Newest at the top
| 2026-02-16 13:48:02 +0100 | lantti | takes a sip |
| 2026-02-16 13:47:07 +0100 | <lantti> | proper tea is theft |
| 2026-02-16 13:46:31 +0100 | haritz | (~hrtz@user/haritz) haritz |
| 2026-02-16 13:46:31 +0100 | haritz | (~hrtz@140.228.70.141) (Changing host) |
| 2026-02-16 13:46:31 +0100 | haritz | (~hrtz@140.228.70.141) |
| 2026-02-16 13:44:46 +0100 | weary-traveler | (~user@user/user363627) user363627 |
| 2026-02-16 13:31:08 +0100 | weary-traveler | (~user@user/user363627) (Remote host closed the connection) |
| 2026-02-16 13:18:03 +0100 | <haskellbridge> | <magic_rb> đź‘Ť youre wrong, but i agree, i dont want to be ruining my morning with this shit again |
| 2026-02-16 13:18:03 +0100 | <haskellbridge> | <Morj> But let's leave this topic |
| 2026-02-16 13:18:03 +0100 | <haskellbridge> | <Morj> magic_rb: yes, information sharing is good and piracy is moral |
| 2026-02-16 13:15:28 +0100 | chele | (~chele@user/chele) chele |
| 2026-02-16 13:15:16 +0100 | chele | (~chele@user/chele) (Remote host closed the connection) |
| 2026-02-16 13:04:33 +0100 | CiaoSen | (~Jura@2a02:8071:64e1:da0:5a47:caff:fe78:33db) (Ping timeout: 260 seconds) |
| 2026-02-16 13:02:03 +0100 | wickedjargon | (~user@2605:8d80:5430:8a6:d1eb:8774:e3ee:c949) (Ping timeout: 252 seconds) |
| 2026-02-16 12:47:53 +0100 | Guest12 | (~Guest12@145.107.83.45) (Ping timeout: 272 seconds) |
| 2026-02-16 12:39:59 +0100 | <lortabac> | both Gemini pro and Claude Opus are very good at matching specs with implementation in my experience |
| 2026-02-16 12:39:12 +0100 | <lortabac> | by "context" I meant explaining the motivation behind the patch |
| 2026-02-16 12:17:22 +0100 | <haskellbridge> | <magic_rb> Right so your argument for paying for LLMs is "the LLM tech giant has stolen knowledge from folks much more knowledgeable than me and i want that knowledge so its okay" |
| 2026-02-16 12:17:19 +0100 | <haskellbridge> | <Morj> Yeah, I know about context. Usually if it's a one-file program or a self-contained module, I will send the whole file to review. If it's a multi-file diff the model won't have context, true |
| 2026-02-16 12:17:19 +0100 | <haskellbridge> | <Morj> My company has subscriptions to gemini and claude, I use those (on my personal projects, hehe) |
| 2026-02-16 12:17:17 +0100 | <haskellbridge> | <Morj> Yea, and I expect there are people who know more about haskell than me. It's not like I write terribly complicated code on the evenings |
| 2026-02-16 12:10:10 +0100 | sp1ff | (~user@2601:1c2:4701:7850::1844) (Ping timeout: 246 seconds) |
| 2026-02-16 12:09:28 +0100 | sp1ff` | (~user@2601:1c2:4701:7850::8cd) |
| 2026-02-16 12:04:48 +0100 | xff0x | (~xff0x@2405:6580:b080:900:6b27:25be:1343:b5d0) |
| 2026-02-16 11:59:44 +0100 | <lortabac> | of course, you should give it some context about the change. A generic "please review this patch" won't do it |
| 2026-02-16 11:58:35 +0100 | <lortabac> | Morj: which models do you use? I find the most recent LLM models very good at code reviews. Sometimes they identify subtle issues that can easily go unnoticed |
| 2026-02-16 11:53:16 +0100 | <jreicher> | What's your problem? The basic design of an LLM is to tell you something somebody else already knows. |
| 2026-02-16 11:47:29 +0100 | <haskellbridge> | <Morj> Good technology, what did I even metaphorically burned a tree for |
| 2026-02-16 11:47:28 +0100 | <haskellbridge> | <Morj> I like how sometimes I ask an LLM to review my code, and it only points to places where I already left FIXMEs |
| 2026-02-16 11:45:35 +0100 | Guest12 | (~Guest12@145.107.83.45) |
| 2026-02-16 11:27:00 +0100 | divlamir | (~divlamir@user/divlamir) divlamir |
| 2026-02-16 11:26:46 +0100 | divlamir | (~divlamir@user/divlamir) (Read error: Connection reset by peer) |
| 2026-02-16 11:08:15 +0100 | xff0x | (~xff0x@fsb6a9491c.tkyc517.ap.nuro.jp) (Ping timeout: 245 seconds) |
| 2026-02-16 11:00:32 +0100 | dutchie | (~dutchie@user/dutchie) dutchie |
| 2026-02-16 10:59:33 +0100 | dutchie | (~dutchie@user/dutchie) (Remote host closed the connection) |
| 2026-02-16 10:58:02 +0100 | Sgeo | (~Sgeo@user/sgeo) (Read error: Connection reset by peer) |
| 2026-02-16 10:53:59 +0100 | xff0x_ | xff0x |
| 2026-02-16 10:53:27 +0100 | Enrico63 | (~Enrico63@host-82-63-21-32.business.telecomitalia.it) (Quit: Client closed) |
| 2026-02-16 10:53:14 +0100 | xff0x | (~xff0x@fsb6a9491c.tkyc517.ap.nuro.jp) (Ping timeout: 252 seconds) |
| 2026-02-16 10:53:00 +0100 | stefan-__ | (~m-yh2rcc@42dots.de) stefan-__ |
| 2026-02-16 10:52:01 +0100 | xff0x_ | (~xff0x@fsb6a9491c.tkyc517.ap.nuro.jp) |
| 2026-02-16 10:39:55 +0100 | <srazkvt> | yea that's because they know nothing |
| 2026-02-16 10:37:02 +0100 | hakutaku | (~textual@user/hakutaku) (Quit: My MacBook has gone to sleep. ZZZzzz…) |
| 2026-02-16 10:33:38 +0100 | <kaol> | Silly AI. It told me to do flip showHex "" to "ensure each byte is converted to a two-digit hexadecimal string". No, it won't. |
| 2026-02-16 10:32:59 +0100 | qqq | (~qqq@185.54.22.240) |
| 2026-02-16 10:31:14 +0100 | hakutaku | (~textual@user/hakutaku) hakutaku |
| 2026-02-16 10:30:35 +0100 | tromp | (~textual@2001:1c00:3487:1b00:b457:3c8c:4ce1:c4df) |
| 2026-02-16 10:27:59 +0100 | <yahb2> | -41 |
| 2026-02-16 10:27:58 +0100 | <pavonia> | % subtract 42 1 |
| 2026-02-16 10:24:06 +0100 | <yahb2> | -41 |