Login | Register For Free | Help
Search for: (Advanced)

Mailing List Archive: Gentoo: OSX

Some Introduction

 

 

Gentoo osx RSS feed   Index | Next | Previous | View Threaded


grobian at gentoo

Aug 7, 2005, 3:22 AM

Post #1 of 15 (1292 views)
Permalink
Some Introduction

Hi all,

I think most of you have already noticed me in some way, either via some
announcements on -dev, by reading a blog, or just by paying attention in
IRC. Anywayz, for those that don't know, I'm a new developer on the OSX
team.

In the small week that I've been officially on the staff now, I was
confronted with many small things that made me ponder. Before going
into a long mail, I'll apologise upfront for my English, it's horrible.

When Lina send me an email after I submitted some bug on bugzilla, I
felt like hey, wow, etc. Basically I still feel like that. It's an
honour to be part of the Gentoo team somehow. That it can't be fun
everytime proves that recent resignation message on -dev I think.
However, let's stick to the main point: I like to contribute some useful
things here, and I hope I can find a corner where I can be that useful
as I hope to be. At the moment I have the terrible feeling of being
useless, doing nothing struggling with everything that gets on my path.

I'm not really an IRC guy. I know what it is, but in general it's great
in distracting you and stopping you from doing what you have to do. Due
to my time zone, I usually miss the important discussions too. Hence,
I'm thinking of a drastical reduction of my IRC online time. I have the
feeling most of the OSX staff is in the #-osx channel, but it simply
doesn't work out so well for me. I prefer the asynchronous way of
email, it also allows me to take some more time to type a response. As
a non-native English typer, I need more time to come up with responses.
And usually, it's time zone free! ;)

I got a fuzzy image of what the OSX team currently consists of. It's
far from a unity, more a group of people somewhere related because of a
shared OS, most of the time. Personally I'm a bit lost in what the
general consensus would be among the team members. Maybe there isn't
even one. There is progressive, darwin, osx, etc. the arch ppc-macos
seems to be a multi-headed dragon.

My vision on Portage for OSX is exactly what the name says; portage on
OSX, thus a portage instance next to the original OS, so I can enjoy the
flexibility and package availability of portage and the sweetness of my
OS. I am willing to accept that I can't install autofs on a Mac OS X
machine. Maybe it sucks, but then you better install Linux on it
afterall. A Mac is different, thinks different, and yet, well... maybe
I just like that. In portage terms this is called "collission-protect".
Great!

Now it seems to me, after paying careful attention to some of the
comments made in the #-osx channel that this vision of mine, which
equals the current 'distribution' I think, can be considered the
unwanted child in the Gentoo family. Ok, it will be always a bastard
child, like Portaris would be, but someone started with this idea, and
got it into portage somehow. How did this whole thing emerge within the
Gentoo community, and what happened afterwards to get into the stage it
is in now?

Ok, this probably all sounds a bit depressing, or put differently, quite
unpromissing. However, all I need for now is some guide into the
wilderness I guess. What are the (common) targets of the team? What is
it 'we' want to achieve? Who thinks what?

Last, but not least, to get a small impression of the people in this
team, I would like a (very) short intro of those people that I haven't
met yet (if there are any ;) )

I hope somehow to become a valuable/active member of the team, but so
far I think I haven't had the opportunity to do so.


--
Fabian Groffen
eBuild && Porting
Gentoo for Mac OS X
--
gentoo-osx [at] gentoo mailing list


kito at gentoo

Aug 7, 2005, 10:48 AM

Post #2 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 7, 2005, at 5:22 AM, Grobian wrote:

> Hi all,

Hi!


> In the small week that I've been officially on the staff now, I was
> confronted with many small things that made me ponder. Before
> going into a long mail, I'll apologise upfront for my English, it's
> horrible.

Don't sell yourself short, your english is pretty damn good IMHO =)


> At the moment I have the terrible feeling of being useless, doing
> nothing struggling with everything that gets on my path.
>

You've only been here a week. There is no hurry, you will get
familiar with all the things you need to...Its opensource development
so people tend to work on their areas of interest.

> I'm not really an IRC guy. I know what it is, but in general it's
> great in distracting you and stopping you from doing what you have
> to do. Due to my time zone, I usually miss the important
> discussions too. Hence, I'm thinking of a drastical reduction of
> my IRC online time. I have the feeling most of the OSX staff is in
> the #-osx channel, but it simply doesn't work out so well for me.
> I prefer the asynchronous way of email, it also allows me to take
> some more time to type a response. As a non-native English typer,
> I need more time to come up with responses. And usually, it's time
> zone free! ;)

Well I'll be sad to not see you in #-osx as much :(

>
> I got a fuzzy image of what the OSX team currently consists of.
> It's far from a unity, more a group of people somewhere related
> because of a shared OS, most of the time. Personally I'm a bit
> lost in what the general consensus would be among the team
> members. Maybe there isn't even one. There is progressive,
> darwin, osx, etc. the arch ppc-macos seems to be a multi-headed
> dragon.
>
> My vision on Portage for OSX is exactly what the name says; portage
> on OSX, thus a portage instance next to the original OS, so I can
> enjoy the flexibility and package availability of portage and the
> sweetness of my OS. I am willing to accept that I can't install
> autofs on a Mac OS X machine. Maybe it sucks, but then you better
> install Linux on it afterall. A Mac is different, thinks
> different, and yet, well... maybe I just like that. In portage
> terms this is called "collission-protect". Great!
>
> Now it seems to me, after paying careful attention to some of the
> comments made in the #-osx channel that this vision of mine, which
> equals the current 'distribution' I think, can be considered the
> unwanted child in the Gentoo family. Ok, it will be always a
> bastard child, like Portaris would be, but someone started with
> this idea, and got it into portage somehow. How did this whole
> thing emerge within the Gentoo community, and what happened
> afterwards to get into the stage it is in now?

Daniel Robbins the Gentoo founder did the initial port to OS X,
Pieter van den Abeele (Found of Gentoo/ppc) released the first pkg
installer, setup what little infrastructure the OS X team has and
actually started recruiting devs and getting stuff in the tree. It
all happened very quickly and half-assed, a lot of people were pushed
through the recruitment/mentoring process(myself included) ,the tree
was broken many many times(myself included), resulting in Gentoo as a
whole hating the OS X project(myself included). Most people since
then have fallen off the project, be it out of shame (several went to
gentoo/ppc), lack of interest(several disappeared all together), or
on to bigger and better things (Pieter is working on some very cool
stuff with freescale/genesii as well as the opensolaris project).

Which pretty much brings us to where we are today....4-5 of us left
wandering around aimlessly.

>
> Ok, this probably all sounds a bit depressing, or put differently,
> quite unpromissing. However, all I need for now is some guide into
> the wilderness I guess. What are the (common) targets of the
> team? What is it 'we' want to achieve? Who thinks what?

Well, my basic feeling is the current method of trying to accommodate
Apple supplied userland is futile, its working against the advantages
of the portage tree. All the apps in portage are tested/tweaked/
hacked/patched/whatever to work with THE APPS IN PORTAGE, not with
3rd party software supplied by arbitrary vendors. In other words, the
path of least resistance is allowing portage to do what it does best,
manage software that is has knowledge of, instead of us constantly
lying to it about deps,etc. i.e. when an ebuild has DEPEND="app-
shells/bash", that means it depends on the bash in portage, if you
have the bash from BeOS/FC4/FBSD/Darwin/NewOS/Whatever its not
supported, and likely to cause problems somewhere down the line...


So, am I saying portage on OS X is a dumb idea? Hell no. Am I saying
the only way I see it working is overwriting Apple files? Hell no.

My personal goals for the project in order of my interest:

A self-hosting Darwin OS build-able and maintainable via portage

A custom Darwin/OS X installer for specialty applications with
portage managing the data (small footprint, SEDarwin/OS X, custom
userlands, Kiosks, etc.)

A fast, up to date binary package repo ala Fink that installs its
software in an arbitrary prefix

Each one of these goals is quite a large subject and beyond the scope
of this email, but hopefully you get the idea of where my interests
lie...

The first 2 goals, I work on now, mostly with help from people
outside of Gentoo. The last one, is the biggie and requires a lot of
work both on the actual ebuilds as well as portage itself. The good
news is it *is* being worked on for the next major portage
release.....but still a ways off regardless.

Sooooo, I guess you can tell I have no interest in the current method
of installing things to / or /usr and constantly trying sidestep,
hack around, and accomodate portage apps co-existing with the apple
userland. Its not practical, is very short sighted, will continue to
pollute the tree with ugly unnecessary cruft, and will never have the
support of Gentoo as a whole IMHO.

I by no means think all of our porting efforts thus far are wasted,
as when/if portage has support for prefixed installs, alot of the
grunt work of getting things building on Darwin/OS X will already
have been done.

>
> Last, but not least, to get a small impression of the people in
> this team, I would like a (very) short intro of those people that I
> haven't met yet (if there are any ;) )
>
> I hope somehow to become a valuable/active member of the team, but
> so far I think I haven't had the opportunity to do so.

I've no doubt you are a huge asset to the project, regardless of
where your niche is/might be.

--Kito

>
>
> --
> Fabian Groffen
> eBuild && Porting
> Gentoo for Mac OS X
> --
> gentoo-osx [at] gentoo mailing list
>
>

--
gentoo-osx [at] gentoo mailing list


fthain at telegraphics

Aug 7, 2005, 9:48 PM

Post #3 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

Firstly can I say thanks for your efforts, I certainly appreciate it, and
the same for the other Gentoo/OS X devs. It is a tough gig.

On Sun, 7 Aug 2005, Kito wrote:

> >
> >Ok, this probably all sounds a bit depressing, or put differently,
> >quite unpromissing. However, all I need for now is some guide into the
> >wilderness I guess. What are the (common) targets of the team? What
> >is it 'we' want to achieve? Who thinks what?
>
> Well, my basic feeling is the current method of trying to accommodate
> Apple supplied userland is futile, its working against the advantages of
> the portage tree.

Grobain, he reason this approach was taken (and this can be found in the
list archives and forum), was that there was no alternative. The mythical
pathspec was supposed to provide it. And, significantly, pathspec wasn't
going to be needed to have a Gentoo/Darwin distro, anyway.

Kito, because of Gentoo/Darwin, this is not futile. And it saves space on
your own machine. OS X has a decent binary package manager, why not
exploit it? Updating a dot release on OS X is easier than emerge -Du
system.

And if portage acquires the ability to pull deps from several prefixes
(which is apparently part of the plan) it sacrifices nothing to use OS X
deps.

> All the apps in portage are tested/tweaked/hacked/patched/whatever to
> work with THE APPS IN PORTAGE, not with 3rd party software supplied by
> arbitrary vendors.

There are two problems here: one is the vendor deps, i.e. the present
practice of package.providing deps that are not equivalent to the ones OS
X provides. This problem goes away as soon as we get ebuilds that can
produce some of apples opensource packages that we use as deps. Apple
being apple, this was always a big ask, but darwinbuild should help.

The second problem is that devs now have the job of making sure that the
portage tree is portable: ebuilds have to work with Portaris,
Gentoo/Darwin, Gentoo/BSD (a lot like Darwin), as well as Gentoo/Linux.
This means a few more guidelines for writing ebuilds, but adds great value
to the portage tree.

> In other words, the path of least resistance is allowing portage to do
> what it does best, manage software that is has knowledge of, instead of
> us constantly lying to it about deps,etc. i.e. when an ebuild has
> DEPEND="app-shells/bash", that means it depends on the bash in portage,
> if you have the bash from BeOS/FC4/FBSD/Darwin/NewOS/Whatever its not
> supported, and likely to cause problems somewhere down the line...

Exactly. And this is not new, see the ML archives for the threads on
"optional os x packages" and "zip, and more, in package.provided", etc.

> So, am I saying portage on OS X is a dumb idea? Hell no. Am I saying the
> only way I see it working is overwriting Apple files? Hell no.
>
> My personal goals for the project in order of my interest:
>
> A self-hosting Darwin OS build-able and maintainable via portage

Would you use a GNU userland (like GNU/Darwin) or an apple one (like
OpenDarwin and OS X)?

The reason I ask is that the former might mean leveraging the existing
portage tree, whereas the latter might mean creating ebuilds for apple
packages, thus solving the old package.provided problem.

-f
--
gentoo-osx [at] gentoo mailing list


kito at gentoo

Aug 7, 2005, 10:38 PM

Post #4 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 7, 2005, at 11:48 PM, Finn Thain wrote:

>
> Firstly can I say thanks for your efforts, I certainly appreciate
> it, and
> the same for the other Gentoo/OS X devs. It is a tough gig.

=)

>
> On Sun, 7 Aug 2005, Kito wrote:
>
>
>>>
>>> Ok, this probably all sounds a bit depressing, or put differently,
>>> quite unpromissing. However, all I need for now is some guide
>>> into the
>>> wilderness I guess. What are the (common) targets of the team?
>>> What
>>> is it 'we' want to achieve? Who thinks what?
>>>
>>
>> Well, my basic feeling is the current method of trying to accommodate
>> Apple supplied userland is futile, its working against the
>> advantages of
>> the portage tree.
>>
>
> Grobain, he reason this approach was taken (and this can be found
> in the
> list archives and forum), was that there was no alternative. The
> mythical
> pathspec was supposed to provide it. And, significantly, pathspec
> wasn't
> going to be needed to have a Gentoo/Darwin distro, anyway.
>
> Kito, because of Gentoo/Darwin, this is not futile. And it saves
> space on
> your own machine.

Maybe you misunderstood, what I think is futile is trying to avoid
overwriting files, and accommodating things portage has no knowledge
of or control over.

> OS X has a decent binary package manager, why not
> exploit it? Updating a dot release on OS X is easier than emerge -Du
> system.

Because relying on Apples proprietary stuff is asking for trouble,
its liable to change at any given moment without notice. However,
emerge'ing from .pkgs is in testing as we speak...

>
> And if portage acquires the ability to pull deps from several prefixes
> (which is apparently part of the plan) it sacrifices nothing to use
> OS X
> deps.

It sacrifices huge amounts of manhours, ebuild consistency, and
again, what works with apples current releases will probably not work
in future releases...Its pretty hard to second guess at team of >30
engineers on a project without live cvs...

>
>
>> All the apps in portage are tested/tweaked/hacked/patched/whatever to
>> work with THE APPS IN PORTAGE, not with 3rd party software
>> supplied by
>> arbitrary vendors.
>>
>
> There are two problems here: one is the vendor deps, i.e. the present
> practice of package.providing deps that are not equivalent to the
> ones OS
> X provides. This problem goes away as soon as we get ebuilds that can
> produce some of apples opensource packages that we use as deps. Apple
> being apple, this was always a big ask, but darwinbuild should help.

I don't think the problem goes away as much as it changes, although
for the better IMHO and this is one of the very things I work on
frequently. darwinbuild is a great tool, I've watched it grow and
helped hack on it a little bit. Still not quite sure how it will fit
into portage in the long run...

>
> The second problem is that devs now have the job of making sure
> that the
> portage tree is portable: ebuilds have to work with Portaris,
> Gentoo/Darwin, Gentoo/BSD (a lot like Darwin),

I wouldn't say a *lot* :p

> as well as Gentoo/Linux.
> This means a few more guidelines for writing ebuilds, but adds
> great value
> to the portage tree.

Agreed.

>
>
>> In other words, the path of least resistance is allowing portage
>> to do
>> what it does best, manage software that is has knowledge of,
>> instead of
>> us constantly lying to it about deps,etc. i.e. when an ebuild has
>> DEPEND="app-shells/bash", that means it depends on the bash in
>> portage,
>> if you have the bash from BeOS/FC4/FBSD/Darwin/NewOS/Whatever its not
>> supported, and likely to cause problems somewhere down the line...
>>
>
> Exactly. And this is not new, see the ML archives for the threads on
> "optional os x packages" and "zip, and more, in package.provided",
> etc.
>
>
>> So, am I saying portage on OS X is a dumb idea? Hell no. Am I
>> saying the
>> only way I see it working is overwriting Apple files? Hell no.
>>
>> My personal goals for the project in order of my interest:
>>
>> A self-hosting Darwin OS build-able and maintainable via portage
>>
>
> Would you use a GNU userland (like GNU/Darwin) or an apple one (like
> OpenDarwin and OS X)?

Probably somewhere in between. I have the base system building right
now, slowly getting some of apples patches into things like python,
bash, bsdmake, etc.

>
> The reason I ask is that the former might mean leveraging the existing
> portage tree, whereas the latter might mean creating ebuilds for apple
> packages, thus solving the old package.provided problem.

I don't think there is a silver bullet here. I already have ebuilds
for 50 or so darwin system packages, although not in the portage tree
yet until things get stabilized more, but I don't think it will be
practical to have one or the other. The main challenge in working
with Darwin is keeping up with Apples engineers...no real live cvs is
a major pain-in-the-ass, so the source just comes in huge dumps every
few months, and with the current G/Darwin dev team of well, me, its a
slow process indeed. Any help is appreciated ;)

Thanks a bunch for your input, its always nice to hear from people
who actually care about this project.

--kito

>
> -f
> --
> gentoo-osx [at] gentoo mailing list
>
>

--
gentoo-osx [at] gentoo mailing list


me at caphector

Aug 7, 2005, 10:56 PM

Post #5 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 7, 2005, at 10:38 PM, Kito wrote:

>
> Because relying on Apples proprietary stuff is asking for trouble,
> its liable to change at any given moment without notice. However,
> emerge'ing from .pkgs is in testing as we speak...

AFAIK, Apple only changes the .pkg format on major releases. Looking
over the release history of Pacifist ( http://www.versiontracker.com/
dyn/moreinfo/macosx/12743 ), it appears that he only needs to update
package handing for major releases.

While it's not perfect, it does keep stuff simpler. You should also
be able to hook into the installer command-line utility and need to
use fewer wrappers.

Cap'n Hector

--
gentoo-osx [at] gentoo mailing list


fthain at telegraphics

Aug 7, 2005, 11:52 PM

Post #6 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Mon, 8 Aug 2005, Kito wrote:

>
> On Aug 7, 2005, at 11:48 PM, Finn Thain wrote:
>
> >
> >On Sun, 7 Aug 2005, Kito wrote:
> >
[snip]
> > >
> > >Well, my basic feeling is the current method of trying to accommodate
> > >Apple supplied userland is futile, its working against the advantages of
> > >the portage tree.
> > >
> >
> >Grobain, he reason this approach was taken (and this can be found in the
> >list archives and forum), was that there was no alternative. The mythical
> >pathspec was supposed to provide it. And, significantly, pathspec wasn't
> >going to be needed to have a Gentoo/Darwin distro, anyway.
> >
> >Kito, because of Gentoo/Darwin, this is not futile. And it saves space on
> >your own machine.
>
> Maybe you misunderstood, what I think is futile is trying to avoid overwriting
> files, and accommodating things portage has no knowledge of or control over.

Yes. Collision-protect was always a hack. Far better to do prefixed
installs for those that want Gentoo to play second-class citizen to OS X.

> >OS X has a decent binary package manager, why not exploit it? Updating
> >a dot release on OS X is easier than emerge -Du system.
>
> Because relying on Apples proprietary stuff is asking for trouble, its
> liable to change at any given moment without notice.

Once the ebuilds for Apple's stuff are in the tree, maybe we could update
package.provided from /Library/Receipts automatically?

> However, emerge'ing from .pkgs is in testing as we speak...

Nice!

>
> >
> >And if portage acquires the ability to pull deps from several prefixes
> >(which is apparently part of the plan) it sacrifices nothing to use OS
> >X deps.
>
> It sacrifices huge amounts of manhours, ebuild consistency

Intuitively, it seems to me that maybe 80% of the logic in the average
ebuild is relevant to the package and doesn't relate to the platform.
(Most free software is highly portable.) The remainder of the ebuild is
either Gentoo specific or Gentoo/Linux specific.

Now obviously, the Linux specific bits are the problem, but it isn't just
Gentoo/OS X and Gentoo/Darwin with this problem, it is the other two
"red-headed stepchildren" as well -- Solaris and BSD. So it is not as if
you are facing the Linux-specific-logic burden alone. More importantly,
this logic can be fixed piecemeal over time, one ebuild at a time, no?

I'm sure the above is not new to you, and I admit to being ignorant of the
extent of the problems you are dealing with.

> and again, what works with apples current releases will probably not
> work in future releases...Its pretty hard to second guess at team of >30
> engineers on a project without live cvs...

This is very true, Apple doesn't have a great record for backward
compatibilty in OS X. But, let's face it. No-one can reasonably expect
every ebuild to keep working after they update their OS. The sad fact is
that they will anyway, and that creates a support issue for you.

What to do? Maybe we could have a file in the profile to say that a given
vendor package whose version exceeds a certain upper bound should be
ignored, unless ~ppc-macos was in effect. i.e. the OS X dep would
effectively disappear (or block?) unless the user did an emerge sync after
Software Update, by which time the problem might be fixed in the tree.

On second thoughts, I wouldn't overload package.provided with this
functionality. I'd probably have an automatically generated
/etc/portage/profile/vendor.provided and a dev issued
/etc/make.profile/vendor.stable...

>
> >
> > >All the apps in portage are tested/tweaked/hacked/patched/whatever to
> > >work with THE APPS IN PORTAGE, not with 3rd party software supplied
> > >by arbitrary vendors.
> > >
> >
> >There are two problems here: one is the vendor deps, i.e. the present
> >practice of package.providing deps that are not equivalent to the ones
> >OS X provides. This problem goes away as soon as we get ebuilds that
> >can produce some of apples opensource packages that we use as deps.
> >Apple being apple, this was always a big ask, but darwinbuild should
> >help.
>
> I don't think the problem goes away as much as it changes, although for
> the better IMHO and this is one of the very things I work on frequently.
> darwinbuild is a great tool, I've watched it grow and helped hack on it
> a little bit. Still not quite sure how it will fit into portage in the
> long run...

It may not need integration into portage. Perhaps what is needed is an
ebuild to install it and create the chroot, plus an eclass to allow an
apple package to get itself built by darwinbuild?

[snip]
> > >
> > >My personal goals for the project in order of my interest:
> > >
> > > A self-hosting Darwin OS build-able and maintainable via portage
> > >
> >
> >Would you use a GNU userland (like GNU/Darwin) or an apple one (like
> >OpenDarwin and OS X)?
>
> Probably somewhere in between. I have the base system building right
> now, slowly getting some of apples patches into things like python,
> bash, bsdmake, etc.

Wow, this is great.

It would also be cool to have an Darwin system that could emerge itself
from stage1 using Apple sources, where the stage1 tarball included
darwinbuild. If only I had time to hack on this...

-f
--
gentoo-osx [at] gentoo mailing list


gongloo at gentoo

Aug 8, 2005, 4:36 PM

Post #7 of 15 (1261 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 7, 2005, at 05:22, Grobian wrote:

> Anywayz, for those that don't know, I'm a new developer on the OSX
> team.

If they don't know, they've been oblivious of your name all over
bugzilla. Good work! :)

> Before going into a long mail, I'll apologise upfront for my
> English, it's horrible.

We know your English is good since you make the same spelling
mistakes us 'native speakers' make. *mumbles something about how us
native speakers aren't so native*

> I like to contribute some useful things here, and I hope I can find
> a corner where I can be that useful as I hope to be.

We're all here on our own free will (hopefully). Those that aren't,
and realize it, generally leave pretty quickly. You've seen this on
gentoo-dev@. As such, we do what we're interested in. Your job as a
dev is to enjoy what you do while avoiding tree breakage. If you're
able to, you might even fix a few things along the way, or improve
the project.

Take Lina. Lina is a biologist first, computer scientist second. As a
result, she took an immediate interest in the sci-biology category.
Most of the packages there have been ported now, and are maintained
by her for ppc-macos. She keeps on top of the relevant RSS feeds, and
keeps track of stable-by dates in iCal. She even maintains a sci-
biology package or two. This is her interest, and her first priority
with respect to Gentoo.

While Lina enjoys more instantaneous gratification, I'm more happy
working on long-term goals and slowly making progress one step at a
time. My interests are different from hers, and we work in different
areas. This is not to say that we don't work together, just that we
work on different sides of the same coin, at the same time, sometimes
cooperating towards a common goal.

You will have to find a corner for yourself. If anyone tells you what
to do, you may or may not like it. You are free to find any corner
that you are sure to like. If you think you like something and change
your mind, just move to another corner. You seem to be into the DB
side of things, perhaps you'd enjoy porting and maintaining some
database services (mysql comes to mind) for ppc-macos. The only
person who knows what you're enjoy is you. Again, devs are here to
have a good time, not to be slaves.

> At the moment I have the terrible feeling of being useless, doing
> nothing struggling with everything that gets on my path.

The majority of our "struggling" comes from inadequate upstream
support for Darwin. This means that the most important thing you can
do in terms of porting software is to send all your source patches,
no matter how trivial, upstream. Sending patches upstream is
nontrivial itself: you have to check out the latest source tree from
CVS (or svn), get it working, and send back the results. You will
probably end up porting less packages per unit time this way, but it
is better in the long run. It's better to patch something once than
for every successive version that comes out. That way, we struggle
once per package.

We obviously can't say why you're feeling useless, but we know that
we personally get that feeling due to the slow nature of making long-
term progress as an arch. Large change takes time, especially when
those changes involve portage itself. As developers, we don't feel
like we're making much progress by porting a few libraries and
applications, but, to the user base, this is more progress than those
large long-term projects Hasan is so fond of. Imagine how many people
you'd affect by porting mysql, for example.

> I'm not really an IRC guy. I know what it is, but in general it's
> great in distracting you and stopping you from doing what you have
> to do. Due to my time zone, I usually miss the important
> discussions too. Hence, I'm thinking of a drastical reduction of
> my IRC online time. I have the feeling most of the OSX staff is in
> the #-osx channel, but it simply doesn't work out so well for me. I
> prefer the asynchronous way of email, it also allows me to take
> some more time to type a response. As a non-native English typer,
> I need more time to come up with responses. And usually, it's time
> zone free! ;)

There are plenty of devs with exactly the same stance. Don't worry --
that's perfectly fine. There's no _requirement_ for you to be
available via IRC.

If you've noticed, you rarely see us on IRC, except when we need to
get in contact with other developers. Like you, we feel it's more
distracting than helpful most of the time. Just drop in once in a
while and let us know that you're still alive, or send an e-mail out
every so often. Your relations thus far have been flawless.

> I got a fuzzy image of what the OSX team currently consists of.
> It's far from a unity, more a group of people somewhere related
> because of a shared OS, most of the time. Personally I'm a bit lost
> in what the general consensus would be among the team members.
> Maybe there isn't even one. There is progressive, darwin, osx,
> etc. the arch ppc-macos seems to be a multi-headed dragon.

Your image is essentially correct, in our opinion. Right now, we have
about as many developers as we do profiles -- well, maybe not quite.
Your perception of the team being far from unity probably stems from
the fact that we more or less have one or two developers working on
different facets of the OSX port. Kito and Robert (we think) are
working on the darwin port and the progressive profile. JoseJX works
on perl and baselayout, mainly on his free time from the ppc team. We
(Lina and Hasan) work on the collision-protect profile, for which you
were recruited.

We hope to increase the number of developers for the Gentoo for Mac
OS X project, but we're doing so slowly since our last recruitment
process was a disaster (one that Hasan and I were part of as new devs
brought in during that time). You were our first pick, so you're
seeing the beginning of the process to get a coherent team together.

Along those lines, we probably need to elect a new strategic and
operational lead as Pieter (pvdabeel) has other priorities at this
time, and Ciaran (ciaranm), a senior developer that stepped in to
help us with QA and operations initially when Pieter got too busy,
has decided to no longer act in that capacity.

We have been doing our best to pick up the slack and start some of
the processes that the leads are responsible for, such as recruitment
and interacting with the portage developers to get necessary bugs
ironed out.

We would like to hold an election for the positions that we have been
more-or-less filling unofficially, as the new metastructure suggests.
We would like to be candidates for Operation Lead (Lina) and
Strategic Lead (Hasan). Obviously, anyone else who wants the position
should announce candidacy. We don't suppose that it is necessary
(pretty much everyone on the team knows each other fairly well by
now), but if anyone would like a short blurb on why we feel we would
be good candidates for the positions, just let us know. Is a week
enough time to allow anouncements for candidacy? We feel that a long
process isn't necessary with so few developers on the team to date.
If everybody would rather have an informal IRC session or email tally
than a vote, that's fine with us.

> My vision on Portage for OSX is exactly what the name says; portage
> on OSX, thus a portage instance next to the original OS, so I can
> enjoy the flexibility and package availability of portage and the
> sweetness of my OS. I am willing to accept that I can't install
> autofs on a Mac OS X machine. Maybe it sucks, but then you better
> install Linux on it afterall. A Mac is different, thinks
> different, and yet, well... maybe I just like that. In portage
> terms this is called "collission-protect". Great!

We have the same vision.

> Now it seems to me, after paying careful attention to some of the
> comments made in the #-osx channel that this vision of mine, which
> equals the current 'distribution' I think, can be considered the
> unwanted child in the Gentoo family. Ok, it will be always a
> bastard child, like Portaris would be, but someone started with
> this idea, and got it into portage somehow. How did this whole
> thing emerge within the Gentoo community, and what happened
> afterwards to get into the stage it is in now?

Kito did a pretty good job summing this up in his own reply. I think
it's important to note that this vision is changing in terms of
respect for Gentoo for Mac OS X developers. A lot of the old-time
devs still voice dislike for the idea of a metadistribution, either
because of the way it's being implemented or because they preferred
things when it was just Gentoo Linux.

> Ok, this probably all sounds a bit depressing, or put differently,
> quite unpromissing. However, all I need for now is some guide into
> the wilderness I guess. What are the (common) targets of the
> team? What is it 'we' want to achieve? Who thinks what?

The team has a number of different targets. Our targets differ from
Kito's, as do our views on the efficacy of portage co-existing with
Mac OS X. I think a common goal is a working implementation of
prefixable installs, but this is a goal for the next major portage
release, so it will take some time. Currently, Lina's goal is to
increase upstream support for darwin, and to maintain Gentoo for Mac
OS X as a viable platform for biologists. Hasan is involved with both
long-term and short-term strategic needs of the project, in terms of
interacting with the portage team, core system tools (pretty much
anything that's base-system), and marketing (getting the most desired
packages working, documentation, etc.).

It's really hard to have a common target without having any leads.
We're working on it. I think that we need to recognize two major
different facets of the Gentoo for Mac OS X project -- the Darwin
side and the Mac OS X side. The strongest tie between these, in terms
of porting, being common linker woes.

> I hope somehow to become a valuable/active member of the team, but
> so far I think I haven't had the opportunity to do so.

You already are. Just don't push yourself too hard. :) Large change/
progress takes time.

--

Hasan Khalil && Lina Pezzella
eBuild and Porting Co-Leads
Gentoo for Mac OS X
Attachments: PGP.sig (0.18 KB)


kito at gentoo

Aug 8, 2005, 5:16 PM

Post #8 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 8, 2005, at 6:36 PM, Hasan Khalil wrote:

> We would like to hold an election for the positions that we have
> been more-or-less filling unofficially, as the new metastructure
> suggests. We would like to be candidates for Operation Lead (Lina)
> and Strategic Lead (Hasan).

I was wondering how long it was going to take you guys....

Who gets to vote in this election?

Shouldn't the entire Gentoo dev team have some input?

Shouldn't this wait until after the council elections are finished?

What will change once you two have been elected?

You should also probably address the fact that there is still a
strategic lead in place (pvdabeel) before deciding amongst
yourselves to hold an election...

I know titles can be fun and addictive, but I'm not sure declaring an
election to lead a team of 4 people, when 2 of them are basically
jointly running, is the Right Thing(tm) at this point in time... yes
the project needs leadership but this smells a little funny.

--Kito
--
gentoo-osx [at] gentoo mailing list


grobian at gentoo

Aug 9, 2005, 12:36 AM

Post #9 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

Kito! Think Different(tm)! :)

Just my two newbee cents:
1) I've heard this too many times now I guess, so I think it would be a
Good Thing(tm) if Pieter -- assuming you all speak the truth and he's gone
for good, or near to it -- would resign from his "lead" job. A king in
another country is of no use, as he cannot reasonably know what his
citizen need.
2) Every country, how small it may be, needs a king! Why? Doesn't that
sound rediculous for only 4 people? Why do you think the Dutch (me) have
a Queen? The Ustated Nights are approx. a million times bigger, and still
we have one. Why? Because that's the person to pick when you want to
have some official contact to the group. Or the country. It's more a
symbolic task, especially in this small group, but good for the image to
the outside. A group without a leader is by definition an unorganised
bunch of eh... hackers.

It's all about perception, if you look professional, people think you are.
Given the fact ppc-macos is considered to be the pain in the place where
the sun don't shine, I think we would benefit from an organised (and
professional) look. New faces, new attitutes, new contacts. Damn, it
looks like I'm giving a marketing course here. Welcome to the world of a
Business Information Systems guy!

So, Lina and Hasan, I think what you want to do is a (c)Good Start(tm)(r).nl


On Tue, August 9, 2005 02:16, Kito wrote:
>
> On Aug 8, 2005, at 6:36 PM, Hasan Khalil wrote:
>
>> We would like to hold an election for the positions that we have
>> been more-or-less filling unofficially, as the new metastructure
>> suggests. We would like to be candidates for Operation Lead (Lina)
>> and Strategic Lead (Hasan).
>
> I was wondering how long it was going to take you guys....
>
> Who gets to vote in this election?
>
> Shouldn't the entire Gentoo dev team have some input?
>
> Shouldn't this wait until after the council elections are finished?
>
> What will change once you two have been elected?
>
> You should also probably address the fact that there is still a
> strategic lead in place (pvdabeel) before deciding amongst
> yourselves to hold an election...
>
> I know titles can be fun and addictive, but I'm not sure declaring an
> election to lead a team of 4 people, when 2 of them are basically
> jointly running, is the Right Thing(tm) at this point in time... yes
> the project needs leadership but this smells a little funny.
>
> --Kito

--
gentoo-osx [at] gentoo mailing list


kito at gentoo

Aug 9, 2005, 11:39 AM

Post #10 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 9, 2005, at 2:36 AM, Grobian wrote:

> Kito! Think Different(tm)! :)
>
> Just my two newbee cents:
> 1) I've heard this too many times now I guess, so I think it would
> be a
> Good Thing(tm) if Pieter -- assuming you all speak the truth and
> he's gone
> for good, or near to it -- would resign from his "lead" job. A
> king in
> another country is of no use, as he cannot reasonably know what his
> citizen need.

I agree, I was just pointing out that it needs to be dealt with
before any changes happen.

> 2) Every country, how small it may be, needs a king! Why? Doesn't
> that
> sound rediculous for only 4 people? Why do you think the Dutch
> (me) have
> a Queen? The Ustated Nights are approx. a million times bigger,
> and still
> we have one. Why? Because that's the person to pick when you want to
> have some official contact to the group. Or the country. It's more a
> symbolic task, especially in this small group, but good for the
> image to
> the outside. A group without a leader is by definition an unorganised
> bunch of eh... hackers.

I <3 your analogies ;)

Again, agreed! I'm not saying we don't need a leader in place, I'm
saying we don't need another leader in name only...IMHO the ideal
situation is a current Gentoo dev-team leader who would be interested
in migrating to OS X.

>
> It's all about perception, if you look professional, people think
> you are.
> Given the fact ppc-macos is considered to be the pain in the place
> where
> the sun don't shine, I think we would benefit from an organised (and
> professional) look. New faces, new attitutes, new contacts. Damn, it
> looks like I'm giving a marketing course here. Welcome to the
> world of a
> Business Information Systems guy!

Its not *all* about perception... lots of PR,hype, and empty titles
without alot of code to back it up is one of the things that got us
in trouble in the first place.

>
> So, Lina and Hasan, I think what you want to do is a (c)Good Start
> (tm)(r).nl
>
>
> On Tue, August 9, 2005 02:16, Kito wrote:
>
>>
>> On Aug 8, 2005, at 6:36 PM, Hasan Khalil wrote:
>>
>>
>>> We would like to hold an election for the positions that we have
>>> been more-or-less filling unofficially, as the new metastructure
>>> suggests. We would like to be candidates for Operation Lead (Lina)
>>> and Strategic Lead (Hasan).
>>>
>>
>> I was wondering how long it was going to take you guys....
>>
>> Who gets to vote in this election?
>>
>> Shouldn't the entire Gentoo dev team have some input?
>>
>> Shouldn't this wait until after the council elections are finished?
>>
>> What will change once you two have been elected?
>>
>> You should also probably address the fact that there is still a
>> strategic lead in place (pvdabeel) before deciding amongst
>> yourselves to hold an election...
>>
>> I know titles can be fun and addictive, but I'm not sure declaring an
>> election to lead a team of 4 people, when 2 of them are basically
>> jointly running, is the Right Thing(tm) at this point in time... yes
>> the project needs leadership but this smells a little funny.
>>
>> --Kito
>>
>
> --
> gentoo-osx [at] gentoo mailing list
>
>

--
gentoo-osx [at] gentoo mailing list


macmonster at myrealbox

Aug 15, 2005, 11:14 PM

Post #11 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

Sorry to come to this so late - I'm just back from holiday.

On Aug 8, 2005, at 6:38 am, Kito wrote:
>
> Maybe you misunderstood, what I think is futile is trying to avoid
> overwriting files, and accommodating things portage has no knowledge
> of or control over.

Unless you avoid over-writing Mac OS's system files Gentoo-OSX will
never become mainstream. I don't know if this is of interest to you or
not, but I for one will never install Gentoo-OSX unless it installs
safely into an /opt (or similar) directory - overwriting Apple's system
files is _bound_ to break something eventually.

You say that accommodating Appe's system is accommodating files portage
has no knowledge of or control over, but at least you have control over
the whole portage tree.

Imagine I spend several ££$£$$£$£$£ on Mathematica, and have a problem
with it; I contact Wolfram support and they say, "Do you have
Gentoo-OSX installed? Sorry, we can't support you - you might have
over-written any of Apple's system files, so we need to you format your
hard-drive & reinstall OS X before submitting this bug".

There is no way that Gentoo can resolve a situation like this to the be
benefit of the user. If, on the other hand, apps-foo/bar-player depends
upon Apple's tar to unpack it, and Apple break that, then you're in a
position to change the ebuild and have a fix rolled-out for all users
next time they sync the tree.

I haven't ever used Portage on OS X, and I don't develop for
Gentoo-Linux (just submit the occasional bug report and do testing),
but I've been using Gentoo for a couple of years now and have been
interested in this project from the start - it was my G4 that Dan
Robbins shelled into to first try porting Portage in June 2003. It
seems to me that an understanding that over-writing Apple's system
files MUST affect system stability, and is hence something that the
majority of users should avoid, would go a long way to getting this
project taken more seriously by devs of other arches.

Stroller.


--
gentoo-osx [at] gentoo mailing list


fthain at telegraphics

Aug 16, 2005, 12:43 AM

Post #12 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Tue, 16 Aug 2005, Stroller wrote:

>
> Sorry to come to this so late - I'm just back from holiday.
>
> On Aug 8, 2005, at 6:38 am, Kito wrote:
> >
> > Maybe you misunderstood, what I think is futile is trying to avoid
> > overwriting files, and accommodating things portage has no knowledge of or
> > control over.
>
> Unless you avoid over-writing Mac OS's system files Gentoo-OSX will
> never become mainstream.

You should take the context into consideration when trying to understand
Kito's comment, which was in reply to my post about Gentoo/Darwin. In that
light, he is absolutely right, collision-protect is counter-productive.

In the context of Gentoo/OS X, I think everyone agrees that prefixed
installs would be preferable to collision-protect.

If you install Gentoo/OS X, you will find collision-protect is used by the
default profile. You concerns seem to be aimed at the progressive profile
that some users prefer. It is your choice.

-f
--
gentoo-osx [at] gentoo mailing list


kito at gentoo

Aug 16, 2005, 9:19 AM

Post #13 of 15 (1261 views)
Permalink
Re: Some Introduction [In reply to]

On Aug 16, 2005, at 1:14 AM, Stroller wrote:

>
> Sorry to come to this so late - I'm just back from holiday.
>
> On Aug 8, 2005, at 6:38 am, Kito wrote:
>>
>> Maybe you misunderstood, what I think is futile is trying to avoid
>> overwriting files, and accommodating things portage has no
>> knowledge of or control over.
>
> Unless you avoid over-writing Mac OS's system files Gentoo-OSX will
> never become mainstream. I don't know if this is of interest to you
> or not, but I for one will never install Gentoo-OSX unless it
> installs safely into an /opt (or similar) directory - overwriting
> Apple's system files is _bound_ to break something eventually.

Yeah, again I'm not saying overwriting files is the way to go. I'm
not sure how much more clear I can be on that. I'm saying installing
things to / and trying to avoid overwriting files is not a good idea.

So, again, just to make sure I'm being understood, installing to an
alternate prefix, i.e. /gentoo, /usr/local, /opt, whatever, is the
way to go and make portage a viable tool for OS X users. As Finn
pointed out in earlier mails, not using a prefix like Fink and
DarwinPorts was never a strategic decision, but a practical one. Its
all that portage knew how to do.

Now, as far as portage having knowledge about what software is
installed...if it lived in its own prefix, that wouldn't be an issue,
as all package dependencies would be handled and installed by
portage, avoiding linking against OS X system libs as much as
possible. This would make it possible for instance to have /gentoo on
say an external drive that could be mounted and used on virtually any
Mac OSX system, regardless of system updates, etc.

That being said, I feel like the name for such a tool shouldn't be
called 'Gentoo for OS X', as thats a very misleading name which seems
to cause a lot of confusion amongst users and developers alike. A
much more apt title IMHO would be 'Portage for Mac OS X', and leave
the title 'Gentoo for Mac OS X' for the profiles that actually manage
system files.

--Kito

> Stroller.
>
>
> --
> gentoo-osx [at] gentoo mailing list
>

--
gentoo-osx [at] gentoo mailing list


grobian at gentoo

Aug 17, 2005, 6:33 AM

Post #14 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

Kito wrote:
> Yeah, again I'm not saying overwriting files is the way to go. I'm not
> sure how much more clear I can be on that. I'm saying installing things
> to / and trying to avoid overwriting files is not a good idea.
>
> So, again, just to make sure I'm being understood, installing to an
> alternate prefix, i.e. /gentoo, /usr/local, /opt, whatever, is the way
> to go and make portage a viable tool for OS X users. As Finn pointed out
> in earlier mails, not using a prefix like Fink and DarwinPorts was never
> a strategic decision, but a practical one. Its all that portage knew how
> to do.
>
> Now, as far as portage having knowledge about what software is
> installed...if it lived in its own prefix, that wouldn't be an issue, as
> all package dependencies would be handled and installed by portage,
> avoiding linking against OS X system libs as much as possible. This
> would make it possible for instance to have /gentoo on say an external
> drive that could be mounted and used on virtually any Mac OSX system,
> regardless of system updates, etc.
>
> That being said, I feel like the name for such a tool shouldn't be
> called 'Gentoo for OS X', as thats a very misleading name which seems to
> cause a lot of confusion amongst users and developers alike. A much more
> apt title IMHO would be 'Portage for Mac OS X', and leave the title
> 'Gentoo for Mac OS X' for the profiles that actually manage system files.

Amen! Couldn't agree more.

--
Fabian Groffen
eBuild && Porting
Gentoo for Mac OS X
--
gentoo-osx [at] gentoo mailing list


fthain at telegraphics

Aug 20, 2005, 3:51 AM

Post #15 of 15 (1260 views)
Permalink
Re: Some Introduction [In reply to]

On Wed, 17 Aug 2005, Grobian wrote:

>
> Kito wrote:
>
[snip]
> >
> > Now, as far as portage having knowledge about what software is
> > installed...if it lived in its own prefix, that wouldn't be an issue,
> > as all package dependencies would be handled and installed by portage,
> > avoiding linking against OS X system libs as much as possible. This
> > would make it possible for instance to have /gentoo on say an external
> > drive that could be mounted and used on virtually any Mac OSX system,
> > regardless of system updates, etc.
> >
> > That being said, I feel like the name for such a tool shouldn't be
> > called 'Gentoo for OS X', as thats a very misleading name which seems
> > to cause a lot of confusion amongst users and developers alike. A much
> > more apt title IMHO would be 'Portage for Mac OS X', and leave the
> > title 'Gentoo for Mac OS X' for the profiles that actually manage
> > system files.
>
> Amen! Couldn't agree more.
>

If you will humour me, what's the difference? Apart from prefix, what
distinguishes profiles that manage the host OS and profiles that don't?
(Given that a profile can change just about every aspect of portage's
behaviour.)

The reason I ask is this: greedily, I want Kito's /gentoo on an external
drive, and I also want to chroot into it sometimes [1].

The implication being, 'Portage for Mac OS X' would have to be equivalent
to 'Gentoo for Mac OS X'.

This idea extrapolates from the original pathspec, which described a
self-similar directory structure. (I guess the idea was that the ebuilds
shouldn't have to care.)

Anyway, as someone who doesn't cut any code for this project, I think I
should stop making outrageous demands now ;-)

-f



[1] This can be done (I learned this trick from Ryan Oliver's Pure LFS
technique). You install everything prefixed /Volumes/Gentoo, then

mkdir /Volumes/Gentoo/Volumes
ln -s ../.. /Volumes/Gentoo/Volumes/Gentoo
chroot /Volumes/Gentoo

Outside the chroot, /etc/make.profile defines the prefix as
/Volumes/Gentoo, whilst inside the chroot, /etc/make.profile uses the same
prefix. The chroot profile may need to be different, if portage is
expecting Mac OS X here... hence my initial question, "What's the
difference?"

I'm assuming that any required OS X libraries have been copied to the
Gentoo volume first, and the psuedo filesystems mounted (as with any
chroot).
--
gentoo-osx [at] gentoo mailing list

Gentoo osx RSS feed   Index | Next | Previous | View Threaded
 
 


Interested in having your list archived? Contact Gossamer Threads
 
  Web Applications & Managed Hosting Powered by Gossamer Threads Inc.