FAQ Search Today's Posts Mark Forums Read
» Video Reviews

» Linux Archive

Linux-archive is a website aiming to archive linux email lists and to make them easily accessible for linux users/developers.


» Sponsor

» Partners

» Sponsor

Go Back   Linux Archive > Redhat > Fedora Advisory Board

 
 
LinkBack Thread Tools
 
Old 05-07-2010, 04:06 PM
Michael Schwendt
 
Default question for board members

On Fri, 7 May 2010 09:47:28 -0500, Matt wrote:

> I don't see the "package review metrics" as focusing on quantity over
> quality. I see it as a way to reward, albeit merely with a "thanks
> for stepping up", for people who review packages. It's an otherwise
> thankless task, and we have a backlog of package review requests, so
> clearly someone is using those packages (if only we could have a legal
> way to get a feel for how many).

The issue here is that the reviewer only contributes reviews to
meet dubious requirements: 1) the ranking of who did the most reviews
in the past week(s), 2) the growing queue of review requests which
is considered a problem by some people, 3) the new packager who has not
been approved yet and might complain loudly, 4) the packager sponsor
who is expected to sponsor new packagers (who in the end might not
contribute what Red Hat… uhm… Fedora leadership is hoping for).

> Sure, it's possible that some
> reviewers might push through with a poor review just to bump their
> review stats, but I'd be surprised if that's a big problem.

They might push through with a _good_ review, and repeat that for dozens of
similar packages (e.g. Perl/Python/Ruby modules), ... and still it isn't
measured whether there is interest in those packages. It may also be that
the packager doesn't maintain the initial quality of the approved package,
because the reviewer won't return to review updates and version upgrades.

Preferably, the review queue would be processed by people, who use the
stuff actually and care enough to either sign up as co-maintainers or as
dedicated test-users to build community package maintenance teams for each
and every package. Instead, we have packagers who own two dozen packages
each (some own a hundred), only to orphan them after a year or so, because
it became boring to update them and because no community developed around
them.

> When we had ~100 packagers, and ~3000 packages, if something broke,
> one of the 100 would either fix it themselves or quickly find the
> owner and get it fixed quickly. Those 100 kept up their quality due to
> the BPB ("brown paper bag"). We're over 2400 packagers now, and
> quickly approaching 10k packages (so quite an increase in scale), and
> there's also less "shame" if a packager or package breaks something -
> "the automated tools should have caught it". Only they don't.

Is it "less shame" due to the number of packagers? Or is it "less shame"
because the package collection contains a larger number of less popular
packages, which don't have a big impact if they break? I think it's the
latter. The Fedora package collection contains software, which simply
isn't used anywhere by Fedora users, and which sometimes doesn't even
install or work at all, because not even the package owner uses the
software on all dist releases.

> [...] get the quality reputation back where it's historically been.

The "development period" is where to work towards that. Fedora suffers
more from Gold releases not being ready -- and being in need of a series
of bug-fix updates -- than from ordinary maintenance updates making it worse.
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 06:06 PM
"Jason L Tibbitts III"
 
Default question for board members

>>>>> "MS" == Michael Schwendt <bugs.michael@gmx.net> writes:

MS> The issue here is that the reviewer only contributes reviews to meet
MS> dubious requirements:

Do you have a particular reviewer in mind, or are you somehow implying
that everyone who reviews packages does so to inflate their numbers?
Because my count was for some time right at the top of the list, and I
can categorically state that said count made no difference at all in
whether I chose to spend my time reviewing packages.

MS> 1) the ranking of who did the most reviews in the past week(s),

I didn't care at all. Still don't.

MS> 2) the growing queue of review requests which is considered a
MS> problem by some people,

I consider it a problem, because we have people who have obviously done
some work to become contributors and they deserve to at least receive
some response to their submission.

MS> 3) the new packager who has not been approved yet and might complain
MS> loudly,

People can complain all they like; the quality of their work is all that
ever mattered to me.

MS> 4) the packager sponsor who is expected to sponsor new packagers
MS> (who in the end might not contribute what Red Hat… uhm… Fedora
MS> leadership is hoping for).

I fail to understand what you're getting at here. I've never felt
pressure to sponsor anyone. "Red Hat" or "Fedora" leadership never gave
any demands, requirements, restrictions or hopes to me regarding
sponsorship.

MS> Instead, we have packagers who own two dozen packages each (some own
MS> a hundred), only to orphan them after a year or so, because it
MS> became boring to update them and because no community developed
MS> around them.

I think that's OK. We shouldn't be afraid to dump packages from the
distro when the maintainer goes away. If someone really wants them,
someone will step up to contribute them.

- J<
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 06:37 PM
Stephen John Smoogen
 
Default question for board members

On Fri, May 7, 2010 at 10:06 AM, Michael Schwendt <bugs.michael@gmx.net> wrote:
> On Fri, 7 May 2010 09:47:28 -0500, Matt wrote:
>
>
> Is it "less shame" due to the number of packagers? Or is it "less shame"
> because the package collection contains a larger number of less popular
> packages, which don't have a big impact if they break? I think it's the
> latter. The Fedora package collection contains software, which simply
> isn't used anywhere by Fedora users, and which sometimes doesn't even
> install or work at all, because not even the package owner uses the
> software on all dist releases.

I think you will have a lot of agreement from people who maybe 2 or 3
years ago were also in the category of "Fedora isn't useful if its not
got as many packages as Debian." How much of our growth over the last
4 years has been due to this mind game collectively we have gotten
into that if we don't have the most users, the most developers, the
most packages we aren't successful or good enough?


>> [...] get the quality reputation back where it's historically been.
>
> The "development period" is where to work towards that. Fedora suffers
> more from Gold releases not being ready -- and being in need of a series
> of bug-fix updates -- than from ordinary maintenance updates making it worse.

>From reading a ton of posts in the last year, I do not know how much
of it is because people think it doesn't matter because we can always
bug fix it later or because we don't spend enough time testing it in
the first place.. or both because they seem part and parcel of each
other. If enough developers think that having a big fat internet
connection is a requirement to use Fedora so you can get continual bug
fixes.. then the idea of having a working Gold release goes away,
because every day is Gold day. And if enough people don't think that
testing is important because it will be fixed later... it just feeds
the cycle.


--
Stephen J Smoogen.
“The core skill of innovators is error recovery, not failure avoidance.”
Randy Nelson, President of Pixar University.
"We have a strategic plan. It's called doing things.""
— Herb Kelleher, founder Southwest Airlines
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 06:42 PM
Jeff Spaleta
 
Default question for board members

On Fri, May 7, 2010 at 10:06 AM, Jason L Tibbitts III <tibbs@math.uh.edu> wrote:
> I think that's OK. *We shouldn't be afraid to dump packages from the
> distro when the maintainer goes away. *If someone really wants them,
> someone will step up to contribute them.


Did we ever develop a probational maintainership path that would make
it easier for people who cared about orphaned packages to help with
the workload as a means to show right-action and and packaging
knowledge so they can gain full sponsorship?

Our sponsorship process is very focused on new package submission.
I'd really like an opportunity to work with a new contributor on
existing packages they care about so they could be given the ability
to do the work in our cvs and our bugzilla..but I stand up as
accountable for tagged builds for that package. Give them 6 months or
so working through me to prove to a sponsor review that they know what
they are doing..but let them do the work and make mistakes but in a
way that other people are confident won't impact users because I'm
trusted to review their work before releasing it into publicly
consumable trees.

-jef
-jef
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 07:13 PM
Bruno Wolff III
 
Default question for board members

On Fri, May 07, 2010 at 10:42:14 -0800,
Jeff Spaleta <jspaleta@gmail.com> wrote:
> On Fri, May 7, 2010 at 10:06 AM, Jason L Tibbitts III <tibbs@math.uh.edu> wrote:
> > I think that's OK. *We shouldn't be afraid to dump packages from the
> > distro when the maintainer goes away. *If someone really wants them,
> > someone will step up to contribute them.
>
>
> Did we ever develop a probational maintainership path that would make
> it easier for people who cared about orphaned packages to help with
> the workload as a means to show right-action and and packaging
> knowledge so they can gain full sponsorship?

I think a few people have gotten approval to be comaintainers as their
entry to packaging, but I haven't followed those cases closely.
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 08:09 PM
Michael Schwendt
 
Default question for board members

On Fri, 07 May 2010 13:06:39 -0500, Jason wrote:

> MS> The issue here is that the reviewer only contributes reviews to meet
> MS> dubious requirements:
>
> Do you have a particular reviewer in mind, or are you somehow implying
> that everyone who reviews packages does so to inflate their numbers?
> Because my count was for some time right at the top of the list, and I
> can categorically state that said count made no difference at all in
> whether I chose to spend my time reviewing packages.

So in your case maybe you didn't consider your position on that list as
rewarding. Do you want to imply that no other reviewer was interested in
the increased visibility of his reviewing activities due to that list?
Or to become a packagersponsor quickly (and then move on with other
activities)?

> MS> 2) the growing queue of review requests which is considered a
> MS> problem by some people,
>
> I consider it a problem, because we have people who have obviously done
> some work to become contributors and they deserve to at least receive
> some response to their submission.

Only if uncontrolled growth of the package collection is one of Fedora's
goals. A system where the number of packages grows faster than the number
of package maintainers. With only a fraction of the packagers doing the
reviews.

> MS> 4) the packager sponsor who is expected to sponsor new packagers
> MS> (who in the end might not contribute what Red Hat… uhm… Fedora
> MS> leadership is hoping for).
>
> I fail to understand what you're getting at here. I've never felt
> pressure to sponsor anyone. "Red Hat" or "Fedora" leadership never gave
> any demands, requirements, restrictions or hopes to me regarding
> sponsorship.

No pressure, but to grow, to grow, to grow has been one of the goals. To
grow with arbitrary packages (provided they passed review), to grow with
arbitrary people (provided they found a packagersponsor to approve them).
To grow out of proportions.

> We shouldn't be afraid to dump packages from the
> distro when the maintainer goes away. If someone really wants them,
> someone will step up to contribute them.

Well, I think not even for every package in the collection there is
somebody who really wants it. I go as far as claiming that some packages
are dumped *into* the collection.
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 08:15 PM
Kevin Fenzi
 
Default question for board members

On Fri, 7 May 2010 10:42:14 -0800
Jeff Spaleta <jspaleta@gmail.com> wrote:

> Did we ever develop a probational maintainership path that would make
> it easier for people who cared about orphaned packages to help with
> the workload as a means to show right-action and and packaging
> knowledge so they can gain full sponsorship?
>
> Our sponsorship process is very focused on new package submission.
> I'd really like an opportunity to work with a new contributor on
> existing packages they care about so they could be given the ability
> to do the work in our cvs and our bugzilla..but I stand up as
> accountable for tagged builds for that package. Give them 6 months or
> so working through me to prove to a sponsor review that they know what
> they are doing..but let them do the work and make mistakes but in a
> way that other people are confident won't impact users because I'm
> trusted to review their work before releasing it into publicly
> consumable trees.

Yes. Several things to note here:

A sponsor can currently sponsor anyone they think is good and able
to maintain things. The traditional route is new package submissions
and pre-reviews, but they can use any other means they like to decide
this.

See https://fedorahosted.org/fesco/ticket/108 and
https://fedorahosted.org/fesco/ticket/275. These are pending implementation.
Basically this adds a 'any packager can commit to my package', updates
the guidelines to specifically show another path to maintainership is
to contact an existing maintainer, they contact a sponsor and that
sponsor sponsors the new person in so they can co-maintain and be
mentored by the maintainer.

kevin
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-07-2010, 08:21 PM
"Jason L Tibbitts III"
 
Default question for board members

>>>>> "MS" == Michael Schwendt <bugs.michael@gmx.net> writes:

MS> So in your case maybe you didn't consider your position on that list
MS> as rewarding. Do you want to imply that no other reviewer was
MS> interested in the increased visibility of his reviewing activities
MS> due to that list?

Erm, you really should at some point look into how to properly conduct
polite discussion. I asked a question and gave an example, nothing
more.

End of my involvement in this discussion since you've confirmed my fears
that it was pointless.

- J<
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 
Old 05-08-2010, 08:51 PM
"Paul W. Frields"
 
Default question for board members

On Thu, May 06, 2010 at 03:07:51PM -0700, Jesse Keating wrote:
> On Thu, 2010-05-06 at 16:18 -0500, Mike McGrath wrote:
> > I joined the board on the platform that I wanted to bring focus to the
> > project. I felt and still feel that our lack of a unified vision has
> > caused anyone and everyone to join the project. Now that they're here and
> > have conflicting views on what Fedora should be, we're seeing lots more
> > in-fighting because of it.
> >
> > In pushing for this unified vision I think I've accomplished just the
> > opposite. The more we as a project thought about the whole "what is
> > fedora" "Who is it for", the more divisive a subject it became. Everyone
> > thought their version of Fedora was the right one. We went in the
> > opposing direction of unity.
> >
> > We've seen the project continue to grow but scale poorly. Our packagers
> > used to be able to do anything they wanted, now have to follow a process.
> > People don't like being told what to do, regardless of if its better for
> > the whole or not. Our processes have gotten more complex and difficult to
> > follow, especially for causal packagers while some needed processes still
> > don't exist.
>
> To add to this, it seems to me that we as a project and as project
> leaders are too afraid of turning away those that joined under the
> 'anything goes' time, or rather the time without direction. When trying
> to decide on a direction and goal and vision, we seem just too unwilling
> to tell people that what you wanted out of the project just isn't what
> we want, and just isn't where we're going. We're too afraid to turn
> people away.
>
> People say that Linux is about choice, and that Fedora should be about
> choice and anybody should be able to do anything they want with Fedora.
> I pretty strongly oppose this view point. To paraphrase Adam Jackson a
> bit here, Linux is absolutely about choice. You can choose to build
> your own distro, or to use and participate in one of the existing ones.
> Fedora does not and should not be about that level of choice. Fedora is
> but one choice in the vast sea of options. And if you don't like them,
> but like some of Fedora, we make it very easy for you to take Fedora and
> make your own thing out of it, outside of the project. I strongly feel
> that in order for Fedora as a project and community to continue to
> scale, we need some hard direction and real ability to say no, and to
> put certain goals above all else. We may lose some people, and that's
> OK, there will be somewhere else for them. What we will gain is shared
> vision, a common goal, and much less infighting about the little things.
>
> When we continue to try to be all things to every person, we continue to
> deliver poorly and disappoint everyone nearly equally.

I don't want to confuse "doing not much better than anyone else" with
"delivering poorly." But that's no excuse for failing to set a higher
bar than we've been doing. The current state of Fedora is that most
of our operations enable us to be an RPM-based Debian, as Seth said in
his blog[1] recently. We do release more quickly, and the releases
are relevant due to heavy development resources and being the upstream
for RHEL, the most important Linux distro to the world at large. One
could argue that as long as Red Hat can build their product from
Fedora, vital resources will keep flowing into Fedora. But it seems
to me that if we're not producing something that captures the
imagination of a large audience in a growing market, the relevance of
Fedora could easily diminish.

We know the Fedora download is the most prominent way the Fedora
Project advertises its work. People view and investigate what we've
made, discover how it's made, and then get interested in participating
in that process. While I would argue that none of the statistics Mike
brought up are great from a scientific basis, there's not an
undeniably huge uptick in the number of Fedora downloads, and that
*is* a concern. It's only one measure of our appeal, but it's an
important one. And as one of the people who makes the case for
resources for Fedora to Red Hat managers, I'd like to be able to show
much higher metrics when I propose increased investments.

Lest anyone think that's a very selfish and unimportant motive, I'd
say that the majority of people who spend their time on Fedora, no
matter whether they're paid to do it or not, care too. They'd have
increased satisfaction by knowing their work reaches an even faster
increasing number of people. There may be some to whom this measure
is completely irrelevant, but they should at least not be *distressed*
by an increase.

At the same time, though, we don't have good ways to find out how well
we're doing at growing the community. Paying attention to mailing
lists isn't sufficient since a lot of work goes on outside lists. And
lists are exceptionally poor barometers when they're dominated, and
our perceptions skewed, by a few toxic personalities. Account numbers
don't tell the tale either. We need better objective metrics that
will help us assess how contribution is happening, and how well it's
growing over time.

This is an important discussion for us to have openly in our
community, and also an example of the transparency we enjoy as opposed
to some of our perceived competitors. Establishing focus for our
distro *doesn't* have to mean shutting down a slew of things the
community wants to work on. But we need clearly established
priorities, and a common understanding of them across the whole
project. Mike's comments about growing while not scaling make a lot
of sense in this respect. Increasingly clear priorities and criteria
for releases, testing, and critical path are a good example. Over the
last few releases we're relying on well-documented and objective
measures to determine release readiness. With a clearer set of
priorities and end goals, we can realize a higher level of
collaboration and coherence.

To make those kinds of decisions is not a step to be taken lightly,
but it must be done definitively. Leaders throughout Fedora, not just
the FPL and the Board, would have to agree we'd travel this road for
the next three or four releases, and collaborate together to execute
against it. Some teams, like the Fedora marketing team, have already
said they are eagerly looking for this sort of priority framework to
help them actively assign work. Others, like Fedora's translation
teams, might not be affected as much. This isn't something I think we
can do by dipping our collective toe in the water for a single
release. We'll also continue to ensure the work of the Fedora team in
Red Hat, who are paid to work on Fedora full-time, is well aligned
with these priorities as well.

For instance, one way we could capture important community metrics
would be to centralize the capture of contributor work such as project
commits, package builds, wiki edits, and so on. That would give us a
basis to know how our strategies and implementations affect our
contributors' efforts. Work has already started on a Fedora messaging
bus that would allow us to provide better automation and communication
for contributor work. That bus would also provide a way for us to
capture the statistics we need to more effectively measure community
work and see how it changes over one or more releases. A joint
responsibility of the Fedora Project Leader and the Fedora Engineering
Manager is to work together closely to support project priorities in
this fashion.

We shouldn't be stringing along or cajoling people to remain in the
project. A healthy community should include some attrition. We also
need to effectively raise the signal-to-noise ratio in our
communications channels to achieve our goals. At the same time, we
need to be very careful not to stifle constructive criticism and
debate, which are part of the lifeblood of open source. This is one
of the most difficult balances to strike in any community, and perhaps
that's why even authoritative works like The Open Source Way[2] don't
address it well yet[3],[4]. While Fedora shouldn't become a
monoculture, without deciding when some issues are closed for the time
being, we waste a lot of time on non-constructive and unprofitable
discussions. I'd propose that we as a community can make such a list
on the wiki, and that list can be used to identify when we have
problems occurring on our mailing lists.

This email is already too long to launch into a set of project
priorities, but I would like people to read and think about this email
before moving into that discussion.


* * *
[1] http://skvidal.wordpress.com/2010/05/07/dissidents/
[2] http://www.theopensourceway.org/
[3] http://www.theopensourceway.org/wiki/Stuff_everyone_knows_and_forgets_anyway#Do_not_let _poisonous_people_bog_down_the_community
[4] http://www.theopensourceway.org/wiki/Stuff_everyone_knows_and_forgets_anyway#Disable_po isonous_communication


--
Paul W. Frields http://paul.frields.org/
gpg fingerprint: 3DA6 A0AC 6D58 FEC4 0233 5906 ACDB C937 BD11 3717
http://redhat.com/ - - - - http://pfrields.fedorapeople.org/
Where open source multiplies: http://opensource.com
_______________________________________________
advisory-board mailing list
advisory-board@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/advisory-board
 

Thread Tools




All times are GMT. The time now is 10:47 AM.

VBulletin, Copyright ©2000 - 2014, Jelsoft Enterprises Ltd.
Content Relevant URLs by vBSEO ©2007, Crawlability, Inc.
Copyright 2007 - 2008, www.linux-archive.org