The circle of technology and FreeBSD

I've been reading the FreeBSD Release Notes for version 1.1, published in 1993. Two things are interesting to me here:

  1. The only name in the old core team I recognize is Jordan Hubbard, and he's not a FreeBSD developer any more (joined Apple for Mac OS X)
  2. Some problems, both technical and organizational, are always repeating themselves.

These things are "normal" and present not only in Open-source projects but in the entire industry (and of course in other industries), and are actually a sign of a healthy enteprise. Developers / employees arrive and leave, some stay as graybeard alumni to teach the next generation, etc. and the technology itself has a sort of cycle to itself that makes old things new again.

While the core list is almost full of unknown people, I recognize some names from the Additional FreeBSD helpers and beta testers list - like Julian Elischer, Jeffrey Hsu, Bruce Evans and Terry Lambert. AFAIK most if not all of them were at some point in time on the core roster. One of the most interesting aspects of the FreeBSD project (edit: actually of all early BSDs - OpenBSD was the first to adopt this trend) is the notable absence of a Linus Torvalds kind of person, one who has actively been the leader from its start. Instead, we have a list of people who have been with the project in many roles practically from its CSRG roots, some 30 years ago. To see why all this is good imagine if the reverse was true - if the original core team was active even now, same old developers working on same old things - it would certanly be a sign of stagnation.

Some organizational issues repeat themselves (again everywhere, not only in Open-source projects) - developers come and go, there are issues of supporting and charging for support, distribution and further development, etc. FreeBSD is also an interesting project from this aspect, as it has risen practically exclusively on its own merit, without support from huge companies pushing their agenda.

At last, technological problems reappear also, even completely mirrored from the past. In 1993 the problem all operating systems had was with ISA-based machines with more than 16 MB of memory (which was the maximum addressible by the bus), and now we have similar problem with low quality hardware (both motherboard and add-on cards) when used with more than 4 GB of memory (which is the maximum amount addressible in 32 bits). Same old, same old.

#1 Re:

Added on 2008-12-27T07:09 by dan

keep writing!

#2 Re:

Added on 2008-12-27T17:33 by Andy Kosela

Great post. I'm studying the history of UNIX and BSD and sometimes it's really hard to get to the facts of what really happened some 30-20 years ago. Also Theo is definetly the "Linus Torvalds" of OpenBSD. He started NetBSD with Chris Demetriou and to my knowledge he is the only one from the first generation BSD hobbyists (circa 1992-1993) still very active in the community. I guess the similar situation is in the Linux camp. There are not too many people from the early nineties that still do heavy kernel development.

#3 Re:

Added on 2008-12-27T20:53 by anon

From the "core" group, I also remember John Dyson. IIRC he was one of the early VM-hackers before Matt Dillon took over (in some interview, Dillon complained that Dyson never put comments in his code ;))

#4 Re: big leaders

Added on 2008-12-27T22:02 by Ivan Voras

Thanks #2; the trend starting with Theo (and later with Matt in DragonflyBSD) is important enough to mention!

#5 Re: big leaders

Added on 2008-12-27T22:56 by Andy Kosela

Yes, I forget about Matt Dillon. Also phk@ is still contributing a lot of new code nowadays. He has been around since very early days... I just wonder one thing - where is David Greenman ? He used to be the "principal architect" of FreeBSD.

#6 Re: big leaders

Added on 2008-12-28T22:18 by hilti

Hm, I think Garrett A. Wollman is still aktive. If I don't confuse him with some other one you should know him as wollman@.

#7 Re: big leaders

Added on 2008-12-31T00:45 by Ivan Voras

I don't recognize Garret Wollman, though I see he has a developer account. Unfortunately it appears he has last logged in in June.

PHK is alive and well, doing interesting work not in the FreeBSD itself but a lot of side-work involving FreeBSD). He's also very active in developers mailing list(s). A living legend - I hope he implements more of his ideas :)

Comments !