Re: Announce: Group projects in Copr

2015-10-15 Thread Alberto Ruiz
Wow, I was about to request this since I was about to start a shared COPR.
You're always one step ahead of me Miroslav! Thank you.

On Thu, Oct 15, 2015 at 10:34 AM, Miroslav Suchý <msu...@redhat.com> wrote:

> It is my pleasure to announce new version of Copr.
>   https://copr.fedoraproject.org/
>
> What's new? Group projects!
>
> When you log in. You can see in right column link "My Groups". There you
> can see list of your FAS groups and you can
> activate them. In fact, you actually create alias for them. E.g. FAS group
> "gitmock" can be named "mock" in Copr.
> You can click on Copr group in right column. That navigate you to url in
> format:
>
>   https://copr.fedoraproject.org/groups/g/mock/coprs/
>
> Where are listed all projects of this group. And you can create new group
> project there too.
> Every member of that FAS group can administer it and build there. Due to
> the OpenID limitations, you have to
> log-out/log-in to see your groups (or whenever you join new group in FAS).
>
> Since copr-cli-1.45 (in updates-testing right now) you can submit package
> to group project like:
>   copr-cli build @GROUP/PROJECT
> e.g
>   copr-cli build @mock/mock-dev
>
> Creation of group project is not possible from command line right now due
> OpenID restriction and you have to use WebUI.
> This may change in near future.
>
> There is no API method in APIv1. But we provide APIv2
>   http://copr-rest-api.readthedocs.org/en/latest/
> which has methods to manipulate with groups.
>
> To create new group in FAS, just follow:
>   https://admin.fedoraproject.org/accounts/group/new
>
> If you have some personal project and you want to change it to group
> project, then please let me know and we change it
> manually.
> --
> Miroslav Suchy, RHCA
> Red Hat, Senior Software Engineer, #brno, #devexp, #fedora-buildsys
> --
> devel mailing list
> devel@lists.fedoraproject.org
> https://admin.fedoraproject.org/mailman/listinfo/devel
> Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct




-- 
Alberto Ruiz
Engineering Supervisor - Desktop Management Tools
Red Hat
-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: Changing default configuration

2015-02-10 Thread Alberto Ruiz
On Tue, 2015-02-10 at 14:38 +0100, Marek Skalický wrote:
 Matthew Miller píše v Út 10. 02. 2015 v 06:19 -0500:
  On Tue, Feb 10, 2015 at 12:12:15PM +0100, Marek Skalický wrote:
   does someone know what are Fedora Guidelines (or something similar)
   saying about this bug
   https://bugzilla.redhat.com/show_bug.cgi?id=1077369 ?
   Is is possible to change default configuration file depending on
   available free space?
  
  Generally, making such a decision at install time is bad, because that
  might not reflect _runtime_. For example, in the cloud image, we use a
  / filesystem as small as Anaconda will allow, but that grows to fill
  available storage when the image is booted.
  
  Is there a way to make this decision when mongodb runs?
  
 
 It should be possible to code this into sysv init script. But I am not
 sure if it is possible to do the same in systemd service...?
 
 My question was also about whether this should be done by mongoDB? Or
 there should be default configuration and user can easily change it in
 configuration file?

It should be done by MongoDB, I would get in touch with the upstream and
discuss the problem with them.

 Marek
 

-- 
Greetings,
Alberto Ruiz
Engineering Manager - Desktop Applications Team
Red Hat, Inc.



-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: FESCo Elections results

2015-02-05 Thread Alberto Ruiz
Hello Kevin,

Could you please elaborate on what's your concern? Am I supposed to
refrain myself from trying to become a FESCo member because I am (among
many other things) a GNOME developer?

On Thu, 2015-02-05 at 14:52 +0100, Kevin Kofler wrote:
 Jaroslav Reznik wrote:
# votes |  name
  - +--
  1427  | Kevin Fenzi
  1247  | Adam Jackson
   919  | Tomas Hozza
   818  | Parag Nemade
   617  | Debarshi Ray ← GNOME developer
  - +--
   540  | Alberto Ruiz ← GNOME developer
   441  | David King ← GNOME developer
 
 What does this tell you? :-)
 
 Kevin Kofler
 

-- 
Greetings,
Alberto Ruiz
Engineering Manager - Desktop Applications Team
Red Hat, Inc.



-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: OpenH264 in Fedora

2013-11-06 Thread Alberto Ruiz
On Wed, 2013-11-06 at 16:15 +0100, Reindl Harald wrote:
  It's only a nightmare because we've steadfastly refused to build the
  tools to a) track library bundling inside app-bundles b) automate bundle
  rebuilds c) force replacement of bundle contents either by sysadmin
  action or by policy.
  
  Let's not confuse the current state of the world with the world we'd
  like to build.
 
 if you are doing all the above statet you re-invent the wheel
 currently existing with the name RPM or DEB.

Hey Reindl,

Quite the contrary, a bundle should not modify the operating system or
compromise its integrity, think about it, if a user installs Chrome in
Fedora today (and if we render the default Firefox experience unusable
for WebRTC, people will) he gets a new Yum repo in its system without
any notice...

If the RPM repo breaks, yum will stop working... (if some repo fails to
answer yum will quit with an error) so with the current model you are
encouraging third parties to push for ways to shoot yourself in the
foot.

Think about it for a moment, we are encouraging third party apps to mess
with our entire system just because we don't have any other channel to
deliver end user applications or third party extensions (codecs,
fonts, ...) than the system wide channel where all the system critical
stuff goes to. As I mentioned in another thread, you can deliver those
bundles as rpms if we wanted, but they have to be scriptless
(pre/postinstall etc) and they need to live in a different yum reposet
and rpmdb than the things we consider integral parts of the operating
system.

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: OpenH264 in Fedora

2013-11-04 Thread Alberto Ruiz
On Sat, 2013-11-02 at 21:29 +0100, Björn Persson wrote:
 Fedora mustn't have third-party repositories like RPM Fusion enabled by
 default. Users must consciously configure them.
 Therefore Fedora mustn't download Cisco's binaries by default. It will
 have to be something that users must consciously configure.

It can ask the user whether he wants to opt-in/out for the plugin installation, 
removing this mechanism at all won't help the users.

 So what's the big deal? If Cisco goes through with this, then there
 will be one more free but patent-encumbered implementation. Another
 implementation doesn't hurt, but I don't see how it solves any
 fundamental problems.

It solves a fundamental problem, you have to pay MPEG-LA a license to
distribute binaries, so now we have a source that is willing to produce
binaries for as many architectures as we need that are licensed by
MPEG-LA.

So now any user is one click away from being able to access content
widely available in the web (which is sadly going through the HTML5
spec).

We still have a problem with MP3, but it does solve a fundamental
problem.

  The only news is that Cisco will hand out blobs
 with gratis patent licenses for some time. Establishing a standard
 based on the assumption that Cisco will continue doing that
 indefinitely seems like a bad idea.

Is better than nothing, we can always fall back to what we do now once
they stop doing that.

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: OpenH264 in Fedora

2013-11-04 Thread Alberto Ruiz
On Sat, 2013-11-02 at 18:36 +0100, Kevin Kofler wrote:
 Michael Catanzaro wrote:
  Another important point is that future versions of Firefox will download
  and install (I'm not quite sure where) the binary from Cisco's
  website, unless some about:config setting is disabled. I imagine this
  could possibly mean H.264 will work only in Firefox but not other
  browsers, which would suck.
 
 IMHO, having such a setting enabled by default in Fedora is unacceptable. A 
 browser must not go and download binaries which are effectively non-Free 
 wherever the patent holds (because exercising your right to modification 
 under the copyright license voids your patent license) behind the user's 
 back.

While I agree that we shouldn't silently install non-free software (and
I'm sure Mozilla doesn't want to either), saying that they are
effectively non-free is a bit inaccurate, the _binaries_ are not
re-distributable under US jurisdiction, access to the source code is
granted, which makes them non-US, the software is free (the source
license does grant 4 freedoms). There are plenty of countries where
software patents are not valid making it perfectly fine.

Mind though, there's pretty much no line of code that is not potentially
covered by some US patent owned by someone these days.

So I'd say that what is broken here is the US patent system not the
code, and we have to deal with it. Strictly speaking, we wouldn't be
promoting non-free software here, but circumventing a policy that is
broken (not so different to what the GPL does).

It's a trade off, would you rather have users not being able to play a
hugely widespread codec that happens to be free software or would you
rather make the default experience in Fedora and Firefox better for our
users?

I think we should _inform_ users of what's going on when the codec is
being downloaded and let them opt-in/out depending on their criteria.
There's plenty of people who just want to play a video and do _not want_
to understand the licensing implications, or how to opt-in for a yum
repository, or to understand what a yum repository is at all.

Passed the point where we made an effort to inform the user I don't
think we should design our software in a way that makes it particularly
hard to make a choice about how much you care about freedom, after all,
you can just go to rpmfusion and/or other repositories and get all sorts
of non-us and non-free software if you chose to, it's just harder to do
that. The problem is that the people who do not have a grasp on how the
underlying system works have a hard time figuring these things out,
which takes their away their freedom to make a choice.

Otherwise, the perception of people is going to be this: heck, I can't
watch videos on Firefox/Fedora and I don't know what to do about it,
hence, Firefox/Fedora is pretty bad. So while trying to preserve
freedom, you are in fact, harming the perception of the quality of our
software and the ability of people to get their stuff done.

So my suggestion is that we wait for Mozilla to come up with an actual
mechanism to install this, we review it, and we make up our minds about
how to preserve freedom of choice while letting the user make an
informed decision.

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: Draft Product Description for Fedora Workstation

2013-11-04 Thread Alberto Ruiz
On Mon, 2013-11-04 at 19:00 +0100, Kevin Kofler wrote:
 Bastien Nocera wrote:
  As an application developer, I'd sure be happy if I didn't have to wait
  6 months for my app to show up in the distribution I use, and for that
  application to be usable on all compatible distributions.
 
 That's a problem with the distribution's update policies, not with the 
 concept of a repository.

Sure, that's why an application container would allow for application
specific repositories (a.k.a. stores) where apps can reviewed as apps
instead of system-wide swappable components.

Then you can have a policy that is more suitable to apps with a
different cadence than the system-wide (rpm/deb/...) repository, in fact
you may not need a cadence at all.

It is outrageous that it's 2013 and I still have to upgrade my whole
system just to get the latest LibreOffice version to name an example.

 
 Kevin Kofler
 

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: OpenH264 in Fedora

2013-11-04 Thread Alberto Ruiz
On Mon, 2013-11-04 at 11:28 -0600, Bruno Wolff III wrote:
 On Mon, Nov 04, 2013 at 15:46:07 +0100,
Alberto Ruiz ar...@redhat.com wrote:
 
 While I agree that we shouldn't silently install non-free software (and
 I'm sure Mozilla doesn't want to either), saying that they are
 effectively non-free is a bit inaccurate, the _binaries_ are not
 re-distributable under US jurisdiction, access to the source code is
 granted, which makes them non-US, the software is free (the source
 license does grant 4 freedoms). There are plenty of countries where
 software patents are not valid making it perfectly fine.
 
 If you don't need to worry about the patents, then x264 (available from 
 RPMFusion) is going to be better code to use for handling h.264.

How is the code from RPMFusion any better? And if getting it through
RPMFusion is acceptable, why is it suddenly unacceptable to do it trough
other means? I don't care about the quality of the code, I just care
that my video is decoded.

 The issue for RTC is that we could be using a royalty free codec, such as 
 VP8 instead. Accepting the binary makes it more likely that h.264 will 
 be made mandatory to implement, which means any company not wanting to 
 implement VP8 can always point to h.264 being mandatory as an excuse 
 not to support VP8.

Google gave up on that battle, Mozilla gave up on that battle, and
somehow you expect that the Fedora community can somehow turn the tides?
There are better ways to push for improvements in this effort (like the
Daala codec).

The only thing you will achieve by actively pushing this codec out is
making the life of the Fedora users more miserable. Nobody cares about
the open web more than the Mozilla organization, I'd say their criteria
(while subject to review and criticism) should be taken into
consideration.

 Another thing to worry about is how the binary is licensed. Accepting that 
 license (even in places where software patents don't apply) could potentially 
 cause issues. I haven't read the license for it yet, but most likely it will 
 be a typcial consumer license that only covers non-commercial use (similar 
 to what people get when they buy digital movie cameras).

Again, the people who have been fighting for open source media Xiph.org
and the Mozilla organization have already acknowledge that while this
situation is not ideal, is an improvement over the current situation,
I'd say we should trust these guys a little when it comes to these
things, don't you think?

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: Draft Product Description for Fedora Workstation

2013-11-04 Thread Alberto Ruiz
On Mon, 2013-11-04 at 12:56 -0500, Matthew Miller wrote:
 On Mon, Nov 04, 2013 at 06:39:54PM +0100, Kevin Kofler wrote:
  The reason we are so strongly opposed to app stores is that we are fairly 
  convinced that the mere fact of having them available WILL:
  * reduce the number of applications actually available in our repositories
(because some upstreams will just upload a bundle and tell you to use
that, and nobody will want to do the work of actually getting the package
through Fedora review),
 
 If they're easily available to our users, and work just as nicely or better,
 why is this a problem? In fact, it kind of sounds like everyone wins! to
 me.

Indeed, this is a relief to the distro community, suddenly you don't
have the burden of maintaining all these apps in your system image,
avoiding any potential problems on upgrades.

It also removes the pressure for the user to have to upgrade his/her
whole system just because they want the latest version of an app.

It also allows for parallel versions of an app, which means that to test
a beta version of, say,  LibreOffice, you don't need the beta version of
your whole operating system.

And the application developer suddenly gains back control on how and
when his app gets delivered to users.

The only downside is that we are going to have to rewire our brains to
stop using yum/dnf to browse/install/remove end user desktop apps.
(This, I think, is why this idea gets so much pushback from the distro
communities)

And, by the way, we've been supporting this kind of model with pip and
gem already, so I really don't get why all the fuss when suddenly we
want to do it with the desktop applications.

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct

Re: OpenH264 in Fedora

2013-11-04 Thread Alberto Ruiz
On Mon, 2013-11-04 at 18:31 +0100, Kevin Kofler wrote:
 Alberto Ruiz wrote:
  We still have a problem with MP3, but it does solve a fundamental
  problem.
 
 We had the same type of solution (just with a different binary producer, 
 Fluendo) offered for MP3. We rejected it, due to both political (effectively 
 unmodifiable binary) and technical (interoperability problems, e.g. with 
 SELinux, that could not be fixed because it was a binary blob) reasons. The 
 compromise that was originally implemented (Codeina, also known as the 
 Codec Buddy, which worked pretty much as you propose) was dropped soon 
 afterwards because the users did NOT like it. (They wanted the libmad plugin 
 rather than the flump3 one, and they also wanted all the other codecs in RPM 
 Fusion for which Codeina either had no alternative or only an expensive paid 
 one. So Codeina just confused them and made it no easier for them to 
 actually get what they wanted. So now we have the PackageKit-based GStreamer 
 codec installer which requires rpmfusion-free-release to be installed to be 
 useful, but then actually offers the codecs that users want.)

A user who actually cares about a certain codec implementation and rpm
package is presumably a user who is perfectly capable of setting up its
own repository (hopefully a fair assumption). They can reject the codec
if they're asked, problem solved.

At that point the current solution doesn't really help to someone who
doesn't know (and I'd go as far as saying, shouldn't know) how to do
that.

As per other technical/political details, Cisco is not Fluendo, they
have shown interest in working with the downstreams on providing
binaries for any given platform that they're asked (it's in their
interest that this codec works well and is widely available), so we
could reach out to them to see if we can solve these problems before
rejecting the idea altogether?

I seriously don't understand why we want to keep playing media files
THIS hard on Fedora. It is the only reason I use RPM fusion. We want to
fight the current situation? Let's endorse the EFF as an organization
and help them spreading the word on their campaigns to reform US policy.
Let's endorse and promote the Daala development to achieve a patent-free
techinically superior codec.

In the meantime we have to be practical, if Mozilla+Google have failed
to push for VP8 and accepted that for now H264 is the way to go, I am
afraid that the reality is that by making it hard to get H264 decoding
in Fedora for users the only thing we are going to achieve is scaring
users away and diminishing the quality perception of Fedora. That is a
net loss for free software.

Asking people to include an extra repository is no fix, it's not the
first nor the second time that my Fedora upgrades break because the
rpmfusion packages (specifically GStreamer ones) suddenly has the
ability to install any sort of software in my system and mess with the
integrity of my rpm database.

A media codec should not be a system wide component (I'd go as far as
saying it should not be user-session wide, but application bundled).

-- 
Cheers,
Alberto Ruiz

-- 
devel mailing list
devel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/devel
Fedora Code of Conduct: http://fedoraproject.org/code-of-conduct