European Wafaa

As many people will know I've spent around 18 months of my spare time packaging and building the Netbook UX for Moblin/MeeGo on openSUSE.  Not on my own, but with a community around me. I would just like to say a huge thank you to that community, you've been great and spurred me on when I felt like the whole world was conspiring against me.

Sound a bit morose? Well in a way it is. Basically by all accounts MeeGo is stopping all work on the Netbook UX. Yup, all our hard work is now almost for nothing :-(

The first piece of evidence is the Release Engineering Plans for 1.2.  If you look at the bottom, you'll notice that the Netbook doesn't get any mention in Features, just bugs. Actually the same is for the 1.1 release.  In all honesty, the biggest change between 1.1 and its predecessor is that all refernces to Moblin was replaced with MeeGo. The next piece of info came at the fabulous FOSDEM XI (I'll do a seperate post on the event later). I spoke to several people directly and indirectly involved with MeeGo, especially the Netbook UX, about the lack of info on the Netbook UX. Each one of them said the same thing - MeeGo is putting the Netbook UX into Maintanence Mode, stopping any further development on it, and only providing bug fixes for major issues.  I asked if there was going to be any announcement, basically the answer was nothing official. They're just going to let it drift away - just like netbooks themselves have.

There are most likely a huge number of reasons, which is fair enough. I was given a couple of reasons - 1: Netbooks arent selling anymore, the Netbook ship didn't really sail away, but was sunk. 2: MeeGo doesn't want to invest anymore time or effort in gtk development, and wants to concentrate on Qt. Now I take this last one with a tiny bit of salt, only because both Intel and Nokia employ (or at least did so) a sizable team of GNOME developers. Saying that though, if they want to keep a roof over their heads they have to do what their employer says, and both originally started working with touch devices. So it does make sense as much as it doesn't.

So, I herby give immediate notice that I am now no longer going to be doing any further work on the MeeGo UX for openSUSE or anyone else.  I wont delete the repo for 1.0 but will clear out the 1.1 and 1.2 repos that I had - might as well give those resources back :-) . Now if anyone would like to continue what I and others had started, then let me know and I would be more than happy.

I cant say what the other distros are going to do, I've spoken to my counter part in Fedora, and they're having a think about it as they were busting their chops to get things in for the F15 release.  Either way I wish all involved luck.

Now before any of you conspiracy theory junkies get worked up - this has nothing to do with the whole MeeGo Trademark issue that Smeegol had with the Linux Foundation.  As it happens I was due to have a conversation with the Linux Foundation about the issues between us and them, but have had to re-schedule due to travel constraints from one party or the other.  I still intend to speak with the LF about the issues, as they are still present in the other UX's that they are developing.

So it's time for me to close off and say, so long and thanks for all the fish! I'll still be around, I'll probably start getting back into the GNOME swing of things, and help fcrozat with GNOME3 etc.  Oh and I've got a lunatic idea for a project which I intend to kick off (stay tuned ;-) ).

MeeGo is a trademark of the Linux Foundation.

I've been stewing over a response on the MeeGo developers mailing list for some time now.  Basically the MeeGo Project and their steward, the Linux Foundation don't seem to be wanting to play this whole open source, open community game.  I wouldn't have any issue with it if it was just Nokia and Intel fumbling with MeeGo, it would be just another lost opportunity that big corporations screwed up on.  Problem is the Linux Foundation is involved and are supposedly the ones guiding the project.

In case you don't know, before I released Smeegol I asked some questions on the MeeGo Dev list with regards to what openSUSE can/can't do/use.  That kind of went nowhere.  Fedora have also tried asking and again got nowhere.  Debian have also asked, and after seven weeks the Linux Foundation answered.

In all honesty, I was insensed by the response. I spoke to my counterparts at Fedora and they too echoed my feelings. To allow those in the openSUSE and wider community who aren't subscribed to the MeeGo lists I'll provide my response.  For those in a rush:

tl;dr - Don't think MeeGo, don't say MeeGo, don't contribute to MeeGo unless it is on our terms, don't use MeeGo unless it is on our crippled distro, MeeGo is going it alone.

For those that want to see me rant some more:

Hi Ibrahim,

On Thu, 2010-12-09 at 13:12 -0600, Ibrahim Haddad wrote:
> Hello Didier,
>
>
> I am responding to you in relation to your inquiry below. First,
> thanks very much for sending the MeeGo Project these questions. We
> really do appreciate it. The MeeGo Project members devoted quite a bit
> of time discussing these questions to make sure the  responses are
> fair and most of all work to the benefit of the MeeGo project, its
> developers and the users of MeeGo.
>

Seven weeks to effectively turn around and say "No, no, no"?  I'm sorry
if I sound bitter or angry; but the truth of the matter is that I'm
actually fairly disgusted with how the Linux Foundation, Nokia and Intel
have all acted towards the wider Linux community.  For all the talk of
openness at the conference in Dublin all I see are closed doors and
barricades.

>
> To start with, the goal is to avoid any confusion around what is and
> what is not MeeGo.  Anything that is or will become associated with
> the MeeGo trademark has to be in conformance with the compliance
> program.  With that in mind, it becomes clear why none of the proposed
> uses of MeeGo  mentioned in your email are in line with the trademark
> policy and guidelines as they create confusion around MeeGo project
> packages, mailing list, project team name, etc.  For example, when you
> append MeeGo to a package name, it would be very reasonable to
> conclude that this is an official MeeGo package coming from MeeGo.com,
> which is not the case and could cause confusion both for your users as
> well MeeGo's.
>

I'm sorry but do you honestly think that users who install the MeeGo UX
on a distro be it openSUSE, Fedora, Debian or whichever are going to
think that it has come from the MeeGo project?  This doesn't happen for
any open source project that is available via a distribution. GNOME, KDE
etc welcome inclusion into distros and want downstream involvement.  Why
do you feel that MeeGo doesn't?

Trying to paint all users with a stupid brush is wrong and condescending
to both users and distros.  You appear to be missing a key factor that
myself and my counterparts are doing here - marketing MeeGo and enabling
the wider community to get involved and help make MeeGo better.  Why
should a developer have to install a new distro to be able to contribute
to the codebase?

Having meego in a package name shows where the upstream contribution
comes from.  Just like with packages from Google etc.  This is the most
basic form of attribution, and is only right.  Why are you trying to
force people to fork?  Yes, fork.  If you don't want people to recognise
your work and re-use it with almost no changes why have it in the open?

Linux users and especially developers can be fickle, they find a distro
they like and stick to it - quite often they are more faithful to their
distro than they are to their partner or loved one.  Forcing them to
choose between distro or contributing to MeeGo the chances are they'll
stick to their distro of choice.

When Moblin was around, it was one of three things - a User eXperience,
a distribution and a reference platform.  This worked well, so why have
you changed it?

>
> We would ask you to move away from using {M,m}-e-e-{G,g}-o or any
> subset of those letters or sounds in that order, alone or in
> combination with other letters, words or marks that would tend to
> cause someone to make a reasonable connection of the reference with
> the MeeGo mark. We specifically discussed one possibility for
> illustration purposes – which is to use MG in the place of MeeGo.  We
> do not think that a plain text MG, when used in reference to the code,
> as in a file or project or team name, would cause a reasonable person
> to be confused. 
>

This fanatical enforcement of the MeeGo trademark is going to do more
damage than good.  People have asked me what's different between MeeGo's
trademark guidelines and openSUSE's, and the simple answer is it is the
execution of the trademark that sets us apart.  Thinking that people are
going to get "confused" between Smeegol and MeeGo is ludicrous.  Sure
I'm not a lawyer/solicitor or any other form of legal professional, but
I don't need to be to apply common sense.

We are trying to promote MeeGo and further it in an already crowded
space.  Making us change package names, and removing any reference to
MeeGo gives you absolutely zero visibility.  At the end of the day,
openSUSE and the other distros have a much wider reach than MeeGo does,
and possibly ever will as a stand alone distro.


>
> We are available for additional discussions if needed and if you
> prefer to have a call to discuss we can certainly do that.
>

I have tried several times to discuss these issues, and have even tried
going off list to facilitate the dialogue in a more protected manner.
MeeGo is far from open to the community, if I was an ISV or ISV then
sure it maybe more open but from what I've heard even partners have a
tough time.

Being an exclusive project is a sure way to failure. Try being
inclusive, you'll have more chance of succeeding.  If you don't want
community participation that's fine, re-license to something proprietary
and you wont get bothered by those that want to help.  Yes you have
already re-licensed some code from open licenses to proprietary ones
without any announcement to the community, not a great open gesture.

>
> Best Regards,
> Ibrahim Haddad [on behalf of The Meego Project]
>
> PS:
> We started an FAQ at http://meego.com/trademark. We have added some
> text already and Q&A will be added there as they come.
>

I don't see a single question or answer in that FAQ, just a statement.
Not especially helpful to the community.

Regards,

Andy

I thought I should let anyone that cares know what the latest status of my work is.  What work?  Getting MeeGo on openSUSE, of course!  I'm calling it Smeegol – SUSE MeeGo Linux, and this just happens to be the Teal Goblin – Teal is the code name for 11.3 ;-)

Well I'm not quite there yet (unfortunately) :-(  I have one major issue which is the Network Panel not displaying properly, and as such is difficult to use.  I also have a smaller issue with the web panel, but that is more of an annoyance rather than a blocker.

So in a nutshell that's the state of play.  Now I'm concentrating on getting things working on 11.3 and getting a working image, but if people would like to see things get into openSUSE 11.4 then there are some low hanging fruit for people to pick up and run with.  In a nutshell before things can make it into Factory, the spec files need to be cleaned up and made to comply with the Geeko's high standards.

So please join in the fun and have a look at the Meego:Netbook repo, and help out where you can.  I wanted to get things working first before I started to look at housekeeping.  Some packages already comply, but others most certainly don't as they were straight imports from upstream.

Oh and if you want to try Smeegol as it stands now, warts and all, you can grab the live image from here. If you're brave (or stupid) you can get the goodness via 1Click for 11.3 or Factory. Yes it is 32bit only at the moment, as when packaging I had too many x86_64 errors.  Please note that the Goblin may chew fluffy bunnies, sodomise gorrilas and generally be an imp!  You can file bugs if you wish over on BNC just select the Moblin Component

So please join in the fun and help clean up the Goblin :-)

I thought it was about time that I enlightened people as to what the
state of play is with my efforts of getting MeeGo 1.0 for Netbooks on
openSUSE.

The good news is I think I'm pretty much there :-)  I believe all the
essential packages are built now, although there is a *lot* of work to
be done in getting the packages into Factory.  The outstanding work is
mostly spec file housekeeping but non the less it will be a relatively
laborious task.

I am currently looking at getting a hybrid iso image built so that
people can install the goodness straight of the bat and have the image
available pretty close to 11.3's release - this may or may not happen as
fast as I would like but we'll see.

For those that are interested all the packages etc live in
Meego:Netbook, and the image stuff lives in the meego-live package in
there.  Any and all help with cleaning spec files and creating an image
is most appreciated.

Talking of thanks, none of this would have happened without the huge
amounts of help from DimStar, darix, dl9pf, joeyli, glin, mchang, mlin, mmeeks
and vuntz. Thank you *SO* much chaps!

I will try and get a 1-Click done whilst I work on an image.  If people could have a look at the repo and let me know if anything is missing that would be great - don't forget you can submit anything that is missing ;-)

As soon as I have anything new I'll let you know.   So if you want to join in the fun, then head over to #opensuse-moblin on IRC and the opensuse-goblin mailing list.  There are lots for people to do, and if you want to start contributing this is a good way to do so :-)

It's been a while since I mentioned anything about Goblin and anything netbook related, sorry.  Anyhow as everyone is aware, Moblin has since been superseded by MeeGo.  The timing was a bit of a PITA for me, as it coincided with the Factory freeze, which means I couldn't update the relevant packages etc in-time for 11.3.

As it stands, 11.3 should have an almost fully functional Moblin environment – I say 'almost' as there were two packages that I just couldn't get to co-operate and build (the modified Moblin browser, and the Web panel).  They should not affect your enjoyment of the UX so I'm not going to kick myself too hard.   I have learnt several lessons though trying to get Moblin into Factory:

1: Getting things into the next distro release is NOT difficult.

I don't care what anyone says, the process is very simple and straight forward.  Importantly, there is a huge amount of help from all parties that deal with the acceptance etc.  If you need help or guidance, just ask! I can't stress enough how much help is available to people.  I would personally like to thank everyone that helped, but I know I will miss some people off due to my feeble brain.  As a start these people have been invaluable to the process - Andreas Jaeger, Marcus Ruckert, Stephan Kulow, Michael Meeks, Henne Vogelsang, Pascal Blesser, Andrea Florio & Dominique Luerenberger.  Thank you all (and sorry to those whom I missed of listing).

2: You need to plan, and plan well.

I started off with great intentions, but I didn't take into account how disruptive my daytime job would be.  Unfortunately I haven't wopn the lottery and as such I have to deal with “The Man”, otherwise I have no food or house :-(  Part of my bad planning led to me missing the cut-off for getting Moblin listed as an installable option from the get go.

3: Spread the love.

There's a saying I've heard many times and from many different places – A burden shared is a burden eased.  Do you know what?  It's absolutely true!  Even if there are only a couple of you dealing with a project, it makes things so much easier, and having more is just plain fun :-)  I tried to get some of you reprobates to join in, but you were all too busy.  As such the pace of progress was fairly slow.

4. Have fun doing it. 

This applies to almost anything you do, but especially when doing any form of contribution. If you feel that it is getting you down, take a break, try and pass the baton to someone else to carry it forward for a bit.  What ever you do, don't let it burn you out!  Loosing a contributor to burn-out is disastrous for the community, and is something we should all be aware of and on the look out for.

So that's the past/present, what about the present/future?  Well Goblin will live on, and incorporate all the new goodness that is coming out of MeeGo.  I do genuinely think that MeeGo is a vast improvement on Moblin, but as always there's room for improvement; and what better way to start improving than by getting the Geeko involved? :-)  As my contribution to HackWeek 5, I have started the importing process for all the MeeGo packages, into the OBS.  Believe it or not, there are only a handful of packages that can be carried over from Moblin which is a bugger.  Regardless I think I have got pretty much all the needed pieces in.

Problem is the way I've done it is pretty hacky – I basically just  dropped the contents of the SRPMs in and hoped for the best.  This was done mainly for speed, and I know that it isn't good enough.  Problem is I'm trying to balance my regular job and dealing with managing the renovation of my parent's house at the same time, so I really don't have a huge amount of spare time :-/  So this is where YOU come in.

Please help me to not only get Goblin updated to the latest and greatest that MeeGo has to offer, but also prove to everyone outside of the openSUSE community that we are a coherent and cohesive force to be reckoned with!  Too many people discount us, and just say that without Novell nothing would get done.  Please help me to help you to prove this assumption to be a load of rubbish.  We have the tools and knowledge available to us, so let's get on with it and get it done!  Oh and we have the real opportunity to be a leader here – the first distro to provide a full MeeGo experience!

So if you can help, then please join in on IRC (#opensuse-moblin), and on the mailing list ( make sure you are subscribed first).  The first step is to get everything building, from there we can move forward.  Oh by the way, this initial push is for the Netbook UX, but I'd love to see use get the Handheld/Tablet UX when that is released ;-)  I'm tempted by Garrett's name suggestion of Smeegol, but let's get it building and working first ;-) 

If anyone has been following my tweets/dents, you will have noticed that I sporadically mention something about "Operation Factory" followed by a pile of tags.  Well for the curious, Operation Factory is basically getting the latest and greatest Moblin into openSUSE.

Basically the process involves the following:

• Get the latest code from the Novell internal Build Service, and import it into the OBS.
• Rebase all the patches and tag them
• Upstream as many patches as possible
• Get as many patches as possible into openSUSE
• Use as many in distro packages as possible
• Put Moblin specific packages in their own devel project on the OBS

OK so that doesn't look too bad now does it? Not entirely no, apart from the fact that my employer doesn't really want to know about what I'm doing and as such would much rather I spend my time doing annoying/boring/beneficial to shareholders work :-( (that's my obligatory moan out of the way).

Michael managed to architect a good plan on how to actually start getting things moving with the help from Aaron and others (I might have stuck my oar in just to feel involved :-) )

Vincent was kind enough to import from the IBS to the OBS, and started linking to distro packages.  I have pretty much linked all the remaining packages that could be with ease (as in there were no patches or differences).

Federico has done a lot of work tagging and highlighting patches that need to be sent upstream/included into openSUSE over in the wiki.  Problem is that he has other critical priorities (like fixing bugs/work beneficial to shareholders/etc) which means he hasn't been able to go through the whole list of packages.  So that means that I or anyone else needs to do the work on the Still ToDo part of the list.  I have started to merge the identified and "cleaned" packages, I still have a way to go but I'm hoping I can juggle things a bit so that I can at least finish that part of the list.

Another thing that needs to be done, is get all the Moblin specific packages submitted to their own devel project before they can be sent into Factory.  We have a devel home ready, it's just a matter of going through all the packages and sr'ing them there.

I have commented on the wiki against the packages that I have linked, patches upstreamed etc. to make it easier for people who may want to join in the fun (just incase no one noticed, that was a hint!).  Basically the repository setup is like this:

• Moblin:2.2 is the new home for building Moblin.  This will effectively be an aggregate for the different devel components.  This is also the repository that will provide the base to build the images for people to install (no I don't have a date for when these will be available yet).  It is also the home for packages that can't get merged into Factory due to too many patches or weird tweaks should there be any.
• Moblin:UI is the new devel project for the Moblin specific components.  This is where things like moblkin-panel-web etc. will live, and where they will be worked on prior to being sr'ed up to Factory. 
• All the other packages should be in distro, so the maintenece aspect of things should be much simpler unlike now.

The keen eyed among you will have noticed the name of the aggregate repo - yes it will house the new release once we have merged everything.  We are starting off with 2.1 and then we can easily see diffs etc between versions when they are submitted.

So all in all things are moving in a positive manner, just not as quickly as I would have liked.  If you are disappointed in the speed at which things are progressing pop into irc or send a mail to the mailing list (make sure you subscribe first).  Please do get involved!  There is lots of useful work for people to do, and importantly lots of fun to be had :-)