Re: Half-year review of Stskeeps's work as distmaster
 

You have no idea how much that frustrates me as well.

Right, so, going all the way back. Mer started as a research project (Mer is Maemo Reconstructed afterall) between friends. We were not happy with the way things were in Maemo and didn't see their model of development and the platform to succeed if things were not changed. Things did change - very similar to our original ideas, in MeeGo, finally. We began to reconstruct Maemo from bottom up - taking the open source pieces and putting them back together in an interesting constellation.

To that respect, we started an iterative process, let's see how far we get if we churn at the problem. Involve more people, churn further at the problem. Now, a lot of work, discoveries and process was developed in this - much which is actually reused in MeeGo or exists in some form now.

The problem was however that the goal post kept moving. This is the same problem for instance GNOME suffers - the goal post of being a 'Windows alternative' keeps on moving as well - ie, what users want. Without day to day collaboration in a platform, we can never reach eachother. That part is too, changed now, in MeeGo. Tracking a mostly closed source distribution is darned hard.

Now, the things that really screwed up things and where we made bad choices:

A research project taking on the task as a Fremantle backport - there was noone else to do it. We had some goals with Mer and this did not always agree with the backport idea and staying close to Fremantle. Things we -should- have done, which I was trying to recitify with Mer^2:

* Ask Nokia to maintain a ARMv6+VFP build of Fremantle alongside the ARMv7 one from early on.
* Activate GTK+ developers in Fremantle alpha to make a look-alike of Fremantle in the old hildon desktop. People have made themes to this regard now, so it was possible.
* In the development process, engage Nokia very early on to get a redistributable binaries for Nokia devices. In that regard, we were perceiving Nokia as slightly more evil than they actually were. Imagine our shock meeting Quim in May in Copenhagen and discovering that they were actually rather relaxed about Nokia binaries for Nokia devices, when we had assumed there was a huge legal framework needed to get things accomplished.

We were operating with a huge bunch of enthusiasm and we developed a lot of things to support a open platform like Mer. Now, this didn't always accomplish the goal of a backport.. it was a hope to keep things maintained than just a backport. That your devices wouldn't go out of fashion. MeeGo's a way to do this, still - and our methods are being used there.

Another thing was the announcement that Nokia was switching to Qt as their main toolkit. It meant that most of our work on GTK+ tools and desktop was well, mostly wasted.

We kept on churning, actually gaining some degree of functionality and usage - except that we couldn't possibly match Fremantle development and functionality with the people we had. And that we didn't know how Fremantle would look like in the end. With Fremantle release, we were not there yet either. And most of the things people wanted, was closed. And we couldn't easily integrate due to previous mistakes.

Now, on to the summit: 300 N900s to the attendants. Most of them the core of N8x0 developers. Bam. Attention to N8x0 dropped like a ball. Who needs N8x0 when we have a shiny new toy? For the rest, I tried to finish up things, but the influx of new contributions had dropped. Most contributions actually came from other device communities.

And then we get to where MeeGo is announced. Mer started like a research project and the experience we gained helps us to be first class participants with skill and merit in this project. Did I waste a year of my own free time and people's time? No, because we learnt things and produced a lot of know-how and tools. How a open mobile OS could be like. And how to tame the N8x0 hardware wise, which we can now use in the MeeGo N8x0 hardware adaptation. We got -really good- at porting to devices.

So, post-mortem of Mer. It's dead. Has no future. But our ideals, ideas, skills and experience still lives on. And has enabled us to be professionals in our own regard in mobile OS'es. Would we have been this without Mer with the situation of OSes on N8x0 before we came to town?

I don't think so. It's time to look forward now. I don't personally use my N8x0 anymore except for random tasks. Why do I still contribute to it and want to? Because I like the device and I know how to do things that others might find worthwhile.

Re: Half-year review of Stskeeps's work as distmaster
 

I think you way overstepped your bounds General, Stskeep was asking what peoples perceptions (good AND bad) were of his own work.

Talk about shooting a messenger (you don't have to agree with him, but shut up about it). How many others do you think will now be honest with Stskeeps when you come charging in and just muddy the waters as usual. I might not agree with him and his ratings of Stskeep; but at least he was being honest.

A BIG Thanks for making people decide it isn't worth being honest. I'm sure that is exactly what Stskeep wanted. Good job!

If my sarcasm escaped you, I'll be really blunt for you -- you really blew it. I'm not even going to respond to the rest of your dribble since it was totally inappropriate -- given he was answering HONESTLY what Stskeeps asked for. A big two thumbs down on your responses. That will probably be the LAST honest response stskeep will get thanks to you.

Nathan

Re: Half-year review of Stskeeps's work as distmaster
 

Ok, here is what I wrote before Stskeeps' most recent reply, but it still seems relevant to me:

Thanks for repeating the explanation. (Referring to GA's comment.)It is just hard for me to understand that Mer was in such bad shape when it seemed to be abandoned. You could actually use Debian programs, importing them with ease -- some at least, I did it. What a dream! All I missed was integrating them into some sort of usable menu system, and there was a crashing problem.

I almost abandoned the regular system and tried to go over entirely to Mer.

So the screwed up drivers prevented an integrated menu system?

My suspicion, which is not based on any knowledge, is that someone fixed their attention on the drivers and then the project was abandoned in a fit of pique, "well, if we can't have what I wanted there, we have to junk the whole thing!" even though there seemed to me to be a lot worthwhile about the project.

The above is probably unfair, sorry. My mind will just not accept that such a beautiful project could be junked, and then we are asked to believe that "oh well, we'll do something similar with Meego."

I give up, but the Dylan lines come to mind,

"I ain't saying you treated me unkind. You just kinda wasted my precious time. But don't think twice, it's alright."

Re: Half-year review of Stskeeps's work as distmaster
 

The problem was that the old desktop was crashing left and right and did not support things like stackable windows, it conflicted with applications wanting to make desktop applets. I am not a GTK+ hacker and there seemed not to be people willing enough to dive into it.. So, we experimented with other desktops but returned to a lot of the technical specifics of the Hildon system.. Yes, the GL drivers were supposed to be a saviour but things really screwed up there as we all know.

Re: Half-year review of Stskeeps's work as distmaster
 

I followed MER, and installed various versions on my N800 to test and play with, which is about as much as I can lend a hand with.

I have always felt, and still feel, that Nokia let the N800 and N810 camp down, and should have put out a "maemo 4.5" with as much of the good stuff from Maemo5 as could be mustered. Stuff like the photo editing/sharing/tagging, media player and tracker, calendar, and some of the IM stack. Personally I think diablo is good enough, but the apps needed development, and backporting, rather than a whole new OS. Granted there may be loads of dependancies within the core OS which may prevent this however.

Mer was a wonderful community effort, and I do appeciate the work that went into it, but it was a very tall order to ask and expect a handful of talented people to make it a viable alternative to Maemo5 on the previous gen hardware.

It is useful and enlightening to read about some of the work and effort which has gone on behind the scenes.

While this should not turn into a blame thread, if there is any blame to be laid, especially for the lack of an "upgrade" path for Diablo, then it rests with Nokia, and the Maemo council for putting too many expectations on Mer.

I should add BTW, that the work launched by STS, and carried on by others regarding the community SSU, pulling together "orphaned" bug fixes etc from Nokia and elsewhere and pulling them together into a update for Diablo is appreciated and is a positive step forward, and a whole lot better than Nokia's "fixed in fremantle" mantra, which really irritated me, especially now when it seems a lot of bugs were fixed in diablo, and would have never seen the light of day without the community SSU.

I would like to request that STS, and the council push to have some of the Maemo5 apps backported by Nokia if possibel, and not too late.

Re: Half-year review of Stskeeps's work as distmaster
 

So, Mer is dead and you have learned something from it. Good for you.

I also learned something from it, I'm selling my n810....

A lot of the stuff you listed as done was actually wasted time, any normal company would have you fired. Hope you do better in the future.

Re: Half-year review of Stskeeps's work as distmaster
 

Well, that was a bit harsh. Good luck finding a seller for the N810 - I hear they still go for a fair bit of cash.

Re: Half-year review of Stskeeps's work as distmaster
 

Stskeeps, thank you very much for all the effort you have put in this, and thanks for being so sincere.
I was very dissapointed too with the (surprising for me) turn of events regarding Mer, but now I understand.
I hope meego can reach the n810, finally, but just the ongoing community SSU is great (I can't thank enough lma for the work he is doing). Given that the n810 has lost so many core community members to the n900 (in just a day), the work you (all) have done since is even more laudable.

Re: Half-year review of Stskeeps's work as distmaster
 

I like to add stskeeps' pushing bme and especially USB-hostmode in direction to a open resp usable state.

Re: Half-year review of Stskeeps's work as distmaster
 

The work you have done has helped me improve my work and inturn helped improve the community even more. When I first started helping on Mer most of my themes were lacking in many ways. When I started working on my Mer theme you helped me you really pushed me to turn out good products. I got to get more involved in the community and in turn wanted to do better work. For me the single best thing you have ever done is get the Maemo5 Theme Template opened up. It made it possible for me to do some really quality themes that I could not have done with the Theme Maker. If anyone is using just about any community theme on their N900 they owe Stskeeps a thanks for getting the template opened up. Keep up the good work.