Hacker News new | past | comments | ask | show | jobs | submit login

This seems like a very similar situation to egcs and gcc: a fork created because the original project proved too painful to work with, and folded back in after sorting out the project governance issues that caused the fork in the first place.



It's not quite as drastic: with egcs they basically just killed gcc and renamed egcs, while here it was more of a merger of the projects. Very similar, though.


They are in a funny way opposite (and I certainly know whereof I speak: look at who wrote this message [0]).

It's hard to remember, but back in 1997 forking was considered really terrible and to be avoided at all costs. When we split off to form egcs the main resistance was to the idea that there would be a fork at all. Gcc, and a bunch of other programs, had a single person who controlled the "official" release. The fact that Cygnus maintained its own release tree for its customers (it was still free -- it simply wasn't identical to the mainline, and in fact advanced a lot faster) was the cause of much angst and even some mistrust.

My reason for making this fork was because the mainline was so far behind us, and we had a commitment to folding all our changes back into the main line. Since the gcc maintainer was the bottleneck, we simply declared our tree a new fork with the support and participation of other major developers. We were trying to be the opposite of exclusivist. But even so it took me months of mailing, calling and negotiating. Many hours on the phone with rms who of course predicted the doom of free software if we went ahead.

And we developed a steering committee, the first as far as I know for a free software project.

In those two ways egcs was a watershed; plenty of developers thankfully now know no other way.

So last year glibc went the opposite way: it decided it no longer needed a steering committee. Cool!

[0] https://gcc.gnu.org/news/announcement.html


rms was (is?) probably still traumatized from the emacs/xemacs split, which is still there and, as far as I can see, does impede emacs development.


The former might be true, but it's very unlikely the latter, hard and harsh experience has shown that in the long term few things can impede development more than having RMS as a gatekeeper.

He's a difficult person, or so I observed when we were in the same social circle, later when I was working for LMI while he was replicating the most essential new Zetalisp functionality, when we both had given up on that and were roommates when he launched the GNU project, and afterwords when for a while I worked for a "Software Hoarder", specifically the company that actually had a license for (James) Goslings Emacs, which GNU Emacs is an illicit fork of. (He even accused that company of setting fire to his apartment building; in truth it was pretty generic arson by "a couple of kids", not surprising because it was in a not so nice part of Cambridge.)


xemacs looks dead to me. 2009 was the last official release. See http://www.xemacs.org/Releases/index.html#Stable


as an outsider, how does the (x)emacs split impede development?

It seems emacs has been steadily, if somewhat slowly, improving (new releases etc).


Well, it was a bit of a hunch, based on the slowness of development. Other commenters have maybe given better explanations.




Join us for AI Startup School this June 16-17 in San Francisco!

Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: