Hacker News Reader

I sometimes wonder if UI/UX passes on open-source tools would be better with very explicit "experimental forks" that keep with upstream when they can. Over time, the successful fork starts to get more traffic from regular users and re-merges. It might save time compared to long debates without much agenda.

View on HN

I'm fairly firmly of this opinion (though I only have a little experience working on large open-source projects). Sure, there can be productivity loss in developing in parallel vs developing a consensus, but nothing highlights the strengths and weaknesses of a design like actually implementing and using it.

The challenge for end-user focussed design is that it is very resource intensive to develop parallel UIs (and for many devs they are the least fun thing to implement), whereas it is not that much of a problem to add an additional parameter to a command line program.

> I sometimes wonder if UI/UX passes on open-source tools would be better with very explicit "experimental forks" that keep with upstream when they can. Over time, the successful fork starts to get more traffic from regular users and re-merges.

Probably the best example of this approach is the Firefox (originally Phoenix, then Firebird) origin story:

https://www.wired.com/2005/02/firefox/

Reply on news.ycombinator.com