Simon Tournier <[email protected]> writes: > Hi, > > Below, a reply to Tomas. :-) > > On Fri, 19 Jun 2026 at 19:11, Ludovic Courtès <[email protected]> wrote: > >>> It looks ok, it would be cleaner to have a single source to maintain >>> but if the carry over can be automatized (I believe it could), it >>> shouldn't be a lot of effort. Why changing the module name though? >> >> Maybe to avoid confusion and/or incompatibilities? Support (guix >> records) evolves in incompatible ways that are okay for Guix and not >> okay for other uses. >> >> Maybe it’s a fine detail though. Maybe let’s just start with (guix >> records) imported as-is and we’ll see. > > About the name change, I’m not too worry with the potential confusion > and/or incompatibilities
I am bit curious, could you add some reasoning regarding why? I assumed it could lead to annoying-to-figure-out bugs, what did I overlook? Tomas -- There are only two hard things in Computer Science: cache invalidation, naming things and off-by-one errors.
