some notes on EGCS

some notes on EGCS

Posted May 15, 2008 4:20 UTC (Thu) by JoeBuck (subscriber, #2330)
Parent article: The Freedom of Fork

The EGCS effort was successful in part because it was carefully designed to make a re-merge, or even a takeover of FSF GCC development, possible, as well as a very careful effort to keep people with completely incompatible desires happy. I have to admit that there was a considerable amount of spin involved: we had to simultaneously persuade people to believe completely opposite things about what EGCS was: a temporary experimental branch that would eventually get merged back, therefore no threat? Complete liberation from dealing with RMS ever again? Freedom for Cygnus management to control GCC? A hacker playground? An effort to put ESR's CatB book into practice? There were people who wanted to believe each of those things.

But the goal from a very early stage, before the public ever heard about egcs, was gcc3: not a fork, but a takeover, so that there would be only one GCC, based on EGCS. For this to be possible, we needed to require papers assigning all the code to the FSF, we followed the GNU style rules. Even so, it took about a year of negotiation for the FSF to accept EGCS as the new GCC and give the EGCS team control of the compiler.

---

(Log in to post comments)