 22:16:31 |
| 25 Jul 2021 |
|  Unsign3dRequ3st joined the room. | 01:42:11 |
