IRC channel logs
2023-06-27.log
back to list of logs
<oriansj>pabs3: Microsoft's .net-core has binaries the last time I checked but I believe one can use Mono to bootstrap those binaries; so it is quite likely Mono will be needed for C# bootstrapping for the ongoing future <fossy>pabs3: it's mostly a manual review process. often generated code has a manual "don't change" header or something like that, so i usually look through "grep generated". another sign is huge diffs for that file in git history (i usually only do this if it's reasonably easy to do). big files are also often pregenerated too <oriansj>janneke: yeah, everyday C# is slipping farther and farther away from having a proper bootstrap. It seems like it needs a parent with a firm hand on it to keep it from running into traffic. <oriansj>muurkha: to your previous question about the inkplate: Chip is ESP32-D0WDQ6 (revision 1) with 4MB flash; so definitely 2 Xtensa 32bit cores <Christoph[m]>There are languages with so much boilerplate that IDEs try to help generate that. Would that be admissible? <oriansj>Christoph[m]: well for 0) it is not source code <oriansj>boilerplate and libraries that you optionally include are still written by people and not generated blobs (sometimes megabytes in size that completely change on every release) <stikonas[m]>Though there are some cases where it's harder to tell <stikonas[m]>Then cleaned up a bit and switched to being handwritten <AwesomeAdam54321>Can programs written in Squeak Smalltalk be bootstrapped? I'm not sure how they can actually be compiled into VM images from source <stikonas[m]>Though it might have been bootstrapped with non free software <stikonas[m]>Though I'm not familiar with squeel Smalltalk myself... <Christoph[m]>I was not aware of the size of the problem! Megabytes of changing code in a repository, that's not the optimal use of git! And it's certainly not reviewable!