[gentoo-project] Status update of Sunrise project?

[gentoo-project] Status update of Sunrise project?

[Ben de Groot]

Looking at the sunrise-reviewed repo, there have been no commits in
the last four months. So it looks like this project died. Can we get
an official update on the status of this project? Also, we should
probably remove this from the layman overlays list, since it is no
longer maintained.

Then we should also redirect users to the proxy-maintainers project as
the best way to contribute ebuilds.

-- 
Cheers,

Ben | yngwin
Gentoo developer

Re: [gentoo-project] Status update of Sunrise project?

[Thomas Sachau]

[-- Attachment #1: Type: text/plain, Size: 1504 bytes --]

Ben de Groot schrieb:
> Looking at the sunrise-reviewed repo, there have been no commits in
> the last four months. So it looks like this project died. Can we get
> an official update on the status of this project? Also, we should
> probably remove this from the layman overlays list, since it is no
> longer maintained.
> 
> Then we should also redirect users to the proxy-maintainers project as
> the best way to contribute ebuilds.
> 

There have been some commits to the "not-yet-reviewed" repo of sunrise
by some people, but i didnt check it, so missed pushing the updates to
the reviewed repo. Since that is done now, you should see that the last
commit was only 1 day before your question.

As an additional note: proxy-maintainers and sunrise are 2 different
projects with different goals and different tools, you cannot replace
one with the other unless you want to move all the packages in sunrise
to the main tree. proxy-maintainers is about users maintaining packages
in the _main tree_ with the help of developers, sunrise is about users
maintaining packages in the _sunrise overlay_. So unless the users finds
a dev willing to add and maintain "his" package(s) in the main tree,
there is no way to replace sunrise with proxy-maintainers.

Also, if some more devs would be willing to help, that would be nice,
being the last active dev for the sunrise overlay for some time now and
i dont like the "bus factor" :-)
-- 

Thomas Sachau
Gentoo Linux Developer


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 299 bytes --]

Re: [gentoo-project] Status update of Sunrise project?

[Ben de Groot]

On 10 April 2015 at 02:57, Thomas Sachau <tommy@gentoo.org> wrote:
> Ben de Groot schrieb:
>> Looking at the sunrise-reviewed repo, there have been no commits in
>> the last four months. So it looks like this project died. Can we get
>> an official update on the status of this project? Also, we should
>> probably remove this from the layman overlays list, since it is no
>> longer maintained.
>>
>> Then we should also redirect users to the proxy-maintainers project as
>> the best way to contribute ebuilds.
>>
>
> There have been some commits to the "not-yet-reviewed" repo of sunrise
> by some people, but i didnt check it, so missed pushing the updates to
> the reviewed repo. Since that is done now, you should see that the last
> commit was only 1 day before your question.

Good to see activity again!

> As an additional note: proxy-maintainers and sunrise are 2 different
> projects with different goals and different tools, you cannot replace
> one with the other unless you want to move all the packages in sunrise
> to the main tree. proxy-maintainers is about users maintaining packages
> in the _main tree_ with the help of developers, sunrise is about users
> maintaining packages in the _sunrise overlay_. So unless the users finds
> a dev willing to add and maintain "his" package(s) in the main tree,
> there is no way to replace sunrise with proxy-maintainers.

I know the difference. But the two projects have similar goals: review
user submitted ebuilds, and commit them to an central repo to make
them available to a wider public. I think it would be good to get more
packages from sunrise into the main gentoo repo.

Are there reasons for some of those packages to be kept out of the main repo?

> Also, if some more devs would be willing to help, that would be nice,
> being the last active dev for the sunrise overlay for some time now and
> i dont like the "bus factor" :-)

Either that, or fold it into the proxy-maintainers project, and get
those packages into the main repo.

-- 
Cheers,

Ben | yngwin
Gentoo developer

Re: [gentoo-project] Status update of Sunrise project?

[Thomas Sachau]

[-- Attachment #1: Type: text/plain, Size: 900 bytes --]

Ben de Groot schrieb:
> I know the difference. But the two projects have similar goals: review
> user submitted ebuilds, and commit them to an central repo to make
> them available to a wider public. I think it would be good to get more
> packages from sunrise into the main gentoo repo.
> 
> Are there reasons for some of those packages to be kept out of the main repo?

I think, the main reason is the required amount of developer time needed
for that move and the continued maintainence of the moved packages.

Sunrise has the advantage, that the packages per dev ratio is higher
then for packages proxied in the main tree as the user does all the work
(prepare, test, commit and update the ebuilds), the devs do just a
review of the ebuild itself during the first commit or for bigger
changes, so less time per packages required.

-- 

Thomas Sachau
Gentoo Linux Developer


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 299 bytes --]

Re: [gentoo-project] Status update of Sunrise project?

[Ben de Groot]

On 11 April 2015 at 00:14, Thomas Sachau <tommy@gentoo.org> wrote:
> Ben de Groot schrieb:
>> I know the difference. But the two projects have similar goals: review
>> user submitted ebuilds, and commit them to an central repo to make
>> them available to a wider public. I think it would be good to get more
>> packages from sunrise into the main gentoo repo.
>>
>> Are there reasons for some of those packages to be kept out of the main repo?
>
> I think, the main reason is the required amount of developer time needed
> for that move and the continued maintainence of the moved packages.
>
> Sunrise has the advantage, that the packages per dev ratio is higher
> then for packages proxied in the main tree as the user does all the work
> (prepare, test, commit and update the ebuilds), the devs do just a
> review of the ebuild itself during the first commit or for bigger
> changes, so less time per packages required.

Since you said you are the only remaining active developer on Sunrise,
and the proxy-maintainers team has quite a few more, and we now have
git pull requests for the main repo, I don't think that reason is
quite so important anymore.

Of course you are free to continue with Sunrise, but in my opinion the
first port of call for user contributions should be proxy-maintainers.
It is better to include useful packages in the main repo, don't you
agree?

-- 
Cheers,

Ben | yngwin
Gentoo developer

Re: [gentoo-project] Status update of Sunrise project?

[hasufell]

On 04/11/2015 07:38 AM, Ben de Groot wrote:
> On 11 April 2015 at 00:14, Thomas Sachau <tommy@gentoo.org> wrote:
>> Ben de Groot schrieb:
>>> I know the difference. But the two projects have similar goals: review
>>> user submitted ebuilds, and commit them to an central repo to make
>>> them available to a wider public. I think it would be good to get more
>>> packages from sunrise into the main gentoo repo.
>>>
>>> Are there reasons for some of those packages to be kept out of the main repo?
>>
>> I think, the main reason is the required amount of developer time needed
>> for that move and the continued maintainence of the moved packages.
>>
>> Sunrise has the advantage, that the packages per dev ratio is higher
>> then for packages proxied in the main tree as the user does all the work
>> (prepare, test, commit and update the ebuilds), the devs do just a
>> review of the ebuild itself during the first commit or for bigger
>> changes, so less time per packages required.
> 
> Since you said you are the only remaining active developer on Sunrise,
> and the proxy-maintainers team has quite a few more, and we now have
> git pull requests for the main repo, I don't think that reason is
> quite so important anymore.
> 
> Of course you are free to continue with Sunrise, but in my opinion the
> first port of call for user contributions should be proxy-maintainers.
> It is better to include useful packages in the main repo, don't you
> agree?
> 

Although you didn't ask me I don't agree, because the statement is too
broad.

Unless gentoo workflow gets fixed (not just by replacing the VCS)...
importing something into the tree often slows down contribution activity
and also version bumps, unless... you _use_ that package yourself.

Proxy-maintainers do not solve that problem. Neither does sunrise. I
have used both workflows and contributed to both and ended up running my
own overlays or contributing to major overlays, because that's currently
the best solution. And a lot of projects do that too and I'd like to
encourage other projects (e.g. office) to try it as well.

Re: [gentoo-project] Status update of Sunrise project?

[Rich Freeman]

On Sat, Apr 11, 2015 at 6:12 AM, hasufell <hasufell@gentoo.org> wrote:
> On 04/11/2015 07:38 AM, Ben de Groot wrote:
>>
>> Since you said you are the only remaining active developer on Sunrise,
>> and the proxy-maintainers team has quite a few more, and we now have
>> git pull requests for the main repo, I don't think that reason is
>> quite so important anymore.
>>
>> Of course you are free to continue with Sunrise, but in my opinion the
>> first port of call for user contributions should be proxy-maintainers.
>> It is better to include useful packages in the main repo, don't you
>> agree?
>>
>
> Although you didn't ask me I don't agree, because the statement is too
> broad.
>
> Unless gentoo workflow gets fixed (not just by replacing the VCS)...
> importing something into the tree often slows down contribution activity
> and also version bumps, unless... you _use_ that package yourself.
>
> Proxy-maintainers do not solve that problem. Neither does sunrise.

Since most of us want the gentoo repository to be as easy to
contribute to as possible, I'd be interested in your discrete answers
to:

1.  What does proxy-maintainers lack in comparison to sunrise
exclusively.  The immediate question is whether sunrise should be
migrated to proxy-maintainers, so this specific comparison is
important.

2.  What does proxy-maintainers lack in comparison to overlays in
general, beyond QA standards?  (By QA standards I'm more concerned
with our QA goals such as not having security-vulnerable packages in
the tree, having consistent depgraphs, having PMS-compliant ebuilds,
etc.  I'd rather not discuss changes to these, unless there really is
something most of us don't think is necessary.  How we go about
achieving those goals is fair game.  ie, one "benefit" of overlays is
that you can commit an ebuild that contains only line noise, and I'm
not so interested in that.  However, maybe another benefit of overlays
is that you can go about quality in a different way that makes it
easier, ultimately reaching a level of quality comparable to the
gentoo repository.)

-- 
Rich

Re: [gentoo-project] Status update of Sunrise project?

[hasufell]

On 04/11/2015 01:54 PM, Rich Freeman wrote:
> 
> 1.  What does proxy-maintainers lack in comparison to sunrise
> exclusively.  The immediate question is whether sunrise should be
> migrated to proxy-maintainers, so this specific comparison is
> important.
> 

proxy-maintainers lack:
1. a repository with a usable VCS
2. an actual review workflow... @proxy-maintainers are just some sort of
backup committers. it's not a hub for contributors to gather, discuss,
get reviews and improve skills
3. means to ensure the tree doesn't break
4. actively look for and educate potential developers, even before the
recruitment process

So it should, if at all, be the other way around: dissolve
proxy-maintainers, fix the sunrise workflow and make it the contribution
hub again it once was. But I'm not actually advocating for that. I think
the sunrise concept doesn't work anymore.

What we need is a said hub, but not under the dominance of one
project/repository. It should rather be a meeting point for various
high-quality overlay developers. One problem with sunrise often was: no
one really knew about e.g. java ebuilds, so we ended up telling people
with those ebuilds to get a review elsewhere.
That said, it makes more sense to directly contribute haskell ebuilds to
the haskell overlay where people have the knowledge about these things
instead of dumping everything in one "user" repository.
But in order to know where to go, there should be some sort of central
meeting point.

> 2.  What does proxy-maintainers lack in comparison to overlays in
> general, beyond QA standards?  (By QA standards I'm more concerned
> with our QA goals such as not having security-vulnerable packages in
> the tree, having consistent depgraphs, having PMS-compliant ebuilds,
> etc.  I'd rather not discuss changes to these, unless there really is
> something most of us don't think is necessary.  How we go about
> achieving those goals is fair game.  ie, one "benefit" of overlays is
> that you can commit an ebuild that contains only line noise, and I'm
> not so interested in that.  However, maybe another benefit of overlays
> is that you can go about quality in a different way that makes it
> easier, ultimately reaching a level of quality comparable to the
> gentoo repository.)
> 

Pretty much the same as already said above. I'd also like to note that
the gentoo tree isn't really the nonplusultra in terms of ebuild
quality, because a lot of people commit stuff without really knowing
what they are doing. So, dedicated overlays sometimes have higher
quality. There's just no patrick or diego running global checks. But
that problem has been solved by exherbo for years now, with far less
manpower. I'll cut the infra-rant. Maybe mgorny will get us there.
Pretty much the only one left who's actively trying to improve things.
And that is both sad and dangerous.

Re: [gentoo-project] Status update of Sunrise project?

[Patrick Lauer]

On Sunday 12 April 2015 11:59:09 hasufell wrote:
> On 04/11/2015 01:54 PM, Rich Freeman wrote:
> > 1.  What does proxy-maintainers lack in comparison to sunrise
> > exclusively.  The immediate question is whether sunrise should be
> > migrated to proxy-maintainers, so this specific comparison is
> > important.
> 
> proxy-maintainers lack:
> 1. a repository with a usable VCS
> 2. an actual review workflow... @proxy-maintainers are just some sort of
> backup committers. it's not a hub for contributors to gather, discuss,
> get reviews and improve skills
> 3. means to ensure the tree doesn't break
> 4. actively look for and educate potential developers, even before the
> recruitment process

Oh my. 

Can you please stop being such a drama queen and accept reality every now and 
then?


 
> So it should, if at all, be the other way around: dissolve
> proxy-maintainers, fix the sunrise workflow and make it the contribution
> hub again it once was. But I'm not actually advocating for that. I think
> the sunrise concept doesn't work anymore.

proxy-maint is the least broken process we have. Unless you have constructive 
criticism I don't see why you waste time whining about everything.
 
Sigh. 

Re: [gentoo-project] Status update of Sunrise project?

[hasufell]

On 04/12/2015 12:02 PM, Patrick Lauer wrote:
> On Sunday 12 April 2015 11:59:09 hasufell wrote:
>> On 04/11/2015 01:54 PM, Rich Freeman wrote:
>>> 1.  What does proxy-maintainers lack in comparison to sunrise
>>> exclusively.  The immediate question is whether sunrise should be
>>> migrated to proxy-maintainers, so this specific comparison is
>>> important.
>>
>> proxy-maintainers lack:
>> 1. a repository with a usable VCS
>> 2. an actual review workflow... @proxy-maintainers are just some sort of
>> backup committers. it's not a hub for contributors to gather, discuss,
>> get reviews and improve skills
>> 3. means to ensure the tree doesn't break
>> 4. actively look for and educate potential developers, even before the
>> recruitment process
> 
> Oh my. 
> 
> Can you please stop being such a drama queen and accept reality every now and 
> then?
> 

I am unable to see how that contributes anything to the discussion.
Could you please stay on the factual level?

> 
>  
>> So it should, if at all, be the other way around: dissolve
>> proxy-maintainers, fix the sunrise workflow and make it the contribution
>> hub again it once was. But I'm not actually advocating for that. I think
>> the sunrise concept doesn't work anymore.
> 
> proxy-maint is the least broken process we have. Unless you have constructive 
> criticism I don't see why you waste time whining about everything.
>  

You have obviously not read the whole email. I did offer an alternative
solution. Please read the whole text.

Re: [gentoo-project] Status update of Sunrise project?

[Patrick Lauer]

On Sunday 12 April 2015 12:04:08 hasufell wrote:
> On 04/12/2015 12:02 PM, Patrick Lauer wrote:
> > On Sunday 12 April 2015 11:59:09 hasufell wrote:
> >> On 04/11/2015 01:54 PM, Rich Freeman wrote:
> >>> 1.  What does proxy-maintainers lack in comparison to sunrise
> >>> exclusively.  The immediate question is whether sunrise should be
> >>> migrated to proxy-maintainers, so this specific comparison is
> >>> important.
> >> 
> >> proxy-maintainers lack:
> >> 1. a repository with a usable VCS
> >> 2. an actual review workflow... @proxy-maintainers are just some sort of
> >> backup committers. it's not a hub for contributors to gather, discuss,
> >> get reviews and improve skills
> >> 3. means to ensure the tree doesn't break
> >> 4. actively look for and educate potential developers, even before the
> >> recruitment process
> > 
> > Oh my.
> > 
> > Can you please stop being such a drama queen and accept reality every now
> > and then?
> 
> I am unable to see how that contributes anything to the discussion.
> Could you please stay on the factual level?

So.

(1) is a completely subjective opinion 
(2) is kinda ... like ... your opinion, if you were actually involved with 
proxy-maint you'd know better
(3) uhm what? That is just absurd speculation without a factual base in our 
common shared reality
(4) That's, like, your opinion, maaan.

If you actually spent time near proxy-maint, or in #gentoo-dev-help, or 
basically anywhere where you'd interact with people, you'd not have these 
absurd fantasies about stuff.

So I suggest you can try to do some proper fact finding first :)

... and then we could even try having a constructive discussion. That'd be, 
like, AAAWESOOME.

> 
> >> So it should, if at all, be the other way around: dissolve
> >> proxy-maintainers, fix the sunrise workflow and make it the contribution
> >> hub again it once was. But I'm not actually advocating for that. I think
> >> the sunrise concept doesn't work anymore.
> > 
> > proxy-maint is the least broken process we have. Unless you have
> > constructive criticism I don't see why you waste time whining about
> > everything.
> You have obviously not read the whole email. I did offer an alternative
> solution. Please read the whole text.

Mhhh yes. No. Thanks for playing.
(Are you trying to get me drunk? Because that's how you get me drunk ...)

Re: [gentoo-project] Status update of Sunrise project?

[hasufell]

On 04/12/2015 12:13 PM, Patrick Lauer wrote:
> 
> So.
> 
> (1) is a completely subjective opinion 
> (2) is kinda ... like ... your opinion, if you were actually involved with 
> proxy-maint you'd know better
> (3) uhm what? That is just absurd speculation without a factual base in our 
> common shared reality
> (4) That's, like, your opinion, maaan.
> 

Of course it's my opinion. That's how a discussion starts.

> If you actually spent time near proxy-maint, or in #gentoo-dev-help, or 
> basically anywhere where you'd interact with people

I do. Not sure where you got the information that I don't.

> So I suggest you can try to do some proper fact finding first :)
> 

Had been a sunrise dev for quite some time and have been involved with a
lot of proxy-maintainers, so yes, that's where I got these "facts".


I'm not sure how your confusions contributes to this discussion. You
just seem to disagree. Ok.

Re: [gentoo-project] Status update of Sunrise project?

[Patrick Lauer]

On 04/12/15 18:21, hasufell wrote:
[snip random noise]
> 
> I'm not sure how your confusions contributes to this discussion. You
> just seem to disagree. Ok.
> 
We don't seem to share enough reality to have a discussion ...

Re: [gentoo-project] Status update of Sunrise project?

[Rich Freeman]

On Sun, Apr 12, 2015 at 5:59 AM, hasufell <hasufell@gentoo.org> wrote:
> On 04/11/2015 01:54 PM, Rich Freeman wrote:
>>
>> 1.  What does proxy-maintainers lack in comparison to sunrise
>> exclusively.  The immediate question is whether sunrise should be
>> migrated to proxy-maintainers, so this specific comparison is
>> important.
>>
>
> proxy-maintainers lack:
> 1. a repository with a usable VCS

proxy-maintainers is using git.  Am I missing something?

> 2. an actual review workflow... @proxy-maintainers are just some sort of
> backup committers. it's not a hub for contributors to gather, discuss,
> get reviews and improve skills

My understanding is that github pull-requests are being used.  I'll
buy that the project is still fairly immature but it seems like they
have a decent foundation for a review-oriented workflow.  Perhaps you
might consider getting involved and building improvements?

> 3. means to ensure the tree doesn't break

Well, we don't even do this well in the main repository.
proxy-maintainers can use repoman, as can committers, but I'll buy
that those aren't the best solution.  There are already efforts to get
travis-ci working on the main tree, and presumably this could be
applied to proxy-maintainers as well.

> 4. actively look for and educate potential developers, even before the
> recruitment process

Sure, but I don't see this as an issue with proxy-maintainers, but
more of another layer to add.

Honestly, I see many of the seeds of things that you're looking for in
the project already.  Change has been slow, but the number of people
doing the work has also been small.  We don't have a huge number of
new devs in general, and most of those new devs are interested in
being package maintainers, not "Gentoo Maintainers" helping to change
how we do things.  I'd be interested in suggestions as to how to
recruit the latter, but I think the nature of our distro tends to
appeal more to users who are more interested in the nuts and bolts
than the organization.

I get your frustration, but I think you should look more at the trend
direction.  Sure, we're not doing a lot in terms of a workflow not
centered around having cvs commit rights, but we're doing a lot more
than we were even a year ago.  If there are others who would like to
make it happen but see barriers to this, I'd be interested in hearing
from them.

-- 
Rich

Re: [gentoo-project] Status update of Sunrise project?

[Michał Górny]

[-- Attachment #1: Type: text/plain, Size: 1792 bytes --]

Dnia 2015-04-12, o godz. 07:22:17
Rich Freeman <rich0@gentoo.org> napisał(a):

> On Sun, Apr 12, 2015 at 5:59 AM, hasufell <hasufell@gentoo.org> wrote:
> > On 04/11/2015 01:54 PM, Rich Freeman wrote:
> >>
> >> 1.  What does proxy-maintainers lack in comparison to sunrise
> >> exclusively.  The immediate question is whether sunrise should be
> >> migrated to proxy-maintainers, so this specific comparison is
> >> important.
> >>
> >
> > proxy-maintainers lack:
> > 1. a repository with a usable VCS
> 
> proxy-maintainers is using git.  Am I missing something?
> 
> > 2. an actual review workflow... @proxy-maintainers are just some sort of
> > backup committers. it's not a hub for contributors to gather, discuss,
> > get reviews and improve skills
> 
> My understanding is that github pull-requests are being used.  I'll
> buy that the project is still fairly immature but it seems like they
> have a decent foundation for a review-oriented workflow.  Perhaps you
> might consider getting involved and building improvements?
> 
> > 3. means to ensure the tree doesn't break
> 
> Well, we don't even do this well in the main repository.
> proxy-maintainers can use repoman, as can committers, but I'll buy
> that those aren't the best solution.  There are already efforts to get
> travis-ci working on the main tree, and presumably this could be
> applied to proxy-maintainers as well.

Some of the proxy-maintainers have already switched to our git Gentoo
repo mirror. There, the pull requests are automatically checked using
repository-wide pcheck runs thanks to travis. This is actually better
than what most of Gentoo developers do, even though I asked them
multiple times to use the mirror when in doubt.

-- 
Best regards,
Michał Górny

[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 949 bytes --]