[gentoo-dev] Decision to remove stage1/2 from installation documentation

[gentoo-dev] Decision to remove stage1/2 from installation documentation

[Kurt Lieber]

[-- Attachment #1: Type: text/plain, Size: 1089 bytes --]

We have received *numerous* complaints from users about the decision to
remove stage 1 and 2 from the installation documentation.  I realize it's
still available if users are willing to dig for it, but not all users do.

In my years of monitoring www@gentoo.org, we've received the most
complaints about this decision than any other single decision.  Is there a
way we can re-introduce the stages into the installation documentation,
perhaps with gigantic warnings saying, "for advanced users only" or "use at
your own risk"?

--kurt

----- Forwarded message from Varun Dhussa <varun.dhussa@gmail.com> -----

Date: Tue, 22 Nov 2005 19:50:07 +0530
From: Varun Dhussa <varun.dhussa@gmail.com>
To: www@gentoo.org
Subject: Complaint

Hello,

Gentoo claims to be giving freedom. However, I was dissapointed to see that
the stage 1 had been removed from gentoo 2005. Infact, even the handbook
makes no refference of it. This takes Gentoo another step closer to other
distros like Ubuntu.

A dissapointed user,
Varun Dhussa
India

----- End forwarded message -----

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Andrea Barisani]

On Tue, Nov 22, 2005 at 02:47:45PM +0000, Kurt Lieber wrote:
> We have received *numerous* complaints from users about the decision to
> remove stage 1 and 2 from the installation documentation.  I realize it's
> still available if users are willing to dig for it, but not all users do.
> 
> In my years of monitoring www@gentoo.org, we've received the most
> complaints about this decision than any other single decision.  Is there a
> way we can re-introduce the stages into the installation documentation,
> perhaps with gigantic warnings saying, "for advanced users only" or "use at
> your own risk"?
> 
> --kurt
> 

I perfectly agree with this request, we should provide the choice and clear
point that out (along with all the correlated risks) instead of simply
"hiding" the option. And I sincerely hope there's no intention to remove
stage1/stage2 tarballs in the future because that would be a really a bad thing
imho.

Cheers

-- 
Andrea Barisani <lcars@gentoo.org>                            .*.
Gentoo Linux Infrastructure Developer                          V
                                                             (   )
PGP-Key 0x864C9B9E http://dev.gentoo.org/~lcars/pubkey.asc   (   )
    0A76 074A 02CD E989 CE7F AC3F DA47 578E 864C 9B9E        ^^_^^
      "Pluralitas non est ponenda sine necessitate"
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1671 bytes --]

On Tue, 2005-11-22 at 15:37 +0100, Andrea Barisani wrote:
> On Tue, Nov 22, 2005 at 02:47:45PM +0000, Kurt Lieber wrote:
> > We have received *numerous* complaints from users about the decision to
> > remove stage 1 and 2 from the installation documentation.  I realize it's
> > still available if users are willing to dig for it, but not all users do.
> > 
> > In my years of monitoring www@gentoo.org, we've received the most
> > complaints about this decision than any other single decision.  Is there a
> > way we can re-introduce the stages into the installation documentation,
> > perhaps with gigantic warnings saying, "for advanced users only" or "use at
> > your own risk"?
> > 
> > --kurt
> > 
> 
> I perfectly agree with this request, we should provide the choice and clear
> point that out (along with all the correlated risks) instead of simply
> "hiding" the option. And I sincerely hope there's no intention to remove
> stage1/stage2 tarballs in the future because that would be a really a bad thing
> imho.

The problem with listing risks and such is the users aren't listening.

They are ignoring our warnings and breaking their own systems, then
filing bugs.  The problem is that these are *not* bugs, but issues with
incompatibility.  It is impossible to install something that requires a
configured kernel before you have a configured kernel.

Now, on the topic of the tarballs.

Give me one example of something that you can do with a stage1 or stage2
tarball that you cannot with a stage3 tarball.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Marc Hildebrand]

Chris Gianelloni wrote:
[..]
> Now, on the topic of the tarballs.
> 
> Give me one example of something that you can do with a stage1 or stage2
> tarball that you cannot with a stage3 tarball.
> 

Answer: Download it in less than 10 minutes.


The question of interest is: Will we keep changing things without a GLEP 
that should *never* be touched without one?

Cheers,

Marc.

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 820 bytes --]

On Tue, 2005-11-22 at 16:26 +0100, Marc Hildebrand wrote:
> Chris Gianelloni wrote:
> [..]
> > Now, on the topic of the tarballs.
> > 
> > Give me one example of something that you can do with a stage1 or stage2
> > tarball that you cannot with a stage3 tarball.
> > 
> 
> Answer: Download it in less than 10 minutes.

I'd love to see you do the same with a stage1 tarball + all the
distfiles you'll need to go from stage1 to stage3.

In case you're wondering, it's more than the size of a stage3 tarball,
by quite a bit.

> The question of interest is: Will we keep changing things without a GLEP 
> that should *never* be touched without one?

Since when is this GLEP material?

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Wernfried Haas]

On Tue, Nov 22, 2005 at 10:48:06AM -0500, Chris Gianelloni wrote:
> > > Give me one example of something that you can do with a stage1 or stage2
> > > tarball that you cannot with a stage3 tarball.

> > Answer: Download it in less than 10 minutes.

> I'd love to see you do the same with a stage1 tarball + all the
> distfiles you'll need to go from stage1 to stage3.

Assuming you keep all distfiles you already downloaded (which some
people do, like me). So you'd just need stage 1, nothing else.
Maybe that does not justify keeping stage1 (imho not even though 
it's useful for me), but it _is_ answering your intial question. ;-)

Btw, if i use stage 3 and then emerge -e world to recompile my whole
system with -omg-optimized i assume stage 3 may lose against stage 1
and compiling -omg-optimized from the beginning. Not that it makes
much sense to do that though.

cheers,
	Wernfried

-- 
Wernfried Haas (amne) - amne at gentoo dot org
Gentoo Forums: http://forums.gentoo.org
IRC: #gentoo-forums on freenode - email: forum-mods at gentoo dot org
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1406 bytes --]

On Tue, 2005-11-22 at 17:15 +0100, Wernfried Haas wrote:
> On Tue, Nov 22, 2005 at 10:48:06AM -0500, Chris Gianelloni wrote:
> > > > Give me one example of something that you can do with a stage1 or stage2
> > > > tarball that you cannot with a stage3 tarball.
> 
> > > Answer: Download it in less than 10 minutes.
> 
> > I'd love to see you do the same with a stage1 tarball + all the
> > distfiles you'll need to go from stage1 to stage3.
> 
> Assuming you keep all distfiles you already downloaded (which some
> people do, like me). So you'd just need stage 1, nothing else.
> Maybe that does not justify keeping stage1 (imho not even though 
> it's useful for me), but it _is_ answering your intial question. ;-)

Just because you downloaded them previously does not mean you didn't
download them.

> Btw, if i use stage 3 and then emerge -e world to recompile my whole
> system with -omg-optimized i assume stage 3 may lose against stage 1
> and compiling -omg-optimized from the beginning. Not that it makes
> much sense to do that though.

*sigh*

You have proven my point.  Thank you.  If you compile the same sources
with the same settings, you get the same output.  It doesn't matter if
you started from a stage1, stage2, stage3, or stage4 tarball.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Wernfried Haas]

On Tue, Nov 22, 2005 at 11:33:04AM -0500, Chris Gianelloni wrote:
> Just because you downloaded them previously does not mean you didn't
> download them.
Yes, but i already have them and don't need to download them any
more in this scenario.

> > Btw, if i use stage 3 and then emerge -e world to recompile my whole
> > system with -omg-optimized i assume stage 3 may lose against stage 1
> > and compiling -omg-optimized from the beginning. Not that it makes
> > much sense to do that though.
> 
> *sigh*
> 
> You have proven my point.  Thank you.  
No i haven't, but you're still welcome. Maybe i was not clear enough
about the definition of losing (see below).

> If you compile the same sources
> with the same settings, you get the same output.  It doesn't matter if
> you started from a stage1, stage2, stage3, or stage4 tarball.
Sorry, I wasn't really refering to the output but the bytes used for
downloading the stage and the distfiles.
Assuming emerge -e world compiles exactly the same packages (same
versions as well) as stage 1 would mean you're a bit better off with 
stage 1 because it's slightly smaller. Hence stage 3 may lose by a few 
bytes here. 

Look, i'm not arguing with you and even though i think stage 1 has
some cool features here i agree removing it may be a good idea. I just
wanted to point out there are _some_ specific situations where you're
better off with stage 1 under certain circumstances because you asked
if there's anything stage 1 can do that 3 doesn't. ;-)

cheers,
	Wernfried

-- 
Wernfried Haas (amne) - amne at gentoo dot org
Gentoo Forums: http://forums.gentoo.org
IRC: #gentoo-forums on freenode - email: forum-mods at gentoo dot org
-- 
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[R Hill]

Chris Gianelloni wrote:
> On Tue, 2005-11-22 at 16:26 +0100, Marc Hildebrand wrote:
>> Chris Gianelloni wrote:
>> [..]
>>> Now, on the topic of the tarballs.
>>>
>>> Give me one example of something that you can do with a stage1 or stage2
>>> tarball that you cannot with a stage3 tarball.
>>>
>> Answer: Download it in less than 10 minutes.
> 
> I'd love to see you do the same with a stage1 tarball + all the
> distfiles you'll need to go from stage1 to stage3.

What about someone on dialup who needs a rescue CD to boot into their system
after they've trashed the MBR?  88MiB vs 14MiB is a big difference in this case.

> In case you're wondering, it's more than the size of a stage3 tarball,
> by quite a bit.
> 
>> The question of interest is: Will we keep changing things without a GLEP 
>> that should *never* be touched without one?
> 
> Since when is this GLEP material?

Are you kidding?  Since it's a fundamentally significant and highly visible
change in the workings of Gentoo.  The three-stage build system is one of the
distinguishing characteristics of Gentoo, up there with source-based, install
from scratch, and highly customizable.  Every review of Gentoo I've ever seen at
least mentions it.

For the record, I don't think it matters if stage 1 goes away.  Make stage 3 the
Official and Supported Way of installing Gentoo, but provide stage 1 as a
minimal LiveCD/RescueCD option.  Make a mention in the install documentation
along the lines of

"It is also possible to do a full install of Gentoo using a minimal Rescue
LiveCD and a network connection.  This method is depreciated and should only be
used if circumstances prevent you using the Universal LiveCD.  Note that we do
_not_ provide support for systems built using minimal installations, so you're
on your own."

(linkity to a new separate stage 1 doc page)



--de.

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Dan Meltzer]

On 11/22/05, R Hill <dirtyepic.sk@gmail.com> wrote:
> Chris Gianelloni wrote:
> > On Tue, 2005-11-22 at 16:26 +0100, Marc Hildebrand wrote:
> >> Chris Gianelloni wrote:
> >> [..]
> >>> Now, on the topic of the tarballs.
> >>>
> >>> Give me one example of something that you can do with a stage1 or stage2
> >>> tarball that you cannot with a stage3 tarball.
> >>>
> >> Answer: Download it in less than 10 minutes.
> >
> > I'd love to see you do the same with a stage1 tarball + all the
> > distfiles you'll need to go from stage1 to stage3.
>
> What about someone on dialup who needs a rescue CD to boot into their system
> after they've trashed the MBR?  88MiB vs 14MiB is a big difference in this case.
Erm, why would I need a stage 1 for a rescue cd?
>
> > In case you're wondering, it's more than the size of a stage3 tarball,
> > by quite a bit.
> >
> >> The question of interest is: Will we keep changing things without a GLEP
> >> that should *never* be touched without one?
> >
> > Since when is this GLEP material?
>
> Are you kidding?  Since it's a fundamentally significant and highly visible
> change in the workings of Gentoo.  The three-stage build system is one of the
> distinguishing characteristics of Gentoo, up there with source-based, install
> from scratch, and highly customizable.  Every review of Gentoo I've ever seen at
> least mentions it.

It removes no functionality, it adds no functionality.  It simply
changes it.  How is this GLEP materiel?
>
> For the record, I don't think it matters if stage 1 goes away.  Make stage 3 the
> Official and Supported Way of installing Gentoo, but provide stage 1 as a
> minimal LiveCD/RescueCD option.  Make a mention in the install documentation
> along the lines of
>
> "It is also possible to do a full install of Gentoo using a minimal Rescue
> LiveCD and a network connection.  This method is depreciated and should only be
> used if circumstances prevent you using the Universal LiveCD.  Note that we do
> _not_ provide support for systems built using minimal installations, so you're
> on your own."
>
> (linkity to a new separate stage 1 doc page)
>
>
>
> --de.
>
> --
> gentoo-dev@gentoo.org mailing list
>
>

-- 
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[R Hill]

Dan Meltzer wrote:
> On 11/22/05, R Hill <dirtyepic.sk@gmail.com> wrote:
>> What about someone on dialup who needs a rescue CD to boot into their system
>> after they've trashed the MBR?  88MiB vs 14MiB is a big difference in this case.

> Erm, why would I need a stage 1 for a rescue cd?

christ, i must be tired.  for some reason i equated stage 1 with the minimal
install cd.  it's been too long since i did an install.  sorry.

the only place i could see a need for stage 1 instructions would be doing a
networkless install on the minimal cd, which isn't possible anyways.


> It removes no functionality, it adds no functionality.  It simply
> changes it.  How is this GLEP materiel?


it's not.

it's obvious that a lot of people (including me) are confused about exactly what
the deal is here.  this is why things need to be made clear before the fact, not
after they happen, and need to be discussed with both devs and users to make
sure everyone's on the same page, instead of spending all day violently
misunderstanding each other.

--de.

-- 
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Duncan]

R Hill posted <dm11g6$m84$1@sea.gmane.org>, excerpted below,  on Wed, 23
Nov 2005 00:16:38 -0600:

> this is why things need to be made clear
> before the fact, not after they happen, and need to be discussed with both
> devs and users to make sure everyone's on the same page, instead of
> spending all day violently misunderstanding each other.

..  All day violently missunderstanding each other...  LOL, but
unfortunately what has seemed to be the case, both with this, the stage-1
thing, and with the mail subdomain issue regarding GLEP 41, the
arch-tester thing.

Certainly lessons for the future.

-- 
Duncan - List replies preferred.   No HTML msgs.
"Every nonfree program has a lord, a master --
and if you use the program, he is your master."  Richard Stallman in
http://www.linuxdevcenter.com/pub/a/linux/2004/12/22/rms_interview.html


-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Abhay Kedia]

[-- Attachment #1: Type: text/plain, Size: 524 bytes --]

On Wednesday 23 Nov 2005 10:28 am, R Hill wrote:
>
> For the record, I don't think it matters if stage 1 goes away.  Make stage
> 3 the Official and Supported Way of installing Gentoo, but provide stage 1
> as a minimal LiveCD/RescueCD option.  Make a mention in the install
> documentation along the lines of
>
He has already said that if Gentoo is releasing something then they have a 
moral onus of supporting it. He doesn't want to release something officially 
and then say that it is not supported.

Abhay

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[solar]

> Now, on the topic of the tarballs.
> 
> Give me one example of something that you can do with a stage1 or stage2
> tarball that you cannot with a stage3 tarball.

Stage1: Changing CHOST= and run ./bootstrap.sh 
(well you can do it but it's dumb)

Stage3: has full cxx/berkdb/ssl/pam/libwrap and all the cruft pulled in
from having use flags enabled thats not easy to get rid of otherwise.

I don't care what you do with the docs, but the stages 1, 3 need to 
stay. stage2 has always been a bonus stage more or less added into the 
mix cuz it's a byproduct of stage building (pre catalyst days).

-- 
solar <solar@gentoo.org>
Gentoo Linux

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Stephen P. Becker]

> Stage1: Changing CHOST= and run ./bootstrap.sh 
> (well you can do it but it's dumb)

You can do the same from a stage3.


> Stage3: has full cxx/berkdb/ssl/pam/libwrap and all the cruft pulled in
> from having use flags enabled thats not easy to get rid of otherwise.

Fair point, however this is the kind of stuff that most users want 
anyway.  I see the embedded angle you are taking with this argument, and 
I would counter by saying that folks who are interested in some sort of 
embedded uclibc type userland probably know what they are doing in the 
first place.


> I don't care what you do with the docs, but the stages 1, 3 need to 
> stay. stage2 has always been a bonus stage more or less added into the 
> mix cuz it's a byproduct of stage building (pre catalyst days).

I don't think anyone has implied that we're not going to distribute 
stage1 anymore.  They are still useful for folks that know what they are 
doing.

-Steve
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Mike Frysinger]

On Tue, Nov 22, 2005 at 10:38:34AM -0500, Stephen P. Becker wrote:
> >I don't care what you do with the docs, but the stages 1, 3 need to 
> >stay. stage2 has always been a bonus stage more or less added into the 
> >mix cuz it's a byproduct of stage building (pre catalyst days).
> 
> I don't think anyone has implied that we're not going to distribute 
> stage1 anymore.  They are still useful for folks that know what they are 
> doing.

i was under the impression we were merely changing our docs and that
we were going to continue to mirror stage1 and stage2 files

this i am OK with
-mike
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1335 bytes --]

On Tue, 2005-11-22 at 10:29 -0500, solar wrote:
> > Now, on the topic of the tarballs.
> > 
> > Give me one example of something that you can do with a stage1 or stage2
> > tarball that you cannot with a stage3 tarball.
> 
> Stage1: Changing CHOST= and run ./bootstrap.sh 
> (well you can do it but it's dumb)

Right.  You can do it.

> Stage3: has full cxx/berkdb/ssl/pam/libwrap and all the cruft pulled in
> from having use flags enabled thats not easy to get rid of otherwise.

You can accomplish this, too.  Maybe we could even use that nice little
"scripts" directory in the portage tree to write a script to assist in
performing this.  I'm sure it would be less error-prone than what we
have now with the broken QA procedures allowing things to go into
system, or system-depended packages, that pulls in all kinds of useless
crap.

> I don't care what you do with the docs, but the stages 1, 3 need to 
> stay. stage2 has always been a bonus stage more or less added into the 
> mix cuz it's a byproduct of stage building (pre catalyst days).

Removing the stage1 and stage2 instructions from the Handbook has
already reduced the number of errors being reported by new users to me.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[solar]

On Tue, 2005-11-22 at 10:58 -0500, Chris Gianelloni wrote:
> On Tue, 2005-11-22 at 10:29 -0500, solar wrote:
> > > Now, on the topic of the tarballs.
> > > 
> > > Give me one example of something that you can do with a stage1 or stage2
> > > tarball that you cannot with a stage3 tarball.
> > 
> > Stage1: Changing CHOST= and run ./bootstrap.sh 
> > (well you can do it but it's dumb)
> 
> Right.  You can do it.
> 
> > Stage3: has full cxx/berkdb/ssl/pam/libwrap and all the cruft pulled in
> > from having use flags enabled thats not easy to get rid of otherwise.
> 
> You can accomplish this, too.  Maybe we could even use that nice little
> "scripts" directory in the portage tree to write a script to assist in
> performing this.  I'm sure it would be less error-prone than what we
> have now with the broken QA procedures allowing things to go into
> system, or system-depended packages, that pulls in all kinds of useless
> crap.
> 
> > I don't care what you do with the docs, but the stages 1, 3 need to 
> > stay. stage2 has always been a bonus stage more or less added into the 
> > mix cuz it's a byproduct of stage building (pre catalyst days).
> 
> Removing the stage1 and stage2 instructions from the Handbook has
> already reduced the number of errors being reported by new users to me.

While I'm glad your getting less bugs/errors reported, I'm a little 
sad that you had stage1 install instructions removed. 
Starting from stage1 is the basic hardened way. Most of our users 
(which seem to be 10% of the userbase) start from that stage due to 
hardened only shipping it's stages as i386-pc-linux-gnu and honestly 
most users are i686-pc-linux-gnu so they end up changing this. 
We ship as i386 to ensure all x86 compatibility.

I hope that you can arrange for either the stage1 install instructions
to be put back in or split off into it's own stage1.xml doc.


-- 
solar <solar@gentoo.org>
Gentoo Linux

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 574 bytes --]

On Tue, 2005-11-22 at 11:25 -0500, solar wrote:
> > Removing the stage1 and stage2 instructions from the Handbook has
> > already reduced the number of errors being reported by new users to me.

> I hope that you can arrange for either the stage1 install instructions
> to be put back in or split off into it's own stage1.xml doc.

I've already agreed that I have no problem with an additional document
that instructs on these advanced concepts.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Andrea Barisani]

On Tue, Nov 22, 2005 at 10:14:04AM -0500, Chris Gianelloni wrote:
> On Tue, 2005-11-22 at 15:37 +0100, Andrea Barisani wrote:
> > On Tue, Nov 22, 2005 at 02:47:45PM +0000, Kurt Lieber wrote:
> > > We have received *numerous* complaints from users about the decision to
> > > remove stage 1 and 2 from the installation documentation.  I realize it's
> > > still available if users are willing to dig for it, but not all users do.
> > > 
> > > In my years of monitoring www@gentoo.org, we've received the most
> > > complaints about this decision than any other single decision.  Is there a
> > > way we can re-introduce the stages into the installation documentation,
> > > perhaps with gigantic warnings saying, "for advanced users only" or "use at
> > > your own risk"?
> > > 
> > > --kurt
> > > 
> > 
> > I perfectly agree with this request, we should provide the choice and clear
> > point that out (along with all the correlated risks) instead of simply
> > "hiding" the option. And I sincerely hope there's no intention to remove
> > stage1/stage2 tarballs in the future because that would be a really a bad thing
> > imho.
> 
> The problem with listing risks and such is the users aren't listening.
> 
> They are ignoring our warnings and breaking their own systems, then
> filing bugs.  The problem is that these are *not* bugs, but issues with
> incompatibility.  It is impossible to install something that requires a
> configured kernel before you have a configured kernel.
>

I still think that pointing things with a *huge* warning shouldn't be
a problem...otherwise we would always end up "hiding" things prone to user
error because we think that users are listening. At least let's draft a nice
and visible document explaining the change and why people should not use this
anymore since judging from the complaints lots of people just don't get it.

> Now, on the topic of the tarballs.
> 
> Give me one example of something that you can do with a stage1 or stage2
> tarball that you cannot with a stage3 tarball.
>

Oh well nothing. I don't doubt that userwise they are not needed...but there
might be other needs developerwise where the two stages are useful.

So fair enough, remove it from the docs...but at least let's explain why we
are doing this since complaints are there (legit or not).

-- 
Andrea Barisani <lcars@gentoo.org>                            .*.
Gentoo Linux Infrastructure Developer                          V
                                                             (   )
PGP-Key 0x864C9B9E http://dev.gentoo.org/~lcars/pubkey.asc   (   )
    0A76 074A 02CD E989 CE7F AC3F DA47 578E 864C 9B9E        ^^_^^
      "Pluralitas non est ponenda sine necessitate"
-- 
gentoo-dev@gentoo.org mailing list

[gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[R Hill]

Andrea Barisani wrote:

> At least let's draft a nice
> and visible document explaining the change and why people should not use this
> anymore since judging from the complaints lots of people just don't get it.

I think that a lot of people use stage 1 because they're under the impression
that they have to in order to change their CHOST on default-linux from
i386-pc-linux-gnu to i686-pc-linux-gnu.  And unless something has changed
recently, to get an NPTL glibc they _do_ have to make that change [1].  If it's
made clear that there is absolutely no difference in the end btwn stage3 +
emerge -e world and stage1, maybe people would be less enthusiastic about using it.


--de.

[1] http://bugs.gentoo.org/show_bug.cgi?id=106556

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Dale]

R Hill wrote:

>I think that a lot of people use stage 1 because they're under the impression
>that they have to in order to change their CHOST on default-linux from
>i386-pc-linux-gnu to i686-pc-linux-gnu.  And unless something has changed
>recently, to get an NPTL glibc they _do_ have to make that change [1].  If it's
>made clear that there is absolutely no difference in the end btwn stage3 +
>emerge -e world and stage1, maybe people would be less enthusiastic about using it.
>
>
>--de.
>
>  
>
Well, I'm not expert for sure, quite a noobie really, but I did my
install about 2 years ago.  I did a stage 3 install and then did a
emerge -ev world.  I must admit though, I did actually read the manual
several times before I started my install.  I usually don't read the
directions but hey, I didn't know crap about Gentoo, don't know much
crap now either.  :/

Some may need the option for a stage 1 but unless you are on one slow
rig, emerge -ev world shortly after booting your new stable kernel
shouldn't take to long then install everything else.  Wouldn't that make
everyone happy?  Shouldn't that also give some devs some time to work on
others things as well?  Getting a sync past 50% with slowing to a crawl
would be nice.  Plenty of people complaining about that.

Don't beat me to much OK.  Be gentle.

Dale
:-)

-- 
To err is human, I'm most certainly human.

 

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Brian Harring]

[-- Attachment #1: Type: text/plain, Size: 380 bytes --]

On Tue, Nov 22, 2005 at 11:36:01PM -0600, Dale wrote:
> R Hill wrote:
> Getting a sync past 50% with slowing to a crawl
> would be nice.  Plenty of people complaining about that.
> 
> Don't beat me to much OK.  Be gentle.
Really need to release a portage that contains 
http://dev.gentoo.org/~ferringb/blog/archives/2005-10.html#e2005-10-12T23_59_53.txt 
...

~harring

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Tuan Van]

R Hill wrote:

> 
> I think that a lot of people use stage 1 because they're under the impression
> that they have to in order to change their CHOST on default-linux from
> i386-pc-linux-gnu to i686-pc-linux-gnu.  And unless something has changed
> recently, to get an NPTL glibc they _do_ have to make that change [1].  If it's
> made clear that there is absolutely no difference in the end btwn stage3 +
> emerge -e world and stage1, maybe people would be less enthusiastic about using it.
> 
if you are on x86, there is stage3 for pentium4 built with
CFLAGS="-O2 -march=penitum4"
CHOST="i686-pc-linux-gnu"
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Mike Owen]

On 11/22/05, Chris Gianelloni <wolf31o2@gentoo.org> wrote:
> Give me one example of something that you can do with a stage1 or stage2
> tarball that you cannot with a stage3 tarball.
>

I may not be the typical user, but I use Stage1 to build servers,
because I can fit a boot image + stage1 tarball on a small usb drive,
boot to that, and then I nfs mount $DISTDIR and $PORTDIR from a
central server.

Mike

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Dan Meltzer]

On 11/23/05, Mike Owen <kyphros@gmail.com> wrote:
> On 11/22/05, Chris Gianelloni <wolf31o2@gentoo.org> wrote:
> > Give me one example of something that you can do with a stage1 or stage2
> > tarball that you cannot with a stage3 tarball.
> >
>
> I may not be the typical user, but I use Stage1 to build servers,
> because I can fit a boot image + stage1 tarball on a small usb drive,
> boot to that, and then I nfs mount $DISTDIR and $PORTDIR from a
> central server.
Correct me if I am wrong, but doesn't the boot image itself have nfs
built in? why a stage1 at all...
>
> Mike
>
> --
> gentoo-dev@gentoo.org mailing list
>
>

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Mike Owen]

On 11/23/05, Dan Meltzer <parallelgrapefruit@gmail.com> wrote:
> On 11/23/05, Mike Owen <kyphros@gmail.com> wrote:
> > I may not be the typical user, but I use Stage1 to build servers,
> > because I can fit a boot image + stage1 tarball on a small usb drive,
> > boot to that, and then I nfs mount $DISTDIR and $PORTDIR from a
> > central server.
> Correct me if I am wrong, but doesn't the boot image itself have nfs
> built in? why a stage1 at all...
> >

Because I'm lazy, and used to doing the stage1 thing :)

Mike

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Dan Meltzer]

On 11/23/05, Mike Owen <kyphros@gmail.com> wrote:
> On 11/23/05, Dan Meltzer <parallelgrapefruit@gmail.com> wrote:
> > On 11/23/05, Mike Owen <kyphros@gmail.com> wrote:
> > > I may not be the typical user, but I use Stage1 to build servers,
> > > because I can fit a boot image + stage1 tarball on a small usb drive,
> > > boot to that, and then I nfs mount $DISTDIR and $PORTDIR from a
> > > central server.
> > Correct me if I am wrong, but doesn't the boot image itself have nfs
> > built in? why a stage1 at all...
> > >
>
> Because I'm lazy, and used to doing the stage1 thing :)

So, to summarize, you don't need stage1 either.
>
> Mike
>
> --
> gentoo-dev@gentoo.org mailing list
>
>

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Bruno]

On Tuesday 22 November 2005 16:14, Chris Gianelloni wrote:
> Give me one example of something that you can do with a stage1 or stage2
> tarball that you cannot with a stage3 tarball.

It's useful if you have to change compiler or other tool-chain part right from 
the start (e.g. use 3.4.* on i386, where 3.3.* is default) on PentiumM in 
order to use -march=pentium-m.

It's certainly possible to start with stage 3, but makes total process last 
longer (Much more to recompile) and is more error-prone.

Example of this risk:
When installing GCC3.4 one may forget to install old libstdc++ (it has to be 
unmasked, and depending on use-flags it me not yes be reauested by portage!) 
and have a missing linking dependency on libstdc++ in python (no more portage 
to recompile python!) once GCC3.3 is unmerged.


For some server-setups it may also be useful to start from a very minimal base 
in order to avoid hidden dependencies caused by unconditionnal operations of 
configure which add unwanted dependencies (e.g. USE-flags disables dep, but 
configure script still uses it, be it directly or indirectly)
Sure you can depclean afterwards to removed unneeded packages, but as a 
precaution a "emerge -e world" would need to be done (loss of time).

It's fine to make stage1/stage2 non-recommended as they bring no advantage 
over stage3 for most desktop systems, but should stay available and 
documented for "minority" who has valid use of it.

Bruno
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Stephen P. Becker]

Kurt Lieber wrote:
> We have received *numerous* complaints from users about the decision to
> remove stage 1 and 2 from the installation documentation.  I realize it's
> still available if users are willing to dig for it, but not all users do.
> 
> In my years of monitoring www@gentoo.org, we've received the most
> complaints about this decision than any other single decision.  Is there a
> way we can re-introduce the stages into the installation documentation,
> perhaps with gigantic warnings saying, "for advanced users only" or "use at
> your own risk"?

Well, if we could educate the users that stage2 tarballs are totally 
pointless, and that running bootstrap.sh followed by emerge -e system 
from a stage3 is pretty much *exactly* the same as starting a stage1 
from scratch...

To me, the email from that user sounds like the typical vocal minority 
of users who make "screw common sense in favor of choice" complaints 
whenever we change something for the better.

-Steve
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1460 bytes --]

On Tue, 2005-11-22 at 09:53 -0500, Stephen P. Becker wrote:
> Kurt Lieber wrote:
> > We have received *numerous* complaints from users about the decision to
> > remove stage 1 and 2 from the installation documentation.  I realize it's
> > still available if users are willing to dig for it, but not all users do.
> > 
> > In my years of monitoring www@gentoo.org, we've received the most
> > complaints about this decision than any other single decision.  Is there a
> > way we can re-introduce the stages into the installation documentation,
> > perhaps with gigantic warnings saying, "for advanced users only" or "use at
> > your own risk"?
> 
> Well, if we could educate the users that stage2 tarballs are totally 
> pointless, and that running bootstrap.sh followed by emerge -e system 
> from a stage3 is pretty much *exactly* the same as starting a stage1 
> from scratch...

It isn't pretty much anymore.  It *is* exactly the same.

> To me, the email from that user sounds like the typical vocal minority 
> of users who make "screw common sense in favor of choice" complaints 
> whenever we change something for the better.

Exactly.

It sounds like we are letting ourselves be swayed by a few heated words
from someone who is obviously shooting for a reaction.

If we give in, the terrorists have won.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

[gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Thomas Kirchner]

[-- Attachment #1: Type: text/plain, Size: 891 bytes --]

* On Nov 22 10:15, Chris Gianelloni (gentoo-dev@lists.gentoo.org) wrote:
> It isn't pretty much anymore.  It *is* exactly the same.

Correct me if I'm wrong, but if you're going to attempt to get the same 
result as a stage1 with a stage3 - (which won't be exactly the same, 
given the defaults of stage3 that we don't all agree with) - don't you 
have to download the huge stage3 tarball *and* all of the distfiles?  
With a stage1, you only need the distfiles.  Seems like a big change, 
especially for those without fast connections.

I'm against this change, personally.  Stage1 has *always* been for 
advanced users.  If someone screws up their own system (which is possible 
in any number of other ways, as well) then it's their fault.  Gentoo 
isn't about babying users.  It is about choice.  And if changing that is 
the only way to reduce your workload, well...
Tom

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1614 bytes --]

On Tue, 2005-11-22 at 11:19 -0500, Thomas Kirchner wrote:
> * On Nov 22 10:15, Chris Gianelloni (gentoo-dev@lists.gentoo.org) wrote:
> > It isn't pretty much anymore.  It *is* exactly the same.
> 
> Correct me if I'm wrong, but if you're going to attempt to get the same 
> result as a stage1 with a stage3 - (which won't be exactly the same, 

You're wrong.  You can still change your virtuals, just like you can
with a stage3.  You can still change CFLAGS/USE flags.

> given the defaults of stage3 that we don't all agree with) - don't you 
> have to download the huge stage3 tarball *and* all of the distfiles?  
> With a stage1, you only need the distfiles.  Seems like a big change, 
> especially for those without fast connections.

Honestly, you only need download what you've changed.  If you have
changed quite a lot, yes, it will be more to download.

> I'm against this change, personally.  Stage1 has *always* been for 
> advanced users.  If someone screws up their own system (which is possible 
> in any number of other ways, as well) then it's their fault.  Gentoo 
> isn't about babying users.  It is about choice.  And if changing that is 
> the only way to reduce your workload, well...

*sigh*

Another "Gentoo is about choice" argument.  Can I ask you something?
Where does it say that Gentoo is about choice?  I see lots of places
that say that Gentoo allows you to customize, but nowhere do I see
anything that says that we are about choice.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Harald van Dijk]

[-- Attachment #1: Type: text/plain, Size: 1787 bytes --]

On Tue, Nov 22, 2005 at 11:39:29AM -0500, Chris Gianelloni wrote:
> Another "Gentoo is about choice" argument.  Can I ask you something?
> Where does it say that Gentoo is about choice?  I see lots of places
> that say that Gentoo allows you to customize, but nowhere do I see
> anything that says that we are about choice.

http://www.gentoo.org/doc/en/handbook/2005.1/handbook-x86.xml?part=1
"About the Gentoo Linux Installation
 Users not familiar with Gentoo do not always know that choice is what
 Gentoo is all about."

And following that link:

http://www.gentoo.org/doc/en/handbook/2005.1/handbook-x86.xml?part=1&chap=1
"Welcome!

 First of all, welcome to Gentoo. You are about to enter the world of
 choices and performance. Gentoo is all about choices. When installing
 Gentoo, this is made clear to you several times -- you can choose how
 much you want to compile yourself, how to install Gentoo, what system
 logger you want, etc.

 Gentoo is a fast, modern metadistribution with a clean and flexible
 design. Gentoo is built around free software and doesn't hide from its
 users what is beneath the hood. Portage, the package maintenance system
 which Gentoo uses, is written in Python, meaning you can easily view and
 modify the source code. Gentoo's packaging system uses source code
 (although support for precompiled packages is included too) and
 configuring Gentoo happens through regular textfiles. In other words,
 openness everywhere.

 It is very important that you understand that choices are what makes
 Gentoo run. We try not to force you onto anything you don't like. If you
 feel like we do, please bugreport it."

(Note that I'm not going to argue either way whether this is a good
thing; I'm merely pointing out that the docs do say we're about choice.)

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Simon Stelling]

Harald van Dijk wrote:
> (Note that I'm not going to argue either way whether this is a good
> thing; I'm merely pointing out that the docs do say we're about choice.)

You still can choose between stage3 and stage3+GRP without having to do anything 
but following the handbook :)

-- 
Simon Stelling
Gentoo/AMD64 Operational Co-Lead
blubb@gentoo.org
-- 
gentoo-dev@gentoo.org mailing list

Re[2]: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 814 bytes --]


22.11.2005, 18:51:12, Simon Stelling wrote:

> Harald van D?k wrote:
>> (Note that I'm not going to argue either way whether this is a good
>> thing; I'm merely pointing out that the docs do say we're about choice.)

> You still can choose between stage3 and stage3+GRP without having to do anything
> but following the handbook :)

Ah, that's a nice choice... ;p

Seriously, I suppose noone plans to remove stage1 from the mirrors; then it
would be nice to have it properly documented (probably outside of handbook and
with warning that it's an unsupported install method if releng wish so). It's
still useful for people and I guess complete removal of docs will just lead to
some half-baked howtos published in forums.g.o. and people complaining on IRC
that it does not work.

-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: Re[2]: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Joe McCann]

[-- Attachment #1: Type: text/plain, Size: 934 bytes --]

On Tue, 2005-11-22 at 18:59 +0100, Jakub Moc wrote:
> 22.11.2005, 18:51:12, Simon Stelling wrote:
> 
> > Harald van D?k wrote:
> >> (Note that I'm not going to argue either way whether this is a good
> >> thing; I'm merely pointing out that the docs do say we're about choice.)
> 
> > You still can choose between stage3 and stage3+GRP without having to do anything
> > but following the handbook :)
> 
> Ah, that's a nice choice... ;p
> 
> Seriously, I suppose noone plans to remove stage1 from the mirrors; then it
> would be nice to have it properly documented (probably outside of handbook and
> with warning that it's an unsupported install method if releng wish so). It's
> still useful for people and I guess complete removal of docs will just lead to
> some half-baked howtos published in forums.g.o. and people complaining on IRC
> that it does not work.
> 

http://www.gentoo.org/doc/en/faq.xml#stage12

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Abhay Kedia]

[-- Attachment #1: Type: text/plain, Size: 3398 bytes --]

On Tuesday 22 Nov 2005 10:09 pm, Chris Gianelloni wrote:
> *sigh*
>
> Another "Gentoo is about choice" argument.  Can I ask you something?
> Where does it say that Gentoo is about choice?  I see lots of places
> that say that Gentoo allows you to customize, but nowhere do I see
> anything that says that we are about choice.

I am a really novice desktop end-user and am following gentoo-dev just for 
learning what all goes through the minds of the uber gentoo developers. I 
have no say in this discussion as it doesn't effect me and am certainly not 
qualified to get into an argument with someone like you but I have read your 
posts mentioning this "Where does it say that Gentoo is about choice?" 
argument lots of time.

Till now I also had a picture in my mind that Gentoo was actually about 
"choice" and when I saw that picture getting shattered by a Lead Developer, I 
went to look for the places that made me think about Gentoo in that way i.e. 
"Gentoo is about choice". These are the few things I could find.

1) On the about page with picture of "Larry The Cow": 
http://www.gentoo.org/main/en/about.xml
"He discovered lots of up-to-date packages that could be auto-built
using the optimizations settings and build-time functionality that
he wanted, rather than what some distro creator thought would be
best for him. All of the sudden, Larry the Cow was in control. And
he liked it."
---rather than what some distro creator thought would be
best for him.
^ that statement makes you think it is about choice.

2) The Philosophy: http://www.gentoo.org/main/en/philosophy.xml
"If the tool forces the user to do things a particular way, then the tool is 
working against, rather than for, the user."

3) Gentoo Social Contract: http://www.gentoo.org/main/en/contract.xml
"A Gentoo operating system should satisfy the self-hosting requirement. In 
other words, the operating system should be able to build itself from scratch 
using the aforementioned tools and metadata. If a product associated with an 
official Gentoo project does not satisfy these requirements, the product does 
not qualify as a Gentoo operating system."

All these things "imply" that there should be a choice for a user to do what 
ever way he/she wants while building his/her system i.e. even from scratch.

Since these documents just implied the "Choice" nature of Gentoo, I went ahead 
and did some googling to actually get the direct connection. Searching for 
"gentoo about choice" leads 653,000 results and just the first two results 
are enough to get the point across for a user.

1) From 1st Link:
Gentoo Weekly Newsletter: March 28th, 2005
http://www.gentoo.org/news/en/gwn/20050328-newsletter.xml
Developer of the week talks
"Gentoo represents choice and freedom for every user to build their computing 
environment to their individual needs, by giving them the tools to do it." -- 
Marcus D. Hanwell (cryos)

2) From 2nd Link:
Trusted Gentoo : by Daniel Black
http://www.gentoo.org/news/20050202-trustedgentoo.xml
"Gentoo is about choice"

The last link should settle it for you?
Can we now comfortably say that "Gentoo is about choice"? The other 652,998 
links might reveal a few more places where we can get the choice idea from 
but I hope that all these links should be sufficient to give anyone this 
idea.

Regards,
Abhay

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Stephen P. Becker]

> The last link should settle it for you?
> Can we now comfortably say that "Gentoo is about choice"? The other 652,998 
> links might reveal a few more places where we can get the choice idea from 
> but I hope that all these links should be sufficient to give anyone this 
> idea.

Ok, fine.  Gentoo is about choice.  So what about developers?  Don't we 
also have a choice?  Sometimes we have to choose what is best for 
ourselves (note, I'm not talking about anything selfish or malicious 
here).  Sometimes we have to choose what is best for Gentoo.  In most 
cases, we are even right, although you would never guess this from the 
amount of complaining that occurs anytime we make such a choice...

-Steve
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Abhay Kedia]

[-- Attachment #1: Type: text/plain, Size: 939 bytes --]

On Tuesday 22 Nov 2005 11:32 pm, Stephen P. Becker wrote:
> Ok, fine.  Gentoo is about choice.  So what about developers?  Don't we
> also have a choice?  Sometimes we have to choose what is best for
> ourselves (note, I'm not talking about anything selfish or malicious
> here).  Sometimes we have to choose what is best for Gentoo.  In most
> cases, we are even right, although you would never guess this from the
> amount of complaining that occurs anytime we make such a choice...

As I said earlier I don't have any credentials/knowledge/expertise (or 
anything you might want to call it) that can enable me to argue with you 
guys. Also I am not effected by the removal of Stage 1 and 2 installation 
docs. I will not be effected even if Stage 1/2 tar balls are removed as well. 
I was simply responding to Chris Gianelloni's comment of "Where does it say 
that Gentoo is about choice?" That is ALL. Period!!!

Abhay

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 7251 bytes --]

On Tue, 2005-11-22 at 23:10 +0530, Abhay Kedia wrote:
> On Tuesday 22 Nov 2005 10:09 pm, Chris Gianelloni wrote:
> > *sigh*
> >
> > Another "Gentoo is about choice" argument.  Can I ask you something?
> > Where does it say that Gentoo is about choice?  I see lots of places
> > that say that Gentoo allows you to customize, but nowhere do I see
> > anything that says that we are about choice.
> 
> I am a really novice desktop end-user and am following gentoo-dev just for 
> learning what all goes through the minds of the uber gentoo developers. I 
> have no say in this discussion as it doesn't effect me and am certainly not 
> qualified to get into an argument with someone like you but I have read your 
> posts mentioning this "Where does it say that Gentoo is about choice?" 
> argument lots of time.

You have just as much right to speak your mind as I.  We aren't special
because we're developers.  We're all just Gentoo users like you.  We
just contribute our time to improve Gentoo.  If you file bug reports or
participate in discussions here, then you're doing the same.

> Till now I also had a picture in my mind that Gentoo was actually about 
> "choice" and when I saw that picture getting shattered by a Lead Developer, I 
> went to look for the places that made me think about Gentoo in that way i.e. 
> "Gentoo is about choice". These are the few things I could find.

The problem with the "Gentoo is about choice" argument is that it is
used to back up any argument where there's not really a good reason for
making the changes *except* for choice.

> 1) On the about page with picture of "Larry The Cow": 
> http://www.gentoo.org/main/en/about.xml
> "He discovered lots of up-to-date packages that could be auto-built
> using the optimizations settings and build-time functionality that
> he wanted, rather than what some distro creator thought would be
> best for him. All of the sudden, Larry the Cow was in control. And
> he liked it."
> ---rather than what some distro creator thought would be
> best for him.
> ^ that statement makes you think it is about choice.

It can imply that, but it does not state it.

Also, remember that having the *ability* to enact change yourself to
make things the way you want is not the same as developers being
*forced* to do something simply so you have a choice.

As I stated before, you're more than able to take a stage3 tarball +
catalyst + the example catalyst spec files and build your own stage1
tarball.  In fact, this is the exact same procedure that Release
Engineering uses in building these tarballs to begin with.  So the
"choice" is still there.

> 2) The Philosophy: http://www.gentoo.org/main/en/philosophy.xml
> "If the tool forces the user to do things a particular way, then the tool is 
> working against, rather than for, the user."

Again, you're still free to do what you chose using the tools we
provide...

...or are you calling *me* a tool?  ;P

> 3) Gentoo Social Contract: http://www.gentoo.org/main/en/contract.xml
> "A Gentoo operating system should satisfy the self-hosting requirement. In 
> other words, the operating system should be able to build itself from scratch 
> using the aforementioned tools and metadata. If a product associated with an 
> official Gentoo project does not satisfy these requirements, the product does 
> not qualify as a Gentoo operating system."

Removing stage1 instructions doesn't go against this in any way.

> All these things "imply" that there should be a choice for a user to do what 
> ever way he/she wants while building his/her system i.e. even from scratch.

Nowhere have I suggested a method of change that would remove this
"choice" from the users.

The problem is that people are confusing "choice" with "developers doing
it for me".  Just because *I* don't build a stage1 tarball, or just
because the instructions are not in the Handbook, does not mean that you
cannot still perform a stage1 installation.  All of the tools for you to
do this are still there, you just have to take the time to use them.

> Since these documents just implied the "Choice" nature of Gentoo, I went ahead 
> and did some googling to actually get the direct connection. Searching for 
> "gentoo about choice" leads 653,000 results and just the first two results 
> are enough to get the point across for a user.

They implied it.  They did not state it.  However, it was pointed out to
me that the Handbook, does indeed have the "Gentoo is about choice"
mantra in it.  I plan on filing a bug against the Handbook to have this
changed to something a bit closer to fact, which I will explain a bit
further below.

> 1) From 1st Link:
> Gentoo Weekly Newsletter: March 28th, 2005
> http://www.gentoo.org/news/en/gwn/20050328-newsletter.xml
> Developer of the week talks
> "Gentoo represents choice and freedom for every user to build their computing 
> environment to their individual needs, by giving them the tools to do it." -- 
> Marcus D. Hanwell (cryos)

Right.  Giving them the tools.  It doesn't say being forced to do
something they think is a bad idea and to continue bad practices simply
to ease user issues with the choices that they have made.  Also, a
single developer's statement does not speak for the entire project.
What Marcus has to say speaks for the entire project no more than what I
say does, which is that I only speak for myself.

> 2) From 2nd Link:
> Trusted Gentoo : by Daniel Black
> http://www.gentoo.org/news/20050202-trustedgentoo.xml
> "Gentoo is about choice"
> 
> The last link should settle it for you?

Not really.

> Can we now comfortably say that "Gentoo is about choice"? The other 652,998 
> links might reveal a few more places where we can get the choice idea from 
> but I hope that all these links should be sufficient to give anyone this 
> idea.

No.

I think instead we should say that Gentoo is about empowerment.  Allow
me to explain.  Saying that Gentoo is about choice implies that we will
take any patches or any feature requests, no matter how pointless or
useless, and increase our workload, simply to give our users more
"choice".  This is, in fact, extremely far from the truth.  Instead, we
give the users the tools that they need to accomplish what they want.
We *empower* them to make choices.  If you want a stage1 tarball, make
one.  Don't like our choices of kernel sources?  Add your own to an
overlay.  Just because you have the ability to enact a change does not
mean that we should be forced to put it into the main distribution and
*support* it.  Because of this, Gentoo is definitely *not* about choice,
but is about *empowering* our users to make choices.  They're only
subtly different, but it is that difference that keeps Gentoo running.

Check bugs.gentoo.org for WONTFIX resolution codes.  Look for places
where we have refused to add features unless they were adopted by
upstream.  You will find countless examples of us *refusing* to add
choices to ensure the quality of Gentoo as a whole.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Abhay Kedia]

[-- Attachment #1: Type: text/plain, Size: 1126 bytes --]

On Wednesday 23 Nov 2005 12:29 am, Chris Gianelloni wrote:
>
> As I stated before, you're more than able to take a stage3 tarball +
> catalyst + the example catalyst spec files and build your own stage1
> tarball.  In fact, this is the exact same procedure that Release
> Engineering uses in building these tarballs to begin with.  So the
> "choice" is still there.
>
I agree with you totally and have no problems at all in having Stage 1/2 
methods or even the tar balls officially removed from Gentoo but all these 
complaints by users (or slashdotters as beejay says) are arising because the 
whole thing has not been documented too well. The users just know one thing 
i.e. Stage 1 and 2 have been removed and hence the knee jerk reaction. They 
don't know that these methods have *not* been removed, just the way to get to 
the point has changed (no matter how worthless it is).

I actually would like to reiterate what Henrik said in one of his posts "If 
our users are explained why stage1/2 installs don't give any benefits over a 
stage3 install, I trust them to acknowledge this fact."


Abhay

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Danny van Dyk]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Thomas Kirchner schrieb:
| I'm against this change, personally.  Stage1 has *always* been for
| advanced users.  If someone screws up their own system (which is possible
| in any number of other ways, as well) then it's their fault.  Gentoo
| isn't about babying users.  It is about choice.  And if changing that is
| the only way to reduce your workload, well...

In which way do we take away your ability to choose by moving
documentation from one place to another one? It's just the aim to not
include this documentation into the handbook (which ends up on the
installcds) and to move it out of the scope of ricers.

Advanced won't need that documentation anyway, as they don't qualify as
adcaneced if they do! (At least in my eyes)

Danny
- --
Danny van Dyk <kugelfang@gentoo.org>
Gentoo/AMD64 Project, Gentoo Scientific Project
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFDg1/QaVNL8NrtU6IRApExAKCnhZf9k+PjJOpWTYlO+C4Np+csigCeO3tC
Eh8xRxPL3AHyqRylAtTjvuc=
=/oto
-----END PGP SIGNATURE-----
-- 
gentoo-dev@gentoo.org mailing list

Re[2]: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 932 bytes --]


22.11.2005, 19:13:36, Danny van Dyk wrote:

> Thomas Kirchner schrieb: | I'm against this change, personally.  Stage1 has
> *always* been for | advanced users.  If someone screws up their own system
> (which is possible | in any number of other ways, as well) then it's their
> fault.  Gentoo | isn't about babying users.  It is about choice.  And if
> changing that is | the only way to reduce your workload, well...

> In which way do we take away your ability to choose by moving
> documentation from one place to another one? It's just the aim to not
> include this documentation into the handbook (which ends up on the
> installcds) and to move it out of the scope of ricers.

If you look at http://www.gentoo.org/doc/en/faq.xml (How do I Install Gentoo
Using a Stage1 or Stage2 Tarball?) I'd not exactly say that the documentations
has been *moved*. Compare that with the original handbook.


-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Danny van Dyk]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Jakub Moc schrieb:
| 22.11.2005, 19:13:36, Danny van Dyk wrote:
|
|
|>Thomas Kirchner schrieb: | I'm against this change, personally.
Stage1 has
|>*always* been for | advanced users.  If someone screws up their own system
|>(which is possible | in any number of other ways, as well) then it's their
|>fault.  Gentoo | isn't about babying users.  It is about choice.  And if
|>changing that is | the only way to reduce your workload, well...
|
|
|>In which way do we take away your ability to choose by moving
|>documentation from one place to another one? It's just the aim to not
|>include this documentation into the handbook (which ends up on the
|>installcds) and to move it out of the scope of ricers.
|
|
| If you look at http://www.gentoo.org/doc/en/faq.xml (How do I Install
Gentoo
| Using a Stage1 or Stage2 Tarball?) I'd not exactly say that the
documentations
| has been *moved*. Compare that with the original handbook.
That's currently a stub. If i recall right, Swift mentioned on #-dev
that he'd need to refurbish this.
Swift: Right?

Danny
- --
Danny van Dyk <kugelfang@gentoo.org>
Gentoo/AMD64 Project, Gentoo Scientific Project
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.1 (GNU/Linux)

iD8DBQFDg2evaVNL8NrtU6IRAqH9AJ9JtN58oKu80O2QPLnbWezZbJte9wCdFKBG
WDdFjRu2HBZF0tYswBp2v6g=
=QcNK
-----END PGP SIGNATURE-----
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Re: Decision to remove stage1/2 from installation documentation

[Sven Vermeulen]

[-- Attachment #1: Type: text/plain, Size: 5336 bytes --]

On Tue, Nov 22, 2005 at 07:47:11PM +0100, Danny van Dyk wrote:
[... About the stage1/2 instructions in the Gentoo FAQ ...]
> That's currently a stub. If i recall right, Swift mentioned on #-dev
> that he'd need to refurbish this.

Lots of heat again. Good thing, because it is quite cold here in Belgium.

When I originally created the Gentoo Handbook, I hoped that it would contain
all Gentoo-specific documentation in one place: the installation
instructions for all architectures, all possible methods. However, that
first attempt had its fundamental flaws, the major one being the stupid
believe that users wouldn't mind reading about other architectures if they
are well guided through the instructions.

So the Handbook changed and split. Architecture-specific instructions were
moved to separate files and each architecture had its own handbook even
though many parts of it were shared, allowing the Gentoo Documentation
Project to maintain all Handbooks without having one or more of them become
too outdated easily (which was what happened with the separate installation
guides).

Yet this attempt still had its difficulties: when the Release Engineering
team decided that a quarterly release was too stressful (they had to do more
packaging and deployment rather than research and development) they also
made the Gentoo Documentation Project split the handbooks in two: one which
contained the Internet-based installation instructions, using the latest
stable packages (baselayout), while the other handbook contained the
instructions that were statically bound to a certain release.

This was needed because, at that time, Gentoo had a history of changing core
system configuration too often making it too darn difficult to keep the
Handbook in good shape. Right now, I believe that those causes are invalid
and that the separate handbooks can be combined again, especially with the
request to move the stage1/2 instructions elsewhere.

I was quite reluctant to move the instructions at first, but when I found
out that the instructions were indeed not perfect, I had two choices: either
update the instructions in the Handbook to be correct, or move the
instructions outside the Handbook first (making sure that the official
installation instructions remain bugfree) and write a separate guide on
bootstrapping.

Based on the input I've gathered from the gentoo-doc mailinglist,
gentoo-releng mailinglist, Gentoo Forums and various other sources it was
quite obvious that a *very* *short* amount of users was aware of the theory
(and practice) behind bootstrapping. In fact, most saw "stage-1" as the
online drug to increate their, quoting Xavier Neys, "ePenis". And not only
that, but I also found that I personally lacked the knowledge to write
something decent about bootstrapping.

Therefore I decided to move the instructions that were in the Gentoo
Handbook to the Gentoo FAQ in the first place. I intended to have the FAQ be
accurate with the information I already had without losing anything
important. I did miss something in that procedure, namely the change of the
CHOST variable, but other than that the FAQ contains the same instructions
as were in the Gentoo Handbook.

The next step for me was (and still is) to investigate what bootstrapping by
itself means. Why whould anyone need to rebuild this toolchain twice? I
could perfectly understand why it was needed the first time, but why the
second time? The only reason I could give myself was that it was to test the
toolchain: if it can rebuild itself, it can build all other packages.

After finally figuring out what bootstrapping is (with input from a nice
forum thread in the "Gentoo Chat" department, information gathered from the
GCC mailinglist and some dev prodding online) I am now trying to work out a
reasonable scenario as to why someone would bootstrap his system as an
example for the guide.

You might be wondering why I didn't first write the damn guide and /then/
update the Gentoo Handbook. Two reasons are behind this. First, the Gentoo
Release Engineering project has asked me to do so, and they were kind enough
to give reasons (like bugreports, but also the theoretical problem with the
bootstrapping/system stuff, circular dependency stuff, etc.). The
installation instructions and the release engineering project are two hands
that should always work together and any discrepancy between them would lead
to confusion of the user.

The second one is my personal motivation: I want to be certain that users
/comprehend/ what they are doing rather than blindly copying over
instructions from one screen to another. We've had (and still have) lots of
users break their initial installation because they "forgot" to edit their
/etc/fstab. For a documentation writer, this is unacceptable. Any failed
installation is seen by me as either a (1.) very stupid user, or (2.)
failure on my part to document the instructions well.

This is my motivation, and this motivation is mine. 

Sincerely,

      Sven Vermeulen

-- 
  Gentoo Foundation Trustee          |  http://foundation.gentoo.org
  Gentoo Documentation Project Lead  |  http://www.gentoo.org/proj/en/gdp
  Gentoo Council Member  

  The Gentoo Project   <<< http://www.gentoo.org >>>

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Grant Goodyear]

[-- Attachment #1: Type: text/plain, Size: 1059 bytes --]

Chris Gianelloni wrote: [Tue Nov 22 2005, 09:15:27AM CST]
> > Well, if we could educate the users that stage2 tarballs are totally 
> > pointless, and that running bootstrap.sh followed by emerge -e system 
> > from a stage3 is pretty much *exactly* the same as starting a stage1 
> > from scratch...
> 
> It isn't pretty much anymore.  It *is* exactly the same.

I keep hearing this, isn't there a real difference between a stage 1 and
a stage 3 install inasmuch as somebody who needs (or wants) to
dramatically tailor what's in the system profile can choose to do so
from a stage 1 or 2, but would have to remove packages after the fact if
starting from a stage 3?  I wouldn't have a problem with that, as long
as we document it, but it just seems that the claim that the old and new
methods produce _exactly_ the same results seems to be stretching things
a bit.

-g2boojum-
-- 
Grant Goodyear	
Gentoo Developer
g2boojum@gentoo.org
http://www.gentoo.org/~g2boojum
GPG Fingerprint: D706 9802 1663 DEF5 81B0  9573 A6DC 7152 E0F6 5B76

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re[2]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 810 bytes --]


22.11.2005, 19:03:49, Grant Goodyear wrote:

> I keep hearing this, isn't there a real difference between a stage 1 and a
> stage 3 install inasmuch as somebody who needs (or wants) to dramatically
> tailor what's in the system profile can choose to do so from a stage 1 or 2,
> but would have to remove packages after the fact if starting from a stage 3?
> I wouldn't have a problem with that, as long as we document it, but it just
> seems that the claim that the old and new methods produce _exactly_ the same
> results seems to be stretching things a bit.

> -g2boojum-

Uhm, which reliable tools do we have for removing no-longer needed packages?
emerge --depclean producing the huge red "I'm broken" warning? Or emerge
--prune with similar warning in man page?

Hmmm...

-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 993 bytes --]

On Tue, 22 Nov 2005 12:03:49 -0600 Grant Goodyear <g2boojum@gentoo.org>
wrote:
| I keep hearing this, isn't there a real difference between a stage 1
| and a stage 3 install inasmuch as somebody who needs (or wants) to
| dramatically tailor what's in the system profile can choose to do so
| from a stage 1 or 2, but would have to remove packages after the fact
| if starting from a stage 3?  I wouldn't have a problem with that, as
| long as we document it

emerge -e world && emerge -e world && emerge depclean

Exactly the same results, except that it won't fall over and die
because of unlisted circular dependencies.

| but it just seems that the claim that the old and new methods produce
| _exactly_ the same results seems to be stretching things a bit.

How do you think stage3s are built in the first place?

-- 
Ciaran McCreesh : Gentoo Developer (Look! Shiny things!)
Mail            : ciaranm at gentoo.org
Web             : http://dev.gentoo.org/~ciaranm


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Grant Goodyear]

[-- Attachment #1: Type: text/plain, Size: 1922 bytes --]

Ciaran McCreesh wrote: [Tue Nov 22 2005, 12:17:47PM CST]
> On Tue, 22 Nov 2005 12:03:49 -0600 Grant Goodyear <g2boojum@gentoo.org>
> wrote:
> | I keep hearing this, isn't there a real difference between a stage 1
> | and a stage 3 install inasmuch as somebody who needs (or wants) to
> | dramatically tailor what's in the system profile can choose to do so
> | from a stage 1 or 2, but would have to remove packages after the fact
> | if starting from a stage 3?  I wouldn't have a problem with that, as
> | long as we document it
> 
> emerge -e world && emerge -e world && emerge depclean

Cool.  Why rebuild twice?  Any chance we could add this to the FAQ?

> | but it just seems that the claim that the old and new methods produce
> | _exactly_ the same results seems to be stretching things a bit.
> 
> How do you think stage3s are built in the first place?

Sorry, poor phrasing on my part.  Of course it's true that if one
follows the handbook (either the current or the previous version), then
one ends up with the same system regardless of whether or not a stage1,
stage2, or stage3 is used.  What I intended to suggest was that
tinkering at the system level is less obviously accomplished when
starting from a stage3, so the occasional assertion I've read that
starting from a stage 1 or stage 2 provides no benefits over starting
from a stage 1 or 2 didn't seem right to me.

In any event, I don't mind the handbook changes, although I'd perhaps
like to see the FAQ for starting from a stage 1 fleshed out a tad, such
as including a paragraph of why one might not want to do that.  Perhaps
steal from whomever posted a treatise on the issue some time ago (either
rac or avenj, I don't remember which)?

-g2boojum-
-- 
Grant Goodyear	
Gentoo Developer
g2boojum@gentoo.org
http://www.gentoo.org/~g2boojum
GPG Fingerprint: D706 9802 1663 DEF5 81B0  9573 A6DC 7152 E0F6 5B76

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 871 bytes --]

On Tue, 2005-11-22 at 13:04 -0600, Grant Goodyear wrote:
> In any event, I don't mind the handbook changes, although I'd perhaps
> like to see the FAQ for starting from a stage 1 fleshed out a tad, such
> as including a paragraph of why one might not want to do that.  Perhaps
> steal from whomever posted a treatise on the issue some time ago (either
> rac or avenj, I don't remember which)?

rac... though most of his arguments are invalid once we updated the
stages.  His arguments were that we should always use stage3 because
stages 1 and 2 caused problems, which they did.  These problems have
been resolved by making the stages fuller.  However, in doing so, we
reduced the legitimate need for having these stages to almost nothing.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Daniel Ostrow]

On Tue, 2005-11-22 at 12:03 -0600, Grant Goodyear wrote:
> Chris Gianelloni wrote: [Tue Nov 22 2005, 09:15:27AM CST]
> > > Well, if we could educate the users that stage2 tarballs are totally 
> > > pointless, and that running bootstrap.sh followed by emerge -e system 
> > > from a stage3 is pretty much *exactly* the same as starting a stage1 
> > > from scratch...
> > 
> > It isn't pretty much anymore.  It *is* exactly the same.
> 
> I keep hearing this, isn't there a real difference between a stage 1 and
> a stage 3 install inasmuch as somebody who needs (or wants) to
> dramatically tailor what's in the system profile can choose to do so
> from a stage 1 or 2, but would have to remove packages after the fact if
> starting from a stage 3?  I wouldn't have a problem with that, as long
> as we document it, but it just seems that the claim that the old and new
> methods produce _exactly_ the same results seems to be stretching things
> a bit.
> 
> -g2boojum-

There are 3 purposes to a stage1/stage2 install (note all 3 can be
achieved with either a stage3 or a custom rolled stage though catalyst):

1). Modify the bootstrap.sh script to change what the "stage2" target
produces.
2). Modify the system target to change what the "stage3" target
produces.
3). Modify the CHOST/CFLAGS/USE et. al. early on to create a customized
and largely unsupported (things like hardened, uclibc, and embedded are
exceptions to the unsupported rule) "stage3" target. 

#3 is where the vast majority of user created bugs occur. The purpose of
highly encouraging users to start with a stage3, by making the handbook
only refer to it, is to make sure that users have a known working
configuration to start their customization from. There are many many
ways to mess up going from a stage1 to a stage3, there are fewer ways to
mess up customizing and recompiling a system that has already been
configured to boot on it's own.

By and large most users will never want to do any of the above for the
reasons that they are really valid for, and any user or developer that
does will have access to both the stages (with relocated documentation)
and catalyst itself to make their own. We are not removing choice
here...just *potentially* making someones initial download time longer. 

Personally I wouldn't be at all opposed to moving the stage1 and stage2
tarballs to another directory on the mirrors (documented of course),
just to make it that much clearer that if you start from a stage1 or a
stage2 RelEng won't support you if any errors occur during system build.

If RelEng ever does get to the point of removing stage's 1 & 2 from the
mirrors (something that has been discussed but isn't on the table at all
right now) end users and developers alike will still be able to generate
them on their own using catalyst and the provided spec files...sure it
is an extra step and all but it's not all that huge...

-- 
Daniel Ostrow
Gentoo Foundation Board of Trustees
Gentoo/{PPC,PPC64,DevRel}
dostrow@gentoo.org

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] [OT] Decision to remove stage1/2 from installation documentation

[lnxg33k]

A few posts here have mentioned Catalyst and its respective documentation. I
remember tossing out some bugs about the docs and was told they were old and
being redone. A quick google seems to bring up some dated stuff. Anyway, are
those docs up yet (I'd even be interested in working copies) and, if so, where
at? Thanks.
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] [OT] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 996 bytes --]

On Thu, 2005-11-24 at 14:49 -0600, lnxg33k wrote:
> A few posts here have mentioned Catalyst and its respective documentation. I
> remember tossing out some bugs about the docs and was told they were old and
> being redone. A quick google seems to bring up some dated stuff. Anyway, are
> those docs up yet (I'd even be interested in working copies) and, if so, where
> at? Thanks.

USE=doc emerge catalyst

That gives the definitive spec file templates which are well documented.
The online documentation is still for catalyst 1.x, which will be phased
out over the next couple weeks for catalyst 2.0, which is current in
pre-release testing and is masked.  Fell free to unmask it and play with
it, as it works fine.  We are just wanting to iron out the last few bugs
before releasing it, so feel free to bang away on it and report any
problems that you find.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] [OT] Decision to remove stage1/2 from installation documentation

[lnxg33k]

Chris Gianelloni wrote:
> On Thu, 2005-11-24 at 14:49 -0600, lnxg33k wrote:
<snip>
> USE=doc emerge catalyst
> 
> That gives the definitive spec file templates which are well documented.
> The online documentation is still for catalyst 1.x, which will be phased
> out over the next couple weeks for catalyst 2.0, which is current in
> pre-release testing and is masked.  Fell free to unmask it and play with
> it, as it works fine.  We are just wanting to iron out the last few bugs
> before releasing it, so feel free to bang away on it and report any
> problems that you find.
> 

Thanks for getting back. I'll give that a try and if you guys don't get it out
before I get a chance to play, then I'll definately unmask and see if I can find
anything.
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1405 bytes --]

On Tue, 2005-11-22 at 12:03 -0600, Grant Goodyear wrote:
> Chris Gianelloni wrote: [Tue Nov 22 2005, 09:15:27AM CST]
> > > Well, if we could educate the users that stage2 tarballs are totally 
> > > pointless, and that running bootstrap.sh followed by emerge -e system 
> > > from a stage3 is pretty much *exactly* the same as starting a stage1 
> > > from scratch...
> > 
> > It isn't pretty much anymore.  It *is* exactly the same.
> 
> I keep hearing this, isn't there a real difference between a stage 1 and
> a stage 3 install inasmuch as somebody who needs (or wants) to
> dramatically tailor what's in the system profile can choose to do so
> from a stage 1 or 2, but would have to remove packages after the fact if
> starting from a stage 3?  I wouldn't have a problem with that, as long
> as we document it, but it just seems that the claim that the old and new
> methods produce _exactly_ the same results seems to be stretching things
> a bit.

Who said that removing something isn't a part of the procedure to get an
identical build?

The point is that following the proper steps, one *can* get the exact
same output.  This would include using --newuse and cleaning out unused
packages, along with any other maintenance items that would be required.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Grant Goodyear]

[-- Attachment #1: Type: text/plain, Size: 1206 bytes --]

Chris Gianelloni wrote: [Tue Nov 22 2005, 01:06:03PM CST]
> Who said that removing something isn't a part of the procedure to get an
> identical build?

Yeah, my phrasing was lousy (which I noted in another e-mail, but I
doubt you had time to see it before replying to this one).
> 
> The point is that following the proper steps, one *can* get the exact
> same output.  This would include using --newuse and cleaning out unused
> packages, along with any other maintenance items that would be required.

That's fine with me.  All I really want to do is ensure that we preserve
our users' ability to tinker with system without making life too painful
for them.  Starting from a stage 1 it was obvious how to do such
tinkering.  I would argue that it's not quite as obvious how to do that
when starting from a stage 3, so a bit of additional documentation on
how to do that would be nice.  If that were done, then I would have no
complaints about the stage 1 and stage 2 tarballs going away altogether.
*Shrug*

-g2boojum-
-- 
Grant Goodyear	
Gentoo Developer
g2boojum@gentoo.org
http://www.gentoo.org/~g2boojum
GPG Fingerprint: D706 9802 1663 DEF5 81B0  9573 A6DC 7152 E0F6 5B76

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 2305 bytes --]

On Tue, 2005-11-22 at 13:28 -0600, Grant Goodyear wrote:
> > The point is that following the proper steps, one *can* get the exact
> > same output.  This would include using --newuse and cleaning out unused
> > packages, along with any other maintenance items that would be required.
> 
> That's fine with me.  All I really want to do is ensure that we preserve
> our users' ability to tinker with system without making life too painful
> for them.  Starting from a stage 1 it was obvious how to do such
> tinkering.  I would argue that it's not quite as obvious how to do that
> when starting from a stage 3, so a bit of additional documentation on
> how to do that would be nice.  If that were done, then I would have no
> complaints about the stage 1 and stage 2 tarballs going away altogether.

That's kinda what I am shooting for down the line.

The idea was to move out the stage1/stage2 docs to somewhere else.  Then
create some sort of "Advanced Installation Topics" guide or something,
to list out the replacement procedures for customizing a system from a
stage3 tarball, then, eventually, drop the stage1 and stage2 tarballs.
I was working on the idea of doing it all in stages.  The "problem"
occurred from people freaking out because they didn't bother reading the
entire news blurb that tells exactly where the instructions moved to,
plus links to the bug # and discussion.  There's also this nice section
in the Handbook.

"A stage3 tarball is an archive containing a minimal Gentoo environment,
suitable to continue the Gentoo installation using the instructions in
this manual. Previously, the Gentoo Handbook described the installation
using one of three stage tarballs. While Gentoo still offers stage1 and
stage2 tarballs, the official installation method uses the stage3
tarball. If you are interested in performing a Gentoo installation using
a stage1 or stage2 tarball, please read the Gentoo FAQ on How do I
Install Gentoo Using a Stage1 or Stage2 Tarball?"

Really, everybody is just up in arms over a knee-jerk reaction to not
reading carefully.  What it boils down to is either not knowing the
facts, or trolling/flaming.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re[2]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 2242 bytes --]


22.11.2005, 20:57:15, Chris Gianelloni wrote:

> The idea was to move out the stage1/stage2 docs to somewhere else.  Then
> create some sort of "Advanced Installation Topics" guide or something, to
> list out the replacement procedures for customizing a system from a stage3
> tarball, then, eventually, drop the stage1 and stage2 tarballs.

Erm, did you read what solar wrote about hardened stages and why should stage1
still stay?

> I was working on the idea of doing it all in stages.  The "problem" occurred
> from people freaking out because they didn't bother reading the entire news
> blurb that tells exactly where the instructions moved to, plus links to the
> bug # and discussion.  There's also this nice section in the Handbook.

> "A stage3 tarball is an archive containing a minimal Gentoo environment,
> suitable to continue the Gentoo installation using the instructions in
> this manual. Previously, the Gentoo Handbook described the installation
> using one of three stage tarballs. While Gentoo still offers stage1 and
> stage2 tarballs, the official installation method uses the stage3
> tarball. If you are interested in performing a Gentoo installation using
> a stage1 or stage2 tarball, please read the Gentoo FAQ on How do I
> Install Gentoo Using a Stage1 or Stage2 Tarball?"

That FAQ section has nothing in common with the original stage1 docs. Sorry,
installing stage3 to remove all the use flags cruft subsequently, bootstrap and
re-emerge the system and then ponder which packages are not needed any more
(again, there's no reliable tool to remove unneeded stuff from system, I've
already mentioned this once) - hmmm... :/

And - once stages 1+2 are removed (as you are suggesting above), then I'll
install the system only to build my own stage1 w/ catalyst, then reformat and
start over with my own stage? Ah, that makes live sooo much easier ;p


> Really, everybody is just up in arms over a knee-jerk reaction to not
> reading carefully.  What it boils down to is either not knowing the
> facts, or trolling/flaming.

Why exactly is evaporating stage1 an ultimate goal here (as it seems to me?).
So don't support it, but why it should not exist?


-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Alexey Chumakov]

Jakub Moc пишет:

>22.11.2005, 20:57:15, Chris Gianelloni wrote:
>
>  
>
>>Really, everybody is just up in arms over a knee-jerk reaction to not
>>reading carefully.  What it boils down to is either not knowing the
>>facts, or trolling/flaming.
>>    
>>
>
>Why exactly is evaporating stage1 an ultimate goal here (as it seems to me?).
>So don't support it, but why it should not exist?
>
>
>  
>
Before I insert my own word -- could somebody tell me, how and by whom
was the initial decision to eliminate the stage1 from mainstream made?

Alexey Chumakov
GDP i18n Russian Lead
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 427 bytes --]

On Tue, 2005-11-22 at 23:28 +0300, Alexey Chumakov wrote:
> Before I insert my own word -- could somebody tell me, how and by whom
> was the initial decision to eliminate the stage1 from mainstream made?

As I said before, it was made by and requested by me, after discussion
with Release Engineering.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Alec Joseph Warner]

Jakub Moc wrote:
> 22.11.2005, 20:57:15, Chris Gianelloni wrote:
> 
> 
>>The idea was to move out the stage1/stage2 docs to somewhere else.  Then
>>create some sort of "Advanced Installation Topics" guide or something, to
>>list out the replacement procedures for customizing a system from a stage3
>>tarball, then, eventually, drop the stage1 and stage2 tarballs.
> 
> 
> Erm, did you read what solar wrote about hardened stages and why should stage1
> still stay?
> 
> 
>>I was working on the idea of doing it all in stages.  The "problem" occurred
>>from people freaking out because they didn't bother reading the entire news
>>blurb that tells exactly where the instructions moved to, plus links to the
>>bug # and discussion.  There's also this nice section in the Handbook.

I'd point out that this was not well executed as a major change should 
have been.  We talked of major package changes, apache config changes, 
of package breakage.  Then one day you up and remove what some consider 
a vital part of installing with no warning.  Announcements earlier 
noting the pending removal of tarballs to say, g-announce and this list 
would probably have stifled much of the complains ( see the news hit 
gentoo-wiki, gentoo-portage, and the community ).  Otherwise yeah, you 
will get a knee-jerk reaction, many users think you just screwed them 
out of something.  Nevermind the fact that they are wrong and uninformed 
( in most cases ) you did a crappy job of conveying the message of what 
when and why.

>>"A stage3 tarball is an archive containing a minimal Gentoo environment,
>>suitable to continue the Gentoo installation using the instructions in
>>this manual. Previously, the Gentoo Handbook described the installation
>>using one of three stage tarballs. While Gentoo still offers stage1 and
>>stage2 tarballs, the official installation method uses the stage3
>>tarball. If you are interested in performing a Gentoo installation using
>>a stage1 or stage2 tarball, please read the Gentoo FAQ on How do I
>>Install Gentoo Using a Stage1 or Stage2 Tarball?"
> 
> 
> That FAQ section has nothing in common with the original stage1 docs. Sorry,
> installing stage3 to remove all the use flags cruft subsequently, bootstrap and
> re-emerge the system and then ponder which packages are not needed any more
> (again, there's no reliable tool to remove unneeded stuff from system, I've
> already mentioned this once) - hmmm... :/
> 
> And - once stages 1+2 are removed (as you are suggesting above), then I'll
> install the system only to build my own stage1 w/ catalyst, then reformat and
> start over with my own stage? Ah, that makes live sooo much easier ;p
> 

Personally if releng is already making stages 1 and 2 for the liveCD's I 
see no reason not to give that work away to the community.  Stick it in 
some unsupported/ section on the mirrors and tell people so.  Why throw 
away the work you did making the liveCD?  Can you quantify the number of 
bugs here?

-Alec Warner (antarus)
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 2282 bytes --]

On Tue, 2005-11-22 at 15:42 -0500, Alec Joseph Warner wrote:
> I'd point out that this was not well executed as a major change should 
> have been.  We talked of major package changes, apache config changes, 
> of package breakage.  Then one day you up and remove what some consider 
> a vital part of installing with no warning.  Announcements earlier 
> noting the pending removal of tarballs to say, g-announce and this list 

*sigh*

Please read the thread you're responding to before making accusations.
Nobody has removed any tarballs.

> would probably have stifled much of the complains ( see the news hit 
> gentoo-wiki, gentoo-portage, and the community ).  Otherwise yeah, you 
> will get a knee-jerk reaction, many users think you just screwed them 
> out of something.  Nevermind the fact that they are wrong and uninformed 
> ( in most cases ) you did a crappy job of conveying the message of what 
> when and why.

No, we changed some text in the Handbook to basically say "If you want
stages 1 or 2, go here" with a link to the new location.

> Personally if releng is already making stages 1 and 2 for the liveCD's I 
> see no reason not to give that work away to the community.  Stick it in 
> some unsupported/ section on the mirrors and tell people so.  Why throw 
> away the work you did making the liveCD?  Can you quantify the number of 
> bugs here?

We don't put out the livecd-stage1 portions of our CD building process,
either.  Why not?  It isn't necessary and it really has no point.

As for quantifying the number of bugs, I could do so by searching
bugzilla, but so could you.  What I cannot quantify, because I haven't
even tried to keep track, is the number of times a user has hit a
circular dependency in #gentoo or on the forums, or by coming and asking
in #gentoo-releng.  I cannot quantify the number of times a person has
asked what is so broken with our releases when they cannot bootstrap due
to some issue where a new USE flag has snuck into the dependency tree
for "system" and is now wanting kernel sources, or has pulled in a MTA
or cron daemon that wasn't the one they wanted.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: Re[2]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 3356 bytes --]

On Tue, 2005-11-22 at 21:16 +0100, Jakub Moc wrote:
> 22.11.2005, 20:57:15, Chris Gianelloni wrote:
> 
> > The idea was to move out the stage1/stage2 docs to somewhere else.  Then
> > create some sort of "Advanced Installation Topics" guide or something, to
> > list out the replacement procedures for customizing a system from a stage3
> > tarball, then, eventually, drop the stage1 and stage2 tarballs.
> 
> Erm, did you read what solar wrote about hardened stages and why should stage1
> still stay?

I read it.  For one, I am speaking of the default stages for the
releases.  If the Hardened team decides that they would prefer to
continue to offer a stage1 tarball, I wouldn't have a problem with that.
While solar's reasoning applies fine to the Hardened releases, it
doesn't apply to the default x86 releases, as we release on multiple
sub-arches, which each have their proper CHOST settings.

> > "A stage3 tarball is an archive containing a minimal Gentoo environment,
> > suitable to continue the Gentoo installation using the instructions in
> > this manual. Previously, the Gentoo Handbook described the installation
> > using one of three stage tarballs. While Gentoo still offers stage1 and
> > stage2 tarballs, the official installation method uses the stage3
> > tarball. If you are interested in performing a Gentoo installation using
> > a stage1 or stage2 tarball, please read the Gentoo FAQ on How do I
> > Install Gentoo Using a Stage1 or Stage2 Tarball?"
> 
> That FAQ section has nothing in common with the original stage1 docs. Sorry,
> installing stage3 to remove all the use flags cruft subsequently, bootstrap and
> re-emerge the system and then ponder which packages are not needed any more
> (again, there's no reliable tool to remove unneeded stuff from system, I've
> already mentioned this once) - hmmm... :/

No.  That FAQ section is there to describe how to install from a stage1
or stage2 tarball and has nothing to do with a stage3 tarball, nor did I
ever say that it would.  I'm not sure I understand what you're getting
at here.

> And - once stages 1+2 are removed (as you are suggesting above), then I'll
> install the system only to build my own stage1 w/ catalyst, then reformat and
> start over with my own stage? Ah, that makes live sooo much easier ;p

You would be more than welcome to, but you would be wasting your time.
I quite personally could care less if you wish to go through this
process or not.  The whole point here is in what we want to support.

> > Really, everybody is just up in arms over a knee-jerk reaction to not
> > reading carefully.  What it boils down to is either not knowing the
> > facts, or trolling/flaming.
> 
> Why exactly is evaporating stage1 an ultimate goal here (as it seems to me?).

It's usefulness is far outweighed by the problems it causes, and it is
really no longer necessary, nor has it been for over a year now.

> So don't support it, but why it should not exist?

I'll explain this just once.  If we release it, we are expected to
support it.  There are *tons* of examples of things we won't do because
we don't want the headache of supporting it.  Why should this be any
different?

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re[4]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 2621 bytes --]


22.11.2005, 21:58:50, Chris Gianelloni wrote:

>> That FAQ section has nothing in common with the original stage1 docs. Sorry,
>> installing stage3 to remove all the use flags cruft subsequently, bootstrap
>> and re-emerge the system and then ponder which packages are not needed any
>> more (again, there's no reliable tool to remove unneeded stuff from system,
>> I've already mentioned this once) - hmmm... :/

> No.  That FAQ section is there to describe how to install from a stage1
> or stage2 tarball and has nothing to do with a stage3 tarball, nor did I
> ever say that it would.  I'm not sure I understand what you're getting
> at here.

Uhm, do I really need to quote it here?

<snip>
"How do I Install Gentoo Using a Stage1 or Stage2 Tarball?

...

However, Gentoo still provides stage1 and stage2 tarballs. This is for
development purposes (the Release Engineering team starts from a stage1 tarball
to obtain a stage3) but shouldn't be used by users: a stage3 tarball can very
well be used to bootstrap the system."
</snip>

Sorry, but that does not answer the original FAQ question at all...
The above does not describe a stage1 install, but a workaround procedure you've
invented because of your strong dislike of stage1 install. However much you
say the result is the same, it's not. E.g. - how exactly I get rid of those
unneeded packages once I've changed the use flags, bootstrapped and rebuilt the
system? Honestly, stage3 is something I don't find useful for a server install
because the default use flags are aimed at desktop systems.

Sure, I can use hardened stage3, compiled for i386 and enjoy the Debian
feeling. ;p

> The whole point here is in what we want to support.

So don't support it, but let it exist!

>> Why exactly is evaporating stage1 an ultimate goal here (as it seems to me?).

> It's usefulness is far outweighed by the problems it causes, and it is
> really no longer necessary, nor has it been for over a year now.

Uhm, I've seen quite a couple of examples in this debate why it is still
necessary and useful.

>> So don't support it, but why it should not exist?

> I'll explain this just once.  If we release it, we are expected to
> support it.  There are *tons* of examples of things we won't do because
> we don't want the headache of supporting it.  Why should this be any
> different?

sigh... You are not required to support it - exactly like you are not expected
or required to support gcc-4 and gcc-4.1 and you can mark any bugs about it as
INVALID (happens every day, quite frankly).


-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: Re[4]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 5570 bytes --]

On Tue, 2005-11-22 at 22:36 +0100, Jakub Moc wrote:
> 22.11.2005, 21:58:50, Chris Gianelloni wrote:
> 
> >> That FAQ section has nothing in common with the original stage1 docs. Sorry,
> >> installing stage3 to remove all the use flags cruft subsequently, bootstrap
> >> and re-emerge the system and then ponder which packages are not needed any
> >> more (again, there's no reliable tool to remove unneeded stuff from system,
> >> I've already mentioned this once) - hmmm... :/
> 
> > No.  That FAQ section is there to describe how to install from a stage1
> > or stage2 tarball and has nothing to do with a stage3 tarball, nor did I
> > ever say that it would.  I'm not sure I understand what you're getting
> > at here.
> 
> Uhm, do I really need to quote it here?

Not really, but you're going to do it anyway.

> <snip>
> "How do I Install Gentoo Using a Stage1 or Stage2 Tarball?
> 
> ...
> 
> However, Gentoo still provides stage1 and stage2 tarballs. This is for
> development purposes (the Release Engineering team starts from a stage1 tarball
> to obtain a stage3) but shouldn't be used by users: a stage3 tarball can very
> well be used to bootstrap the system."
> </snip>
> 
> Sorry, but that does not answer the original FAQ question at all...

Umm... yeah.  So you snip it RIGHT BEFORE THE INSTALL INSTRUCTIONS...
Good show... *rolleyes*

> The above does not describe a stage1 install, but a workaround procedure you've
> invented because of your strong dislike of stage1 install. However much you
> say the result is the same, it's not. E.g. - how exactly I get rid of those
> unneeded packages once I've changed the use flags, bootstrapped and rebuilt the
> system? Honestly, stage3 is something I don't find useful for a server install
> because the default use flags are aimed at desktop systems.

emerge -e world && emerge -e world && emerge depclean

This was already answered for you.  Your refusal to accept the answer,
is not my problem.

I'm tired of arguing with you.  You refuse to listen to what I am
saying.  A properly maintained and built system will be identical to one
built from a stage1 tarball.  You cannot argue this point just because
you do not personally know how to do it.  I have already said that we
are working on documenting the process for the users.  This will be done
well before we ever remove a stage1 or stage2 tarball from the mirrors.

> Sure, I can use hardened stage3, compiled for i386 and enjoy the Debian
> feeling. ;p

You can do whatever you like.  Nobody is forcing you to do anything.

That being said, you are not going to force *me* to do anything, either.

> > The whole point here is in what we want to support.
> 
> So don't support it, but let it exist!

Why?  Why would I even bother distributing something that is not worth
distributing?  We don't want testing on them.  We know they are broken.
We don't want users using it.  We know it is broken.  What purpose is
served by putting out something that we KNOW is broken and have no
intentions on fixing due to it being broken BY DESIGN?

> >> Why exactly is evaporating stage1 an ultimate goal here (as it seems to me?).
> 
> > It's usefulness is far outweighed by the problems it causes, and it is
> > really no longer necessary, nor has it been for over a year now.
> 
> Uhm, I've seen quite a couple of examples in this debate why it is still
> necessary and useful.

No.  You really haven't.  You might think that you have, but you have
not.  We also are not advocating anything for either Hardened or
Embedded.  They are their own projects with their own Release Engineers
and their own support infrastructure.  If they want to support a stage1
tarball until the Sun explodes, I don't care.

> >> So don't support it, but why it should not exist?
> 
> > I'll explain this just once.  If we release it, we are expected to
> > support it.  There are *tons* of examples of things we won't do because
> > we don't want the headache of supporting it.  Why should this be any
> > different?
> 
> sigh... You are not required to support it - exactly like you are not expected
> or required to support gcc-4 and gcc-4.1 and you can mark any bugs about it as
> INVALID (happens every day, quite frankly).

Look.  I don't care what you think I should do.  I really don't.  You
can argue this point until you're blue in the face, but until I see you
volunteering to do THE WORK you really have no say.  This really is
something that is an internal decision to Release Engineering.  We have
discussed it and we're in agreement here.  Now, the one thing that I've
not seen *anyone* here do is step up to help with any of this.  Instead,
all I see is flames, name calling, and other useless arguments.  We
decided that we do not want to put out unsupported, known broken, crap.

Do you really not understand the fact that we are making an attempt to
improve the quality of our distribution.  We are trying to improve the
end user experience.  We have already seen that users are not following
the documentation, as it is.  The Handbook keeps growing in size and
complexity, and there's no end in sight.  All the while, the quality is
going to shit because we crossed the line where we can feasibly test
what we're producing a long, LONG time ago.  You're more than welcome to
argue this for as long as you want, but I am done.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re[6]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 3534 bytes --]


23.11.2005, 0:26:03, Chris Gianelloni wrote:

>> However, Gentoo still provides stage1 and stage2 tarballs. This is for
>> development purposes (the Release Engineering team starts from a stage1
>> tarball to obtain a stage3) but shouldn't be used by users: a stage3 tarball
>> can very well be used to bootstrap the system." </snip>
>> 
>> Sorry, but that does not answer the original FAQ question at all...

> Umm... yeah.  So you snip it RIGHT BEFORE THE INSTALL INSTRUCTIONS...
> Good show... *rolleyes*

I can summarize those "ommited" instructions for you, looks pretty much like
this: How do I make a cup of coffee? Uhm, you first make a cup of tea, then
pour it out into the kitchen sink, and then make your coffee.

> emerge -e world && emerge -e world && emerge depclean

You've missed revdep-rebuild to fix the borkage that emerge depclean produced. ;)

>> Sure, I can use hardened stage3, compiled for i386 and enjoy the Debian
>> feeling. ;p

> You can do whatever you like.  Nobody is forcing you to do anything.

> That being said, you are not going to force *me* to do anything, either.

Hmm, have I missed an argument here? Actually, the above is incorrect. You
*are* forcing me to use stage3, but whatever... after all I still have the nice
choice to not use GRP, as already mentioned previously, so no need to complain.

> Look.  I don't care what you think I should do.  I really don't.  You can
> argue this point until you're blue in the face, but until I see you
> volunteering to do THE WORK you really have no say.  This really is something
> that is an internal decision to Release Engineering.  We have discussed it
> and we're in agreement here.  Now, the one thing that I've not seen *anyone*
> here do is step up to help with any of this.  Instead, all I see is flames,
> name calling, and other useless arguments.  We decided that we do not want to
> put out unsupported, known broken, crap.

> Do you really not understand the fact that we are making an attempt to
> improve the quality of our distribution.  We are trying to improve the end
> user experience.  We have already seen that users are not following the
> documentation, as it is.  The Handbook keeps growing in size and complexity,
> and there's no end in sight.  All the while, the quality is going to shit
> because we crossed the line where we can feasibly test what we're producing a
> long, LONG time ago.  You're more than welcome to argue this for as long as
> you want, but I am done.

<sarcasm class="strong">
   Yeah, as I see it, this will only reach the acceptable quality when it goes
GLI click click click way, of course also additionally hiding the dangerous use
flags from users so that they cannot possibly break anything when installing,
since they don't read the instruction properly. By that time, most of the
people who cared will have switched to LFS, and the rest won't mind really. And
additionally, this might attract a considerable Manra^^Wiva user base, so
everyone will benefit. ;p
</sarcasm>

*Now* I hope I've finally been sarcastic enough to justify the incredibly
pissed-off tone you've shown in your previous reply. I've not exactly seen any
flames or name calling here, and I'm not the one to blame for the fact
that you're feeling overloaded. Jump back in when you are in more constructive
mood. With this level of irritation caused by anyone who does not jump happily
on stage1 grave, the debate lacks any sense. Bleh...

-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Donnie Berkholz]

Jakub Moc wrote:
> *Now* I hope I've finally been sarcastic enough to justify the incredibly
> pissed-off tone you've shown in your previous reply. I've not exactly seen any
> flames or name calling here, and I'm not the one to blame for the fact
> that you're feeling overloaded. Jump back in when you are in more constructive
> mood. With this level of irritation caused by anyone who does not jump happily
> on stage1 grave, the debate lacks any sense. Bleh...

I hope you realize that pissing people off is a really terrible way to 
get them to change their minds. It's more of a way to make them lose 
motivation and quit doing what they're doing entirely.

Donnie
-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Mark Loeser]

[-- Attachment #1: Type: text/plain, Size: 1027 bytes --]

Jakub Moc <jakub@gentoo.org> said:
> > You can do whatever you like.  Nobody is forcing you to do anything.
> 
> > That being said, you are not going to force *me* to do anything, either.
> 
> Hmm, have I missed an argument here? Actually, the above is incorrect. You
> *are* forcing me to use stage3, but whatever... after all I still have the nice
> choice to not use GRP, as already mentioned previously, so no need to complain.

Actually, he's not forcing you to do anything since the stage1 and stage2
tarballs are still being produced.  If you want to use them, then you can
download them and go ahead with your installation.  If you don't understand the
process and what you are doing already, then you shouldn't be using them.

That being said, I have not seen anyone here offer to support these installation
methods.  Its nice and all to say we should keep something, but the releng team
cannot support everything currently.  If you want to "save the stage1", then
step up and offer to help.

Mark

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]

Re: Re[6]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[George Prowse]

No stage 1 or 2? Excellent. People seem to wear Stage 1 like a badge
of honour but never seem to realise that the earlier you play around
with your system the deeper the mistakes.

Why are the instructions for the stage 1 and 2 in the handbook at all?
They should be in Gentoo dev space and the only way new users find out
about it should be through word of mouth. The people who *really*
need/want a stage 1 will have the knowlege to ask releng about it

George

-- 
gentoo-dev@gentoo.org mailing list

Re: Re[6]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Paul de Vrieze]

[-- Attachment #1: Type: text/plain, Size: 1081 bytes --]

On Wednesday 23 November 2005 01:55, Jakub Moc wrote:
> > emerge -e world && emerge -e world && emerge depclean
>
> You've missed revdep-rebuild to fix the borkage that emerge depclean
> produced. ;)

After double rebuilding of the complete world I would seriously doubt it 
that any stray dependencies were still around. If there were, it would be 
because of broken ebuilds. That should be reported as bugs and fixed 
instead of relying on revdep-rebuild.

About revdep-rebuild, that is an ugly kludge that is only there to 
aleviate that portage does not record the metadata needed to know what 
should be rebuilded without having lib files searched. In 99% of the 
cases it is also not needed to use it, and those people advocating its 
use should probably put their attention into fixing portage that it 
records and uses information about what package was used to satisfy what 
dependency. (and to verify the package tree state after each install)

Paul

-- 
Paul de Vrieze
Gentoo Developer
Mail: pauldv@gentoo.org
Homepage: http://www.devrieze.net

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re[8]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 924 bytes --]


23.11.2005, 11:25:58, Paul de Vrieze wrote:

> On Wednesday 23 November 2005 01:55, Jakub Moc wrote:
>> > emerge -e world && emerge -e world && emerge depclean
>>
>> You've missed revdep-rebuild to fix the borkage that emerge depclean
>> produced. ;)

> After double rebuilding of the complete world I would seriously doubt it 
> that any stray dependencies were still around. If there were, it would be 
> because of broken ebuilds. That should be reported as bugs and fixed 
> instead of relying on revdep-rebuild.

You've probably missed the point. It's emerge depclean that's broken; again -
we are lacking any reliable way to punt unneded packages.

BTW, I'd still like to know how I'll get nptl(only) hardened install once
stage1 is gone. i386 does not have nptl, and I've done change CHOST && emerge
-e system && emerge -e world job a couple of times and it never went smoothly.


-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: Re[8]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Ned Ludd]

On Wed, 2005-11-23 at 12:06 +0100, Jakub Moc wrote:

> BTW, I'd still like to know how I'll get nptl(only) hardened install once
> stage1 is gone. i386 does not have nptl, and I've done change CHOST && emerge
> -e system && emerge -e world job a couple of times and it never went smoothly.

Please calm down. Chances are for 2006.0 hardened will no longer produce
a set of 2.4.x stages, by providing 1 less set of stages it frees up
some 
of our mirror space for providing a set if i386-gentoo-linux-gnu and a 
set of i686-gentoo-linux-gnu set of stages. If your start from a stage1
and remove the hardened USE flags it's functionality the equivalent as 
starting from a vanilla set by the time you ./bootstrap.sh and then 
your emerge -e world would remove any remaining traces. Of course I 
think it's silly to remove the hardened USE flag.


Please calm down. Chances are for 2006.0 hardened will no longer produce
a set of 2.4.x stages, by providing 1 less set of stages it frees up
some of our mirror space for providing a set if i386-gentoo-linux-gnu
and a set of i686-gentoo-linux-gnu set of stages. If your start from a 
stage1 and remove the hardened USE flags it's functionality the
equivalent as starting from a vanilla set by the time you ./bootstrap.sh
and then your emerge -e world would remove any remaining traces. Of
course I think it's silly to remove the hardened USE flag.

You can have your cake and eat it too as long as catalyst support
remains.

-- 
Ned Ludd <solar@gentoo.org>
Gentoo Linux

-- 
gentoo-dev@gentoo.org mailing list

Re: Re[8]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Ned Ludd]

On Wed, 2005-11-23 at 08:01 -0500, Ned Ludd wrote:
> On Wed, 2005-11-23 at 12:06 +0100, Jakub Moc wrote:
> 
> > BTW, I'd still like to know how I'll get nptl(only) hardened install once
> > stage1 is gone. i386 does not have nptl, and I've done change CHOST && emerge
> > -e system && emerge -e world job a couple of times and it never went smoothly.
> 
> Please calm down. Chances are for 2006.0 hardened will no longer produce
> a set of 2.4.x stages, by providing 1 less set of stages it frees up
> some 
> of our mirror space for providing a set if i386-gentoo-linux-gnu and a 
> set of i686-gentoo-linux-gnu set of stages. If your start from a stage1
> and remove the hardened USE flags it's functionality the equivalent as 
> starting from a vanilla set by the time you ./bootstrap.sh and then 
> your emerge -e world would remove any remaining traces. Of course I 
> think it's silly to remove the hardened USE flag.
> 
> 
> Please calm down. Chances are for 2006.0 hardened will no longer produce
> a set of 2.4.x stages, by providing 1 less set of stages it frees up
> some of our mirror space for providing a set if i386-gentoo-linux-gnu
> and a set of i686-gentoo-linux-gnu set of stages. If your start from a 
> stage1 and remove the hardened USE flags it's functionality the
> equivalent as starting from a vanilla set by the time you ./bootstrap.sh
> and then your emerge -e world would remove any remaining traces. Of
> course I think it's silly to remove the hardened USE flag.
> 
> You can have your cake and eat it too as long as catalyst support
> remains.


ha ok new rule for me. drink coffee before sending first mail of the
morn.


> -- 
> Ned Ludd <solar@gentoo.org>
> Gentoo Linux
> 
-- 
Ned Ludd <solar@gentoo.org>
Gentoo Linux

-- 
gentoo-dev@gentoo.org mailing list

Re: Re[8]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Henrik Brix Andersen]

[-- Attachment #1: Type: text/plain, Size: 421 bytes --]

On Wed, Nov 23, 2005 at 08:19:40AM -0500, Ned Ludd wrote:
> ha ok new rule for me. drink coffee before sending first mail of the
> morn.

What's this supposed to imply? That you didn't intend to repeat
yourself in the original mail - or that hardened will continue to ship
2.4.x stages for 2006.0+? ;)

Regards,
Brix
-- 
Henrik Brix Andersen <brix@gentoo.org>
Gentoo Metadistribution | Mobile computing herd

[-- Attachment #2: Type: application/pgp-signature, Size: 211 bytes --]

Re: Re[8]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[solar]

On Wed, 2005-11-23 at 16:57 +0100, Henrik Brix Andersen wrote:
> On Wed, Nov 23, 2005 at 08:19:40AM -0500, Ned Ludd wrote:
> > ha ok new rule for me. drink coffee before sending first mail of the
> > morn.
> 
> What's this supposed to imply? That you didn't intend to repeat
> yourself in the original mail - or that hardened will continue to ship
> 2.4.x stages for 2006.0+? ;)

That I did not intend to repeat myself. Before hitting send I usually 
take whatever mail copy and paste it into $EDITOR then wrap it at 72 
chars wide and delete the unwrapped block. Clearly I forgot the last 
step. As for hardened and 2.4.x it seems most of our users are wanting 
2.6.x now and unless users/devs show interest I can't really see us 
needing to produce a new set of 2.4.x based 2006.x stages.

-- 
solar <solar@gentoo.org>
Gentoo Linux

-- 
gentoo-dev@gentoo.org mailing list

Re: Re[8]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Paul de Vrieze]

[-- Attachment #1: Type: text/plain, Size: 1801 bytes --]

On Wednesday 23 November 2005 12:06, Jakub Moc wrote:
> 23.11.2005, 11:25:58, Paul de Vrieze wrote:
> > On Wednesday 23 November 2005 01:55, Jakub Moc wrote:
> >> > emerge -e world && emerge -e world && emerge depclean
> >>
> >> You've missed revdep-rebuild to fix the borkage that emerge depclean
> >> produced. ;)
> >
> > After double rebuilding of the complete world I would seriously doubt
> > it that any stray dependencies were still around. If there were, it
> > would be because of broken ebuilds. That should be reported as bugs
> > and fixed instead of relying on revdep-rebuild.
>
> You've probably missed the point. It's emerge depclean that's broken;
> again - we are lacking any reliable way to punt unneded packages.

Emerge depclean is very reliable in its behaviour. That behaviour is not 
always what is desired though. When however all packages on the system 
are consistent with the present USEFLAGS and eachother, depclean does 
exactly what it should do. The fault is that it asumes that packages have 
been build with the current environment. This is generally not true 
(thanks to for example auto-use, missing dependencies, and configure 
script automagic).

> BTW, I'd still like to know how I'll get nptl(only) hardened install
> once stage1 is gone. i386 does not have nptl, and I've done change
> CHOST && emerge -e system && emerge -e world job a couple of times and
> it never went smoothly.

Probably what you want is to create static bash gcc binutils etc. stuff 
like in stage1. With those one can change the c library at hearts 
content. I never tried it, but I believe that there is a USE flag for 
building them that way.

Paul

-- 
Paul de Vrieze
Gentoo Developer
Mail: pauldv@gentoo.org
Homepage: http://www.devrieze.net

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 4112 bytes --]

On Tue, 2005-11-22 at 14:47 +0000, Kurt Lieber wrote:
> We have received *numerous* complaints from users about the decision to
> remove stage 1 and 2 from the installation documentation.  I realize it's
> still available if users are willing to dig for it, but not all users do.
> 
> In my years of monitoring www@gentoo.org, we've received the most
> complaints about this decision than any other single decision.  Is there a
> way we can re-introduce the stages into the installation documentation,
> perhaps with gigantic warnings saying, "for advanced users only" or "use at
> your own risk"?

The problem is that we (releng) cannot possibly keep up with the number
of possible bugs that are being introduced via USE flags.  It used to be
that if someone introduced a USE flag into *any* package that would show
up under "system" that they would make sure the damn thing would pass a
stage1->stage3 process.  Now, we're receiving bugs and emails quite
often from problems where things like "hal" are being pulled in to
system, which is a major problem, as it requires a configured kernel,
which, of course, doesn't exist at this point.

As I am now not only the Release Engineering lead, but also the x86
Release Coordinator, I am fielding nearly 100% of these issues.

I DO NOT HAVE THE TIME TO DO OTHER PEOPLE'S QA FOR THEM.

Because of this, *I* requested to have the instructions removed.  They
were causing more problems than they are worth, and since *not a single
person* stepped up when I asked for help after beejay left, I'm just
going to do what I need to do with the things that I maintain.  If this
means requesting Handbook changes to reduce my workload, I have and will
continue this trend.

Personally, I would like to see stage1 and stage2 go away completely.
They serve no real purpose anymore after the changes we have made to the
stages to include a complete /var/db before 2005.0's release.  They take
longer to use for installation, and give you exactly 0 advantages over a
stage3 that cannot be done with a stage3 tarball itself.  I would have
no problem with us documenting these more advanced methods somewhere,
but I would have a definite problem with resurrecting the obsolete
materials just because a few users that are ignorant to the actual
issues are flaming and otherwise provoking www@g.o with this.

Besides, there's *nothing* stopping a user from continuing to use a
stage1 tarball.  There's *nothing* stopping a user from taking a stage3
tarball, the example catalyst specs, and building their own stage1
tarball.  We aren't taking away their "freedom" in any way.  However,
anything that we release, we *are* expected to do QA on and make sure it
works, along with resolving bugs.  Almost all of these bugs are
user-created due to their lack of knowledge of USE flags, Gentoo in
general, and the bootstrap process.  We cannot expect every user that
might think about using a stage1 tarball to know this.  That means
they'll be filing bugs.  I'll be getting them.  I came up with a
resolution for these bugs and enacted it.  While it will not prevent the
problem 100%, it will reduce my workload greatly.

I truly do appreciate and adore our users, but if a few people are going
to get pissed off and leave over this.  Fine.  Let them.  They're
probably not the kind of people we want associated with us anyway.

> ----- Forwarded message from Varun Dhussa <varun.dhussa@gmail.com> -----
> 
> Date: Tue, 22 Nov 2005 19:50:07 +0530
> From: Varun Dhussa <varun.dhussa@gmail.com>
> To: www@gentoo.org
> Subject: Complaint
> 
> Hello,
> 
> Gentoo claims to be giving freedom. However, I was dissapointed to see that
> the stage 1 had been removed from gentoo 2005. Infact, even the handbook
> makes no refference of it. This takes Gentoo another step closer to other
> distros like Ubuntu.
> 
> A dissapointed user,
> Varun Dhussa
> India
> 
> ----- End forwarded message -----

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Henrik Brix Andersen]

[-- Attachment #1: Type: text/plain, Size: 1004 bytes --]

On Tue, Nov 22, 2005 at 10:10:14AM -0500, Chris Gianelloni wrote:
> Personally, I would like to see stage1 and stage2 go away completely.
> They serve no real purpose anymore after the changes we have made to the
> stages to include a complete /var/db before 2005.0's release.  They take
> longer to use for installation, and give you exactly 0 advantages over a
> stage3 that cannot be done with a stage3 tarball itself.

Perhaps we should just document these facts better? Write a news item
explaining the reasons behind removing stage1/2 instructions from our
handbooks - and explain _why_ stage1 and 2 are not really better than
a stage3 install + bootstrap.sh + emerge -e world. Perhaps also add a
small note about this to the handbook.

If our users are explained why stage1/2 installs don't give any
benefits over a stage3 install, I trust them to aknowledge this fact.

Regards,
Brix
-- 
Henrik Brix Andersen <brix@gentoo.org>
Gentoo Metadistribution | Mobile computing herd

[-- Attachment #2: Type: application/pgp-signature, Size: 211 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1288 bytes --]

On Tue, 2005-11-22 at 16:26 +0100, Henrik Brix Andersen wrote:
> On Tue, Nov 22, 2005 at 10:10:14AM -0500, Chris Gianelloni wrote:
> > Personally, I would like to see stage1 and stage2 go away completely.
> > They serve no real purpose anymore after the changes we have made to the
> > stages to include a complete /var/db before 2005.0's release.  They take
> > longer to use for installation, and give you exactly 0 advantages over a
> > stage3 that cannot be done with a stage3 tarball itself.
> 
> Perhaps we should just document these facts better? Write a news item
> explaining the reasons behind removing stage1/2 instructions from our
> handbooks - and explain _why_ stage1 and 2 are not really better than
> a stage3 install + bootstrap.sh + emerge -e world. Perhaps also add a
> small note about this to the handbook.

Well, the information was in the bug linked with the GWN article, along
with also in the gentoo-docs mailing list archive that was posted.

> If our users are explained why stage1/2 installs don't give any
> benefits over a stage3 install, I trust them to aknowledge this fact.

Your faith is much stronger than mine.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Lance Albertson]

[-- Attachment #1: Type: text/plain, Size: 2253 bytes --]

Chris Gianelloni wrote:
> On Tue, 2005-11-22 at 14:47 +0000, Kurt Lieber wrote:
> 
>>We have received *numerous* complaints from users about the decision to
>>remove stage 1 and 2 from the installation documentation.  I realize it's
>>still available if users are willing to dig for it, but not all users do.
>>
>>In my years of monitoring www@gentoo.org, we've received the most
>>complaints about this decision than any other single decision.  Is there a
>>way we can re-introduce the stages into the installation documentation,
>>perhaps with gigantic warnings saying, "for advanced users only" or "use at
>>your own risk"?
> 
> 
> The problem is that we (releng) cannot possibly keep up with the number
> of possible bugs that are being introduced via USE flags.  It used to be
> that if someone introduced a USE flag into *any* package that would show
> up under "system" that they would make sure the damn thing would pass a
> stage1->stage3 process.  Now, we're receiving bugs and emails quite
> often from problems where things like "hal" are being pulled in to
> system, which is a major problem, as it requires a configured kernel,
> which, of course, doesn't exist at this point.
> 
> As I am now not only the Release Engineering lead, but also the x86
> Release Coordinator, I am fielding nearly 100% of these issues.
> 
> I DO NOT HAVE THE TIME TO DO OTHER PEOPLE'S QA FOR THEM.

If you are that overworked, perhaps you should find more people to help
with releng and the duties you have? I've been in a similar position as
yourself where its hard to find good quality folks that stick around,
and then you get used to doing everything yourself. I haven't seen any
emails asking for help or people for releng. This is the first I've
heard of your troubles with not having adequate time for all the duties
you do yourself.

If you need help, please ask for it and at least try and get some of
that load off of you so that we don't take things out of gentoo simply
because of the lack of time.

Cheers-

-- 
Lance Albertson <ramereth@gentoo.org>
Gentoo Infrastructure | Operations Manager

---
GPG Public Key:  <http://www.ramereth.net/lance.asc>
Key fingerprint: 0423 92F3 544A 1282 5AB1  4D07 416F A15D 27F4 B742

ramereth/irc.freenode.net

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 186 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Andrew Gaffney]

Lance Albertson wrote:
> If you are that overworked, perhaps you should find more people to help
> with releng and the duties you have? I've been in a similar position as
> yourself where its hard to find good quality folks that stick around,
> and then you get used to doing everything yourself. I haven't seen any
> emails asking for help or people for releng. This is the first I've
> heard of your troubles with not having adequate time for all the duties
> you do yourself.
> 
> If you need help, please ask for it and at least try and get some of
> that load off of you so that we don't take things out of gentoo simply
> because of the lack of time.

Apparently you missed his email to -core on the 27th of October with the subject 
"x86 Release Coordinator" where he asks for someone to step up to take beejay's 
old job...the one that he's doing now.

-- 
Andrew Gaffney                            http://dev.gentoo.org/~agaffney/
Gentoo Linux Developer                                   Installer Project

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Lance Albertson]

[-- Attachment #1: Type: text/plain, Size: 1738 bytes --]

Andrew Gaffney wrote:
> Lance Albertson wrote:
> 
>> If you are that overworked, perhaps you should find more people to help
>> with releng and the duties you have? I've been in a similar position as
>> yourself where its hard to find good quality folks that stick around,
>> and then you get used to doing everything yourself. I haven't seen any
>> emails asking for help or people for releng. This is the first I've
>> heard of your troubles with not having adequate time for all the duties
>> you do yourself.
>>
>> If you need help, please ask for it and at least try and get some of
>> that load off of you so that we don't take things out of gentoo simply
>> because of the lack of time.
> 
> 
> Apparently you missed his email to -core on the 27th of October with the
> subject "x86 Release Coordinator" where he asks for someone to step up
> to take beejay's old job...the one that he's doing now.

Yeah, my bad. I guess I was just looking at the -dev mailing list since
it has a larger reader-base. I still think he should have sent that to
-dev to get more input. I understand that he probably was looking
internally for someone, but its best to keep your options open. You
never know what person you might find out of the blue to help with the
project.

Also, I noticed there's no request for this position on the [1] staffing
needs page. So you may want to include something there as well.

My bad on missing the email on -core.

[1] http://www.gentoo.org/proj/en/devrel/staffing-needs/

-- 
Lance Albertson <ramereth@gentoo.org>
Gentoo Infrastructure | Operations Manager

---
GPG Public Key:  <http://www.ramereth.net/lance.asc>
Key fingerprint: 0423 92F3 544A 1282 5AB1  4D07 416F A15D 27F4 B742

ramereth/irc.freenode.net

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 186 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 1767 bytes --]

On Tue, 2005-11-22 at 09:54 -0600, Lance Albertson wrote:
> > As I am now not only the Release Engineering lead, but also the x86
> > Release Coordinator, I am fielding nearly 100% of these issues.
> > 
> > I DO NOT HAVE THE TIME TO DO OTHER PEOPLE'S QA FOR THEM.
> 
> If you are that overworked, perhaps you should find more people to help
> with releng and the duties you have? I've been in a similar position as
> yourself where its hard to find good quality folks that stick around,
> and then you get used to doing everything yourself. I haven't seen any
> emails asking for help or people for releng. This is the first I've
> heard of your troubles with not having adequate time for all the duties
> you do yourself.

Really?  Did you read -core on October 27th?

Also, the problem is not so much needing manpower for testing as far as
Release Engineering is concerned.  It is instead having some method in
place where devs actually perform QA on their own packages.  A prime
example of this is bug #110383.  I was always under the impression that
if you were adding a flag to a package that affected "system" that it
was your responsibility to ensure that "system" still works, rather than
passing it off onto the Release Engineering team.  Now, I don't know
what package it is that is pulling in hal for this user, so it most
likely is not hal's fault, but it illustrates the point perfectly.

> If you need help, please ask for it and at least try and get some of
> that load off of you so that we don't take things out of gentoo simply
> because of the lack of time.

I did.

I got exactly *0* responses.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Lance Albertson]

[-- Attachment #1: Type: text/plain, Size: 1587 bytes --]

Chris Gianelloni wrote:
> On Tue, 2005-11-22 at 09:54 -0600, Lance Albertson wrote:
> 
>>>As I am now not only the Release Engineering lead, but also the x86
>>>Release Coordinator, I am fielding nearly 100% of these issues.
>>>
>>>I DO NOT HAVE THE TIME TO DO OTHER PEOPLE'S QA FOR THEM.
>>
>>If you are that overworked, perhaps you should find more people to help
>>with releng and the duties you have? I've been in a similar position as
>>yourself where its hard to find good quality folks that stick around,
>>and then you get used to doing everything yourself. I haven't seen any
>>emails asking for help or people for releng. This is the first I've
>>heard of your troubles with not having adequate time for all the duties
>>you do yourself.
> 
> 
> Really?  Did you read -core on October 27th?

Yeah, as i just sent, My bad on missing that. Sorry about not catching that.

>>If you need help, please ask for it and at least try and get some of
>>that load off of you so that we don't take things out of gentoo simply
>>because of the lack of time.
> 
> 
> I did.
> 
> I got exactly *0* responses. 

Only thing I have to say to that is try sending that email on -dev and
try putting it up on the staffing needs page. That way its documented in
a public form and other people will see it.

No need to start a flame war about this :-)

Cheers-

-- 
Lance Albertson <ramereth@gentoo.org>
Gentoo Infrastructure | Operations Manager

---
GPG Public Key:  <http://www.ramereth.net/lance.asc>
Key fingerprint: 0423 92F3 544A 1282 5AB1  4D07 416F A15D 27F4 B742

ramereth/irc.freenode.net

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 186 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 2023 bytes --]

On Tue, 2005-11-22 at 11:15 -0600, Lance Albertson wrote:
> Chris Gianelloni wrote:
> > On Tue, 2005-11-22 at 09:54 -0600, Lance Albertson wrote:
> > 
> >>>As I am now not only the Release Engineering lead, but also the x86
> >>>Release Coordinator, I am fielding nearly 100% of these issues.
> >>>
> >>>I DO NOT HAVE THE TIME TO DO OTHER PEOPLE'S QA FOR THEM.
> >>
> >>If you are that overworked, perhaps you should find more people to help
> >>with releng and the duties you have? I've been in a similar position as
> >>yourself where its hard to find good quality folks that stick around,
> >>and then you get used to doing everything yourself. I haven't seen any
> >>emails asking for help or people for releng. This is the first I've
> >>heard of your troubles with not having adequate time for all the duties
> >>you do yourself.
> > 
> > 
> > Really?  Did you read -core on October 27th?
> 
> Yeah, as i just sent, My bad on missing that. Sorry about not catching that.

I'd responded before seeing your response.

> >>If you need help, please ask for it and at least try and get some of
> >>that load off of you so that we don't take things out of gentoo simply
> >>because of the lack of time.
> > 
> > 
> > I did.
> > 
> > I got exactly *0* responses. 
> 
> Only thing I have to say to that is try sending that email on -dev and
> try putting it up on the staffing needs page. That way its documented in
> a public form and other people will see it.

Actually, I've had a volunteer come forward from among the developer
ranks, thunder.  I'm going to try to get him up-to-speed for the
upcoming release.  The main reason why I did not send this to -dev
before was that I am already at a stretch for time, and do not have time
to bring someone through the recruitment process for this.

> No need to start a flame war about this :-)

Definitely not.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re[2]: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Jakub Moc]

[-- Attachment #1: Type: text/plain, Size: 963 bytes --]


22.11.2005, 17:30:50, Chris Gianelloni wrote:

> On Tue, 2005-11-22 at 09:54 -0600, Lance Albertson wrote:


> Also, the problem is not so much needing manpower for testing as far as
> Release Engineering is concerned.  It is instead having some method in
> place where devs actually perform QA on their own packages.  A prime
> example of this is bug #110383.  I was always under the impression that
> if you were adding a flag to a package that affected "system" that it
> was your responsibility to ensure that "system" still works, rather than
> passing it off onto the Release Engineering team.  Now, I don't know
> what package it is that is pulling in hal for this user, so it most
> likely is not hal's fault, but it illustrates the point perfectly.

Blame vapier for that one ;p (Bug 99533 half-fixed for ~arch only);
alternatively, you can blame usata for adding emacs support to gpm in the first
place (Bug 80217).


-- 

jakub

[-- Attachment #2: Type: application/pgp-signature, Size: 183 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Benjamin Judas]

[-- Attachment #1: Type: text/plain, Size: 1560 bytes --]

And again, we have the same situation that lead to my resignation:
People who have absolutely no clue of how releng works scream. Not about
"BAD QA!!!11" this time, but about a decision that was made to make work
easier.

What is wrong with you, guys? You all have so good and enlightening
ideas! Why does nobody want to be my successor? Come on - my footprints
aren't that big!

Seriously, the reaction of some developers and users smells a bit like
Slashdot to me: "OMG! They are removing the 133735t part of gent00!!!".
Sounds silly, right? Those people that read the install-docs *carefully*
are the only ones being  r e a d y  to do a stage1-install. If they read
carefully they will see the link "Hey, you can still do stage1 - if you
want to then read-on over here".

With some exceptions, most of the people replying to this thread
obviously have no clue what tracking down a bug in a stage1 installation
means: It's a lottery - especially if the user has nearly no knowledge
of the backgrounds and can only tell you: "!!! ERROR in foo/bar-0.8.15".

I'd like all of you to keep this in mind. Gentoo isn't a
garage-distribution anymore and I tend to say (though I don't know the
exact number of received mails) that those who complained at www@ didn't
read the whole story about the "removal" and are really a minority.

Also, why a GLEP for that? A GLEP for removing something from the
handbook? Wow! Bureaucracy-wise Gentoo seems to get more and more
european.

Regards

Benjamin "beejay, the former x86-monkey" Judas

[-- Attachment #2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Henrik Brix Andersen]

[-- Attachment #1: Type: text/plain, Size: 359 bytes --]

On Tue, Nov 22, 2005 at 05:56:52PM +0100, Benjamin Judas wrote:
> Also, why a GLEP for that? A GLEP for removing something from the
> handbook? Wow! Bureaucracy-wise Gentoo seems to get more and more
> european.

You mean more and more American? ;)

./Brix
-- 
Henrik Brix Andersen <brix@gentoo.org>
Gentoo Metadistribution | Mobile computing herd

[-- Attachment #2: Type: application/pgp-signature, Size: 211 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Ciaran McCreesh]

[-- Attachment #1: Type: text/plain, Size: 651 bytes --]

On Tue, 22 Nov 2005 14:47:45 +0000 Kurt Lieber <klieber@gentoo.org>
wrote:
| In my years of monitoring www@gentoo.org, we've received the most
| complaints about this decision than any other single decision.

How many of those complaints were from users who understood the issues
involved, and how many were knee-jerk reactions from morons who thought
that there was a reason to use stage1s other than for seed stages and
to get around a nasty bug in stager (which we no longer use)?

-- 
Ciaran McCreesh : Gentoo Developer (Look! Shiny things!)
Mail            : ciaranm at gentoo.org
Web             : http://dev.gentoo.org/~ciaranm


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Andrew Muraco]

Kurt Lieber wrote:

>We have received *numerous* complaints from users about the decision to
>remove stage 1 and 2 from the installation documentation.  I realize it's
>still available if users are willing to dig for it, but not all users do.
>
>In my years of monitoring www@gentoo.org, we've received the most
>complaints about this decision than any other single decision.  Is there a
>way we can re-introduce the stages into the installation documentation,
>perhaps with gigantic warnings saying, "for advanced users only" or "use at
>your own risk"?
>
>--kurt
>
(I've read all of the comments up until now, but my response is not 
directed at any particular post.)

Facts: (according to me, and what I've read)
-The releng team DID make a good decision by making stage 3 default in 
the instructions.
-The releng team _DID_NOT_ do a sufficient job of making the community 
aware of the changes BEFORE they occurred (I didn't know about this 
change until after it was done, and Gentoo.org is my home page, I read 
the GWN)
-Stage 1 & 2 tar balls and instructions ARE available

OPINION:
- This change should be GLEP'd, as it effects everyone that installs 
Gentoo (to some degree, most do not suffer tho)
- Stage 1 SHOULD continue to be released and maintained, instructions 
clearly stating risks and LACK of SUPPORT and easily visibility from the 
install docs (which it seems it does not have (according to posts), 
although, It is perfectly clear to me.)

to the releng team: Good Decision, Bad Execution Thats whats leading to 
this entire reaction.
------------
Andrew
www.leetworks.com
tuxp3@leetworks.com

-- 
gentoo-dev@gentoo.org mailing list

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Paul de Vrieze]

[-- Attachment #1: Type: text/plain, Size: 1781 bytes --]

On Wednesday 23 November 2005 05:01, Andrew Muraco wrote:
>
> (I've read all of the comments up until now, but my response is not
> directed at any particular post.)
>
> Facts: (according to me, and what I've read)
> -The releng team DID make a good decision by making stage 3 default in
> the instructions.
> -The releng team _DID_NOT_ do a sufficient job of making the community
> aware of the changes BEFORE they occurred (I didn't know about this
> change until after it was done, and Gentoo.org is my home page, I read
> the GWN)
> -Stage 1 & 2 tar balls and instructions ARE available
>
> OPINION:
> - This change should be GLEP'd, as it effects everyone that installs
> Gentoo (to some degree, most do not suffer tho)
> - Stage 1 SHOULD continue to be released and maintained, instructions
> clearly stating risks and LACK of SUPPORT and easily visibility from
> the install docs (which it seems it does not have (according to posts),
> although, It is perfectly clear to me.)

This whole issue has been discussed on gentoo-releng@g.o before. While not 
quite as noisy as dev, everyone had their say. It is clear to me that the 
decision is right the way it was made. A GLEP wasn't necessary as it was 
discussed and approved by all involved (on the releng list).

Most people that complain are probably misinformed about the usefulness of 
stages 1 and 2. They are really only useful if you know what you're doing 
and don't really need the handbook that much. Those users should be able 
to find the alternative installation docs. I do agree however that there 
should be some link to the relevant documentation from the handbook.

Paul

-- 
Paul de Vrieze
Gentoo Developer
Mail: pauldv@gentoo.org
Homepage: http://www.devrieze.net

[-- Attachment #2: Type: application/pgp-signature, Size: 198 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Chris Gianelloni]

[-- Attachment #1: Type: text/plain, Size: 938 bytes --]

On Wed, 2005-11-23 at 10:24 +0100, Paul de Vrieze wrote:
> Most people that complain are probably misinformed about the usefulness of 
> stages 1 and 2. They are really only useful if you know what you're doing 
> and don't really need the handbook that much. Those users should be able 
> to find the alternative installation docs. I do agree however that there 
> should be some link to the relevant documentation from the handbook.

There is a link.  It was in since the docs were moved.  Also, I plan on
working with the documentation team to come up with an "Advanced
Installation Topics" type guide that will not only give information on
the lower stages, but also how to make a "stage1 install" from a stage3
tarball.  It will likely also cover things like Hardened, provided they
want it that way.

-- 
Chris Gianelloni
Release Engineering - Strategic Lead
x86 Architecture Team
Games - Developer
Gentoo Linux

[-- Attachment #2: This is a digitally signed message part --]
[-- Type: application/pgp-signature, Size: 189 bytes --]

Re: [gentoo-dev] Decision to remove stage1/2 from installation documentation

[Sven Vermeulen]

[-- Attachment #1: Type: text/plain, Size: 1007 bytes --]

On Wed, Nov 23, 2005 at 10:24:55AM +0100, Paul de Vrieze wrote:
> Most people that complain are probably misinformed about the usefulness of 
> stages 1 and 2. They are really only useful if you know what you're doing 
> and don't really need the handbook that much. Those users should be able 
> to find the alternative installation docs. I do agree however that there 
> should be some link to the relevant documentation from the handbook.

Actually the migration process did all that in one step:

- Update the Handbook
  . Remove stage1/2 instructions
  . Add a /couple/ of references to the Gentoo FAQ
- Update the Gentoo FAQ
- Update the Gentoo Handbook FAQ

Perhaps I should use the <blink>...</blink> tags more often.

Wkr,
      Sven Vermeulen

-- 
  Gentoo Foundation Trustee          |  http://foundation.gentoo.org
  Gentoo Documentation Project Lead  |  http://www.gentoo.org/proj/en/gdp
  Gentoo Council Member  

  The Gentoo Project   <<< http://www.gentoo.org >>>

[-- Attachment #2: Type: application/pgp-signature, Size: 189 bytes --]