Refreshing
Refreshing
Posted Jun 9, 2021 20:51 UTC (Wed) by khim (subscriber, #9252)In reply to: Refreshing by nix
Parent article: Rewriting the GNU Coreutils in Rust
You have made me really curious.
Specifically that:
> nearly-unreadable ultra-low-contrast dark blue on black
And of course what I have find is something which is really well-readable and nice looking. While yellow used for README.md and Makefile is a bit hard for me to read (but not hard enough to declare them not usable).
Finding good colors is hard. People are different and significant percentage of males have issues there. Thus I'm not sure “oh, default colors don't look nice to me — means the whole project is unsuitable for everyone” is a good attitude.
> but the whole point of exa is to make better decisions by defaultBut what is “better decisions by default”? We may aim to create better peception for the majority of people (the ones with no defects) or we may try to make sure our color scheme is poor-but-acceptable for most people (including the ones with vision defects). My guess would be that ls and axe aim for the 1st (which makes sense for highly-configurable tool) while web sites are usually designed for the 2nd.
This creates strange dilemma: colors which are best for actual program are not best for the website dedicated to that program… so what should be done about that? Should the website use some non-standard palette to make it acceptable for the majority or use the actual default colors program uses? I don't think these are easy decisions and I don't think exa is big enough project to afford market research…
