login
Search
As originally submitted to KernelTrap by gncuster and reported on OSNews, the AtheOS operating system has been forked by Bill Hayden. The new creation is temporarily called New Atheos while Bill secures the domain names and trademark for the official and as-of-yet unnannounced name. Essentially he has merged the AtheOS and BeOS API's, porting it all to run on the Linux kernel. This results in significantly increased driver support, powerPC support, and the ability for most BeOS programs to compile and run with little or no changes. There has not yet been any source released, nor a date set for the official release.
Bill says, "I forked Atheos about 6 months ago and have been continuously developing it since that time. I've taken it in some very new directions. I should warn you that some of you will absolutely love the changes, and some of you will perhaps feel that the "dream" of Atheos has been sold out." Bill's announcement and much of the resulting thread follows. Find more information about AtheOS in this earlier review from OSNews.
From: Bill Hayden
Date: 03/26/2002 06:59:50
Subject: [Atheos-developer] Atheos Fork Announcement
Well, it was not my intent to announce this quite this soon, but given
the recent conversation on the list, I feel that it's best not to wait
any longer.
I forked Atheos about 6 months ago and have been continuously developing
it since that time. I've taken it in some very new directions. I
should warn you that some of you will absolutely love the changes, and
some of you will perhaps feel that the 'dream' of Atheos has been sold out.
The new project has had a name since the beginning, but I'm going to
hold off on releasing that until I can verify that the domain names and
trademark are secure. So I'll call it "New Atheos" for the purposes of
this e-mail.
New Atheos has the following major new features:
o Runs on top of the Linux kernel, not the Atheos kernel
o Atheos API has been merged with the BeOS API
o PowerPC support
o gcc 3.0.X compatiblity
o OpenTracker/Deskbar desktop manager
These features give the following benefits:
o Most BeOS programs compile and run with little or no changes
o Linux kernel means that CD-ROM, CD booting and installing, DHCP, etc.
work
o Linux kernel means that driver support is excellent
o Mac users get a piece of the action
Things I haven't even started on:
o Printing
o Media Kit
o Replicants
Existing Atheos programs will need changes to compile. I haven't found
one that took me longer than a few minutes to "convert". Where Atheos
and BeOS use different semantics, I chose the BeOS method.
I am going to hold off on a release until I can successfully compile and
run OpenTracker and Deskbar. They use just about every obsolete and
goofy BeOS construct that exists. I'm most of the way there, though,
especially for Deskbar. Kurt wasn't lying when he said it would be a
nightmare to port them. Of course, I'm doing an "anti-port". When some
BeOS program won't compile, I change the API to match it instead of
changing the program itself.
The first BeOS program that successfully came up was Pulse, and there
was no small amoung of satisfaction to see good ole' Pulse running on my
new system. Nostalgic BeOS users can perhaps understand.
I'm writing in a hurry, so hopefully I haven't forgotten something
important. And no, I can't give a release date yet. I hope to have a
CVS server up at the time of release.
Thanks,
Bill Hayden
From: J.Brown
Date: 03/26/2002 07:41:39
Subject: RE: [Atheos-developer] Atheos Fork Announcement
This in itself makes it a quite horrible prospect.
While I appretiate the work you may have put into it, Atheos was
fundementally it's kernel. What you've done is little more than port the
atheos API (fairly simple), and invented a pretty new GUI. Yuck... IMHO
it's quite silly :)
> New Atheos has the following major new features:
> o Runs on top of the Linux kernel, not the Atheos kernel
From: Bill Hayden
Date: 03/26/2002 07:57:40
Subject: RE: [Atheos-developer] Atheos Fork Announcement
Obviously you fall into the "killing the dream of Atheos" camp. :-)
Can you please explain to me what makes the Atheos kernel so compelling?
It an honest question -- I'm interested to hear specifically what are
its advantages. Here's what I looked at:
o Prospect of future support
o Cross-platform
o Driver availability
o Speed (granted, not a lot of study has been done here)
Does the Atheos kernel win any of these battles? Will it in the next
year? in 2 years? I'm betting on "no", and feeling pretty safe about that.
The compelling feature of Atheos has always been the appserver/gui IMHO.
Everyone agrees that XWindows needs to die, but no one has put forth a
worthy replacement... yet. :-)
Thanks,
Bill Hayden
From: Jon Doda
Date: 03/26/2002 08:18:51
Subject: RE: [Atheos-developer] Atheos Fork Announcement
While Bill Hayden's work is a little far in left field to have a direct
implication on AtheOS it does raise an idea that becomes more attractive
every time I think about it. Namely, porting the linux kernel to
AtheOS. We are not talking here about making AtheOS "just another
window manager" but using the linux kernel as a virtual drop in
replacement for AtheOS' current native kernel. If it can be made
transparent to the end user then the advantages of such a move are
obvious and staggering.
The real question then becomes how do we make the change transparent?
It probably wouldn't be all that hard as the linux kernel provides
virtually all the functionality that the AtheOS kernel does, save for
one area, the vfs. Node-watching, extended attributes, attribute
indexing and live queries would all have to be added, and then an AFS
driver would have to be written. These changes may require almost an
entire rewrite of the linux vfs but I think it would be worth the effort.
In exchange we get drastically improved driver support, a kernel that
most open source developers are familiar with, all the CLI tools you can
eat, excellent stability and security, and a host of other good things
:). I think it's something to at least think about instead of just
dismissing it out of hand.
--
Jon Doda
From: Bill Hayden
Date: 03/26/2002 08:52:03
Subject: RE: [Atheos-developer] Atheos Fork Announcement
Jon Doda wrote:
> Node-watching, extended attributes, attribute indexing and live
> queries would all have to be added...
Linux can do node watching as of kernel 2.4. It is implemented in "New
Atheos".
> In exchange we get drastically improved driver support, a kernel that
> most open source developers are familiar with, all the CLI tools you
> can eat, excellent stability and security, and a host of other good
> things :). I think it's something to at least think about instead of
> just dismissing it out of hand.
You seem to be promoting almost exactly what I've done. Except for
attribute/query support, and an AFS driver, it's just what you've asked for.
Thanks,
Bill Hayden
From: Jon Doda
Date: 03/26/2002 09:12:44
Subject: RE: [Atheos-developer] Atheos Fork Announcement
It seems more like you're using the AtheOS appserver as a starting point
for a BeOS clone than actually continuing AtheOS development. Now,
that's a pretty clever thing to do if you want a BeOS clone, but it's
not AtheOS.
--
Jon Doda
> I forked Atheos about 6 months ago and have been continuously developing
> it since that time. I've taken it in some very new directions. I
> should warn you that some of you will absolutely love the changes, and
> some of you will perhaps feel that the "dream" of Atheos has been sold
> out.
Well.. when I saw this, I thought : Wow, great idea.
> o Runs on top of the Linux kernel, not the Atheos kernel
When I saw this atrocity, I wanted to die.!
Come on, Linux kernel is awful, has a terrible design idea since its
beginning, it has more support that atheos kernel now, but that's not
the point. Making that sort of change to the OS kills it's whole idea,
everything what is great in atheos kernel, is a disaster on Linux. (even
if some functions are not implemented yet)
Speaking about the Atheos API has been merged with the BeOS API, and the
GCC3 compatibility, well, that can be discussed.(I start liking that
idea :)
And I ask you Bill, would it be too hard to re-port your new AtheosApi
to work with the real Atheos kernel?, because, we could have 2 distros,
one more functional (at the beginning) with the Linux kernel and another
one with real Atheos kernel; but where the high level api's, the GUI,
and everything else than the kernel are binary and code compatible.
It's an option.
Just my Ar$0.06
Pablo M
Pablius.net
Everyone,
End this thread now. Give Kurt a week or two to respond, it's his baby
after all- give the man some respect. If he doesn't reply, THEN talk
about how and what you want to fork. It's incredibly rude and
presumptuous to be talking about how to go about doing a fork before you
give him due time to speak his mind.
Furthermore, Bill, a fork needs an agreed upon leader. Not a self
appointed one. Otherwise atheos won't fork, it'll splinter into a
million useless threads of development.
-Dale
From: Bill Hayden
Date: 03/26/2002 10:37:47
Subject: RE: [Atheos-developer] Atheos Fork Announcement
Dale wrote:
> End this thread now. Give Kurt a week or two to respond, it's his baby
> after all- give the man some respect. If he doesn't reply, THEN talk
> about how and what you want to fork. It's incredibly rude and
> presumptuous to be talking about how to go about doing a fork before
> you give him due time to speak his mind.
My project is not Atheos. Atheos will continue to exist.
> Furthermore, Bill, a fork needs an agreed upon leader. Not a self
> appointed one.
My project is not Atheos. Atheos will continue to exist. (repeat to
self 5 times)
I am not requesting to be the new leader of of Atheos, and will not
serve if elected. :-) I consider what I am doing to be a new product
entirely. Like Kurt, I'm did this for myself, and if others see value
in it they can use it too. The contention that I am trying to be the
new Atheos "leader" or just doing it for self-aggrandizement is
ludicrous. I just thought some folks might be interested, and everybody
with an opinion started coming out of the woodwork. Interestingly, with
few exceptions the positive comments have been addressed to me
personally, and the negative comments have been addressed to the list.
There's probably a psychologic study in that fact somewhere.
> Otherwise atheos won't fork, it'll splinter into a million useless
> threads of development.
I would submit that the /current/ thread of development is useless
unless drastic action is taken. Which I did. Free software is free.
You can use it or not, and do with it what you will. That, of course,
is the point.
Thanks,
Bill Hayden
Atheos has great things, that linux lacks:
-No module to insert, so no reboot is required.
-Node-watching
-extended attributes
-attribute indexing
-live queries (working 100%?)
-micro kernel
-a real modular design
-fast booting
-real consistence between kernel and gui
-server's idea (be like)
Obviously, its not perfect, and need a lot of development to gain Linux
functionality, but Atheos kernel is well designed since its beginning,
so it will take not a lot of time until it reaches linux level (with our
help and work).
From: Bill Hayden
Date: 03/26/2002 10:48:28
Subject: RE: [Atheos-developer] Atheos Fork Announcement
Pablo M wrote:
>Atheos has great things, that linux lacks:
>
>-No module to insert, so no reboot is required.
>
Linux has had this, what, forever? You must be thinking of Windows.
>-Node-watching
>
Linux >= 2.4 has this.
>-extended attributes
>-attribute indexing
>-live queries (working 100%?)
>
These would be very nice, and are the only real argument against this
project.
>
>-micro kernel
>
Atheos in its present state is not much of a microkernel at all. Kurt
says so himself on the website. Microkernels are not the answer to all
kernel woes anyway. Their only real advantage is portability, but Linux
doesn't suffer from a problem there, so it's moot.
>-a real modular design
>
I'm concerned about the real world. No businessperson is going to say
"Everyone usesWindows/Linux, but I think I'll use Atheos because it has
modular design!". Would you rather have ReiserFS, CD-ROM support, DHCP
support, etc., or have a warm fuzzy feeling because your kernel has
modular design?
>-fast booting
>
If Atheos supported all the features of Linux, the boot time would be
similar.
>
>-real consistence between kernel and gui
>
Not sure what you mean here. The kernel has very little to do with the
GUI. The Atheos Appserver/GUI on Linux *is* the Appserver/GUI on Atheos.
>
>-server's idea (be like)
>
This is "Microkernel" listed a second time. See above.
Cool ideas and practical ideas are two different beasts. I took the
practical route.
Thanks,
Bill Hayden
Pablo M wrote:
> Atheos has great things, that linux lacks:
> -No module to insert, so no reboot is required.
Kernel and FS drivers are kernel modules just like in Linux. They're
loaded automatically in AtheOS, which is nice, but not something you
couldn't do on Linux. The driver model in AtheOS is quite a bit nicer
than Linux's right now, but the Linux driver model is getting overhauled
in the 2.5 tree and it looks like it will be quite nice by the end of it.
> -Node-watching
Exists in Linux, apparently :)
> -extended attributes
This is a biggie, but it's probably less work to add it to Linux then it
is to add all the stuff the AtheOS kernel needs that the Linux kernel
already has.
> -attribute indexing
> -live queries (working 100%?)
Neither of these are implemented in AtheOS either, though they are
definitely neccesary.
> -micro kernel
AtheOS is no more a microkernel than Linux is. Both are modular
monolithic kernels.
> -a real modular design
er, see above.
> -fast booting
I'll give you this one.
> -real consistence between kernel and gui
The gui isn't really tied to the kernel in any real way. It's already
been ported to linux by two different people.
> -server's idea (be like)
This has nothing to do with the kernel at all.
I don't want to give the impression that I'm really pushing for using
the linux kernel. I'm still very undecided on the matter. However, it
is an interesting idea and it deserves some real thought, not just
uninformed FUD.
--
Jon Doda
On Tue, 26 Mar 2002 15:10:35 -0300
"Pablo M" wrote:
> Atheos has great things, that linux lacks:
> -extended attributes
Just a note to everyone who is complaining about this on. It's done and going
to be/being merged in 2.5 right now. Not only that, XFS at least has support
for it, and ext2/3 has experimental support for it. Reiserfs 4 will have
support (October). Back to your regularly scheduled argument
Daniel
What do you guys want from AtheOS?
The only thing I want is a distracting little hobby.
The beauty of AtheOS is the threading, the messaging
and the appserver. The filesystem is also neat, and
those attributes are just crying out for fresh uses.
The move to push AtheOS into the mainstream and the
moves to support POSIX and BeOS and this that and the
other is really distracting from the beauty of AtheOS
IMO.
AtheOS is basically an empty house. Why are we trying
to bring furnishings from all our other houses?
Instead we should be sitting in the workshop making
really cool *new* stuff. Play around with innovative
ideas. See if they work, or don't or whatnot.
There seems to be two types of AtheOSist- the dreamer
who is trying to make new things and try new ideas;
and the porters who want to make AtheOS into a M$
killer by making it more and more like 1000 other
operating systems...
If you want Linux/BeOS/*BSD, why not just use
Linux/BeOS/*BSD?
Guess I am getting as passionate about this as
everyone else on the list, so I'll finish with a :-)
Excuse me one moment while I do this....GRRRRRRAAAAAAAAAGHHHHHHH (Imagine
the noise the Incredible Hulk makes...thats me now) Its been a bad 24
hours. The computer has died, I have spent hours trying to get the net
connection working in Windows95 on the second box, and now I have to use
Webmail which sucks badly to send this message. Now I have that off of
my chest, I shall continue...
First of all, I'm as frustrated as Brent. Secondly, Bill, ignore the nay-sayers.
Putting parts of AtheOS onto the Linux kernel is an obvious solution to
the current shortcommings of AtheOS (namely the VM & drivers). However,
it sounds like you're trying to write something more like OpenBeOS or BlueOS,
and that isn't AtheOS. If thats what you want to do though, go for it,
hack code, have fun!
Now, onto the forking issue. I have suggested the idea of a fork myself
on several occasions, sometimes on the list and sometimes not. AtheOS
is in serious jepordy of stagnating, and now faces compatition from OpenBeOS,
BlueOS (And Bills system too now I guess!). A fork of the code, along
with a more open development model, would address these issues. If you
view AtheOS as a serious attempt to provide a useful Open source dekstop
then I'm guessing that you'll agree with a port of AtheOS & accelerating
development by doing so. If you view AtheOS as a hobby, then I guess you'll
be horrified at the idea. Guess which camp I fall into....
So, if someone is willing to head a project as large a forking an entire
OS (Pablo?) then why not? The worst you can do is fail. Kurt will more
than likely still be around & hacking on his source, so AtheOS is unlikely
to be damaged in the short term by the emergance of a port.
I'm going to stop now, this is the second time I have written this email...
--
Vanders
http://www.shagged.org/~vanders/
From: Pablo M
Date: 03/26/2002 17:32:10
Sunject: RE: [Atheos-developer] Atheos Fork Announcement
> So, if someone is willing to head a project as large a forking an entire
> OS (Pablo?) then why not? The worst you can do is fail. Kurt will more
> than likely still be around & hacking on his source, so AtheOS is unlikely
> to be damaged in the short term by the emergance of a port.
Well, the magical words have been said, I'll fork this beast, until Kurt
reappear and takes over the control, or something like that.
I'll be needing some help here coz it's going to be a huge task, anybody
interested contact me privately.
In the next 24 hours I'll release some sort of web page with information
about the new fork :)
I hope you Brent have no opposition to have this new fork appearing
under Kamidake (at the end, this is AtheOS)
As I said before, anybody that has made modifications or upgrades to the
Atheos core and has not publicly released them (or has but are not part
of the official distro) please send them to me so I start to apply them
to the kernel.
Also anybody has any CVS server lying around so we can use it?
I have a DSL connection that can be used, but it will be really slow
(128Kb Uplink)
Pablo M
Pablius.net
