Slashdot Mirror
FreeBSD For The iMac And Other Eye-Openers
Anonymous Coward writes: "In this interview, part III, over at GNULinux.com you can read what Jordan Hubbard, CEO of FreeBSD Inc., has to say about the future of FreeBSD. '[...] Finally, the notion of the PC is changing. One could even argue that the PC has widened to encompass the PowerPC, because there are all these iMacs on peoples' desks, and according to the original mandate we should be looking at those iMacs, too, which is what we're doing,' sounds pretty sweet." This article provides a positive (but sober) overview of recent and anticipated progress from the devil-suit side of free software. You might also be interested in Jordan's answer to the question, "Can FreeBSD scale to the PDA?"
Once again, you have it backwards. Solaris is the result of a merger between SunOS and SVR4. SunOS is based on 4.1BSD and 4.2BSD which predate Solaris by 5-10 years.
The main obstacles are technical and political; the NetBSD people have insisted on a few substantial differences in kernel structures. Still, code has been shared in the past, and probably will be in the future.
Anyway, where do you think the Alpha port came from? Lots and lots of code was "borrowed". (In fact, this is one of the old grudges; some of the people who did the BSD-on-Alpha work originally feel they didn't get fairly credited for their work when it got borrowed.)
Anyway, the code is out there, and anyone can use it. Neat, huh? It's sort of like *actually* free software.
My blog: http://www.seebs.net/log/ --- My iPhone/iPad app: http://www.seebs.net/seebsfrac/
Since I started running things like Linux and FreeBSD, in the height of the PC/Mac wars, suddenly I was no longer taking sides.
Indeed, I don't care anymore what my hardware is. Even a 486 or a Sparcstation 1 is something I can use. Macs and PC's look pretty much the same to me unless I want to add some hardware to them, which I am comfortable with either way.
This is a really positive thing. Once the software runs on everything, the hardware will come to the aid of the software and I think things might become a bit more uniform.
Well, that's just wishful thinking. We aren't nearly as much a slave to the hardware manufacturers anymore, though, are we. "No, that costs too much for what it's worth. I will jump to architecture X instead. No skin off my nose."
--Gabriel
NetBSD/macppc already runs on the iMac. We've been there for over a year (since the 1.4 release in May 1999).
The PPC is known to be one of the worst targets, right up there with Alpha.
With IBMs commitment to Linux on the RS/6000 series I'd assume that they are going to be throwing some weight behind a gcc optimization for PPC. If they come out with an updated RS/6000 Thinkpad, I'm there, baby, I'm soooo there...
by Mike Buddha -- Someday the mountain might get him, but the law never will.
and PDA's - http://www.netbsd.org/Ports/hpcmips/.
Since the iMac now apparently has enough people using it to warrant a FreeBSD port, does this mean {the FreeBSD team, someone else} will help work on gcc optimization? The PPC is known to be one of the worst targets, right up there with Alpha.
Unless there is an MMU available, you can't run FreeBSD on it-- which excludes most PDAs right off the bat.
Is there a uFreeBSD project like uLinux? Maybe there should be...
by Mike Buddha -- Someday the mountain might get him, but the law never will.
I freely admit to not knowing as much about the BSD license as I do about the GPL: how much trouble, both technically and legally, would it be to merge the platform specific code from the various NetBSD ports into the FreeBSD tree?
(and would there be any point)
On a side note -- I presume "Gift-F" was the interviewer mishearing "#ifdef", right? So neither the author nor the editor bothered to understand their own sentence...
--
Lots of work by many experienced gurus has already gone into adapting FreeBSD for small applications. Currently, there is a development kit known as PicoBSD that enables one to automatically build floppy images that contain a minimal FreeBSD. Other people have created their own custom embedded versions of FreeBSD; the tight integration and cleanness of FreeBSD and its source tree make it quite easy for even an amateur to roll his/her own version. I myself am currently working on an improved development kit for building embeddded versions of FreeBSD quickly and easily. It's output is currently running off of an 8MB DiskOnChip on the desk to my right.
For more information, see Small FreeBSD Home Page. It's a bit outdated, but work is still actively going on. A maintainer is currently working on improving the site. To get at the very heart of things, subscribe to the freebsd-small mailing list (info here) or read the archive.
Washington, DC: It's like Hollywood for ugly people.
"the problem may have been that there was a stonehenge monument on the stage that was in danger of being crushed by a dwarf!" - David St. Hubbins, Spinal Tap
Interesting to read the bit about how FreeBSD linked up with Walnut Creek CDROM. Is it possible that without Walnut Creeks kind response, FreeBSD as we know it would either not exist, be later coming along???
"A goldfish was his muse, eternally amused"
Vs lbh pna ernq guvf, ybt bss abj. Tb bhgfvqr. Syl n xvgr.
386BSD came out, and the FreeBSD guys (who obviously weren't FreeBSD guys then) thought "Cool, UNIX for the masses. And it's free. We like this." But 386BSD never went anywhere. It jumped out of the blocks, but then stopped, and so FreeBSD started where 386BSD left off.
Their goal was to provide UNIX for the people out there. That meant providing BSD for the common hardware, the x86. It wasn't that they thought other platforms weren't good, or didn't deserve FreeBSD, it was (as I understand it) more a case of saying "We want to provide the best unix to the most people" and to devote their limited reources to small platforms was counter-productive.
FreeBSD was never supposed to be x86 for the sake of x86, any more than Linux was (which started off in 386 assembly). It was (and is) x86 because that's where people most needed it. Despite running on many architectures, Linux is still most popular on x86. It's where the market is. FreeBSD is highly pragmatic - don't do it cause it's cool, do it cause it acheives a purpose.
So, the conclusion then is, if the FreeBSD team feels that pouring some of the effort into a new architecture is going to provide greater overall benefit to the users, then it is within their original goals to do so.
x86 was a starting point, not an end.
--
Read more of this story at Slashdot.Read more of this story at Slashdot.Read more of this story at Slashdot.
"You are cordially invited to the marriage of frilly, fru-fru industrial design and the suave sophistication of an impenetrable user interface.
"You'll be the first on your block to have a computer no one wants to look at *or* use."
Every BSD iMac will come with a pocket protector and a black beret.
-- He's fantastic, made of plastic....
Are you on crack?!?
Sun paid SCO nearly a billion dollars for the rights to SVR4!
you won't be able to run Mac OS X binaries on Free- or NetBSD because they won't have access to the Carbon, Cocoa, or Quartz API's.
They can do Quartz/Cocoa once gnustep is done.
"There's very little progress on GNUStep!"
I assume this will see more progress as fellas will want to run their OS 10 apps without rebooting their boxen.
Will I retire or break 10K?