3 ms·
Can break down where the money likely went, how it was spent?
by solarkraft 9y ago
Can break down where the money likely went, how it was spent?
- pacaro 9y agoI remember writing code for Psion (later Symbian) Epoc32 in the late 90s and early thousands. It was a big deal that Cygnus had done the work to make gcc support ARM because the vendor compiler was (IIRC) some weird combiniation of expensive and not very good. We now assume ARM as being a default necessary platform to support, 20 years ago it really wasn’t clear
- gumby 9y agoThis was between 1989 (when we started in my living room) and 1999 when I left. By 1999 we had about 160-170 developers working on gcc, gdb, binutils, glibc, Cygwin etc. The money was a mixture of 1> ports (typically paid for by the manufacturer), 2> maintenance (mostly paid for by the corporate end user -- for $100K/year we'd fix your bug whether we spent $100 to make a trivial fix or $300K on a major problem) 3> maintenance on very old versions that people had to freeze on for one reason or another and 4> the public tree (we wouldn't work on non-free software, apart from a few abortive experiments after I had left). Making sure everything we wrote was compatible with the public tree was expensive. In the end, one of the last things I did saved a lot of money and improved code quality for everyone: I forked gcc from the FSF (and made the egcs tre the de facto, and eventually de jure version). Nowadays forking is considered routine, and even beneficial, and projects have steering committees that are responsible for master releases. At that time, forks were considered a tragedy, and it took me over 6 months of discussion with various people to build a reasonable consensus (you can see the people's name at the bottom of the letter I wrote announcing it ( https://gcc.gnu.org/news/announcement.html https://gcc.gnu.org/news/announcement.html ). It's been a good model. Note that RMS's name is not on the letter -- he was furious and was sure this would destroy the FSF. Instead it has strengthened it. By the way, to compare these numbers: Red Hat went public with, AFAIK less than 10 engineers and when Cygnus and Red Hat merged, though both companies had about 200 people, Cygnus had over 160 engineers while Red hat had a couple of dozen. Of course we had a lot more revenue, and in the years since RHAT has invested heavily in free software development. But their company value at that time was something like 3X ours -- a very valuable lesson.
- oblio 9y agoHow did Cygwin come about?