Re: community managers

[Brian Cameron <brian cameron oracle com>]

Jon/Dave:

On Fri, Nov 16, 2012 at 10:50 AM, Dave Neary <dneary gnome org> wrote:
On 11/16/12 11:28 AM, William Jon McCann wrote:
> I was attracted to the GNOME project because of the vision and because I
> was excited by it.  [...]  I continue to be inspired by it - going on 10
> years now. Inspired enough to put up with a lot of negativity.

It's good for your karma.  :)

> Despite how some would portray it, this vision is shared by the current
> core contributors.
>
> You'll find evidence of this everywhere, if you look. That said, we
> should do more to make it explicit. Make it clear. Not because of
> threads like this but because we are proud of it. Because we want to
> shout it and we know people will respond.

I agree GNOME Rocks!

> We can be different,
> have different ideas, have different goals, and still be friends.
> Sharing where it is mutually beneficial but still appearing separate and
> distinct. Standing on our own, proudly. With individual rights and
> responsibilities.

I very much agree.  You and I, for example, have had many differences
over design choices relating to GDM, yet I also respect the significant
amount of work you do leading the project.  Friendship is more like a
spectrum than an on-off switch, so I think it can mean different things
amongst different people, but I am proud to be associated with so many
brilliant GNOME engineers.

George Lebl, the maintainer of GDM before me, indicated in his source
code comments that the believed that fixing crack gave one a certain
license to introduce more.  Modernizing GDM has broken configuration
features and caused pain for users.

Criticism aside, I do think George would applaud the fact that GDM is
finally ported to using sensible interfaces like D-Bus.  While the
"new" GDM is a step back in certain ways (such as XDMCP support since
you cannot launch the GDM chooser from the GUI anymore), I think it was
overall a step in the right direction.  I am not sure if this lack
of XDMCP chooser support breaks LTSP, though I wonder.

Interfaces like GTK+ and the entire GNOME Platform have had a stellar
ABI stability over the years, yet stability seems to be breaking down
recently.

I think ABI stability and providing existing users important updates
like security fixes are important parts of project management.  GNOME
should accept that interfaces exposed to users, such as theming
interfaces, need to be better supported if we want to build a stronger
relationship with the actual userbase.  GNOME will benefit from the 
stronger interface stability that comes with maturity, but now is
probably a good time to consider what configuration interfaces should
be more stable, such as GTK+ theming, obviously.

In the GNOME 2 cycle, it was GNOME 2.16 before GNOME really started
being usable when HAL started fixing a lot of serious desktop bugs and
GStreamer started being used.  I would say that GNOME 3.6 is already
much farther along than 2.16 was at its stage in the development cycle.
So, there is progress.  :)

> I have absolutely no problem with Cinnamon. I think I give them more
> credit than you do. They took a name, on purpose. To differentiate
> themselves - to allow people the freedom to choose a different user
> experience. They have different goals. A different appearance. Different
> behaviors. A different future. And that is fine.

Does the GNOME community have a plan for how to deal with providing
GNOME 2 users important fixes like security bug fixes?  By making a
small committment to release new GNOME 2 tarballs with security updates
as needed and making sure that updates to things like D-Bus do not
break the GNOME 2 experience, then I think GNOME maintains stronger
control over the GNOME 2 source code.  People should want to use the
GNOME source code repository if that's where they get security fixes.
Does the GNOME community have any recommendations about how a distro
should deliver a secure GNOME 2 experience?

> >     I think it's a shame that Cinnamon users don't realise, for the most
> >     part, that they are using GNOME Shell, and the rest of the GNOME 3 stack

How could they be expected to realize unless GNOME were to support them
with the GNOME brand.  GNOME provides too little guidance to distros
that use GNOME, such as OLPC, about how to reference the GNOME brand in
their products.  Or do you think GNOME should not work to encourage the
GNOME brand gets effective placement in products that use it?

> It is not a shame that users aren't concerned with or interested in
> implementation details. That is as it should be. We welcome it.

Users are concerned, though, with brands.  The implementation detail
of how GNOME makes effective use of its brand is something of their
concern.  How do you think Cinnamon should use the GNOME brand?

> We are not sending any message other than:
>
> "We are deeply sorry that we could not agree on goals. We are always
> willing to have a conversation about how we may find common ground. We
> respect your difference of opinion and your right to identify and
> differentiate yourself. We would still like to collaborate on
> implementation details and shared technology. Perhaps we can agree on
> commonality in the application development experience, and application
> delivery and installation."

This seems apologetic to me.  When does respecting difference translate
to the fair use of the GNOME brand?  If we have more clear rules that
explain how the GNOME brand should be used in products that use GNOME
depending on level of integration, then we would really be supporting
the sort of collaboration you suggest.

Do you support any existing forks to the point that you would think they
should be supported by the GNOME brand?  Should one of them be
delegated responsibility for keeping things securely updated?

> Suggesting otherwise isn't helping anyone.
>
> >     Personally, I think that it'd be cool to have our community be the
> >     community of people who can go wild on the platform - "let a
> >     thousand flowers bloom". That the core GNOME project is solid and
> >     useful, but that we encourage experimentation, respins, freedom for
> >     our users. That seems inconsistent with the current GNOME messaging.

I agree.  I think the GNOME community does not make effective use of its
brand.  I think this makes it harder to connect with users and new
developers.  More people would be proud of GNOME if they were exposed
to the brand and its ties to software freedom.  Users will only be
inspired to become developers if we encourage experimentation, as you
say.  How can we do this if real users are never exposed to the brand
unless they seek it themselves?

Wasn't Bastien just reminding us that OLPC does not yet support GNOME
3?  It does not make sense to me that GNOME does not ensure products
like OLPC have effective GNOME brand placement, or to remove Fallback
Mode before they, or any other Enterprise GNU/Linux distro has started
to support GNOME 3.  Products like OLPC should have a good support story
for security and other critical fixes, no?

Maintaining these modules at a limited level to just provide security
fixes seems a sort of minimal baseline of support that the GNOME
community should, I think, strive to surpass.

It is one thing to say we encourage people to experiment, but which
experimentors get endorsed with the GNOME brand, if any?  If none,
wouldn't that encourage forking?

I agree with the sentiment that the GNOME community could do a better
job of including real users (e.g. OLPC user, insert name here) as a
part of this circle of friendship.  This will bear fruit, if not
flowers.

I appreciate the opportunity to vent a bit, though I hope that I am
still being friendly.  Let me know if you think I am just off my rocker
or whatever.

Brian