(GPL)
After my previous post about Openmoko 2007.2 distribution and the Openmoko community I got some comments proving that like me, some other people feel that the community is lost and uninformed. This post is a follow up to the previous post and the comments I got. And just to make it clear: I think Freerunner is a great piece of hardware and I really hope it’ll grow big!
I’ve understood that 2007.2 still ships with the phones. As soon as one gets his new c00l Freerunner linux phone they need to flash the phone with a new distro (2008.x or Debian, FDOM, SHR, Gentoo, Qtopia, FSO..) since the software it ships with is obsolete. OK, so you get your 2008.x there and find that it doesn’t really do all the magic and would like to report a bug.
Bugs
I don’t know how others feel but I really don’t feel like reporting bugs in the bug tracker. OK, I’m running an unsupported distro (2007.2) so there’s one reason but I also feel that it’s okay to post kernel bugs only there, not anything that a normal user would see in his GUI. Is this correct or just a feeling or misunderstanding? Remember that all Freerunner owners and people here in the community are not kernel developers. I don’t even know how to save back traces or hack the source code but I’m happy to report if something doesn’t work and then when someone more skillful finds what’s wrong and eventually fixes it, I’m again happy to try if the fix works for me. For some reason I feel that there’s no space in the community for users like me.
Distros
So it’s 2008.x now. No, wait.. there are FDOM and SHR around that are not far from 2008.x. Is 2007.2 still developed and lead by Openmoko or by the community? Do the community developers have access to the files so that new releases can be made or is SHR the closest to 2007.2 you can get to? My feeling is that there was a lack of information about why 2007.2 was forgotten and new 2008.x series was started and that something weird is happening in the official distros that creates a need for FDOM and SHR – why don’t the developers feel ok to contribute directly to 2007.x and 2008.x but ‘fork’ their own distros?
Information
I think that what makes me feel lost is the lack of information about what’s happening in the community. I don’t know what’s the development status of the software or what’s the general direction the community and/or Openmoko is heading to. Having a better view on the general situation makes one feel much more comfortable and secure: now I feel that I just wait to see what the next release’s like not being able to know what to wait for.
Next steps
People working at Openmoko and other software developers:
- Please check ‘Community Management as Open Source’s Core Competency‘ by David Eaves and have a good look at the Openmoko community. It needs management!
- Please write a blog post once a week or so to planet.openmoko.org and the community mailing list telling the community what’s going on. Five to ten lines is enough to help the community feel better!
- Please respect your community!
Community members:
- How do you feel about the current situation? How happy are you with the community?
- Do you feel well informed? Do you know what’s the direction we’re heading to?
- How would you like to contribute?
Thanks!
This message has been posted also on the Openmoko Community mailing list.
Related posts:
- 2.12. Tämä viesti on osa vuoden 2008 adventtikalenteria. This post is...
Re: Information
You can find weekly news at:
http://wiki.openmoko.org/wiki/Weekly_Engineering_News
Re: Information
Someone’s doing a regular online development review of 2008.x:
http://onlinedev.blogspot.com/search/label/om2008.8%20review
I have more or less the same feeling. The OM Community is not as friendly as it could (althought I have to say – I think that my national community is very good).
1. We have 4 stacks. 2 of them are unstable (FSO, ASU), 1 is obsolete (2007.2) and last is borrowed (QT Extended). 2007.2 was enough good to keep it in place. If we really needed to move somewhere we should do it to one place – possibly FSO. Then we could slowly ‘close’ the 2007.2 – simillary to what Torvald done to 2.2 and 2.4 – as far as I remember the 2.2.something was released at the beginning of the 2.6.
2. I’ve tried sometimes to ask on #openmoko with relativly simple questions. Hardly ever I was responed. I understend it if there is 19 people on channel (possibly they forgot to whitch client off etc.) – but with >300 it creates bad feeling. Also I do not see what should go where on mailing lists. The tech support is mixed with announcies etc. It can be done if there is low activity (<25 posts/day).
3. Don’t get me wrong. I’m really happy I’ve bought OM and it is the best phone I ever had. But it could be much better.
I couldn’t agree more: all points of this blog post. The worst in this all, that normally I DO know how to get a backtrace or hack a source code. Not in Openmoko. I failed with everything I tried.
If Openmoko Inc. switched to 2008.8 from 2007.2, I respect their decision and I want to help with actual code bug reports, not something old. So I did switch too. (I can still run the original Openmoko apps here, I think they could be developed and included, regardless of the changes. For example Openmoko-mediaplayer2. I don’t care about the things which are completely replaced and not available or needed anymore (Today/2007.2?). Fine. Let’s move on! Let’s upgrade to Openmoko’s current code!)
I have bought Neo1973. I never able to get it running good enough to become my test phone. Now with Freerunner is out I have the feeling that it is not supported any more. I still believe that Neo1973 could be very useful as development platform, but I can’t figure out which distribution will work on it. One of benefit of Linux it that can run on old hardware, but right now I feel completely abandoned by OM.
OK, I’m going to play devils advocate here. All of the following is just my opinion and I am by no means an expert.
From what I understand 2007 was dropped because it was GTK only and it was felt that the platform should be more open and allow all toolkits (Qt and E). 2008 pulled lots of stuff from Qtopia in order to turn the freerunner into a useable smartphone as soon as possible. I guess this hasn’t gone a well as was expected but I feel like it is getting there and everybody was warned over and over again that this is a development release and not a replacement for a commercially available smartphone at this stage.
Some people really objected to the dropping of the 2007 apps and wanted to keep them so they started the SHR project to keep these apps alive. This has nothing to do with Openmoko Inc.
FDOM is just 2008 with extra apps and a few tweeks included, it can be compared to something like mythbuntu (a version of ubuntu using xfce for running mythtv) where optional packages and configuration is done for you to save time in setting up the sort of system you like. It’s not really a fork and will become unnecessary as time goes on.
FSO is the future but it is not intended as a full distribution it’s a layer to allow applications to interface easiliy and cleanly with the hardware. In an ideal world this would have been under development from the beginning but it’s easy to criticise with the benefit of hindsight. FSO is also not just about the freerunner it is intended as the model for future hardware (not just from FIC) to allow trivial porting of apps to all “FSO compliant” phones. I think that the next “big” release (2009?) will use FSO as the base and be the start of the “final” Openmoko distribution, but there will always be alternatives, that’s what open-source is all about.
Creating a truly open-source phone platform is a monumental task, Openmoko are battling closed-hardware and every user wants something different. It shouldn’t really be a surprise to anyone that there has been fragmentation, this project will take years but has the potential to make a real difference to mobile computing and communication.
I am not saying that everything is perfect, clearly there has been some in-fighting at OM and good people have come and gone because of it. I’m sure that it’s difficult to get an answer out of OM but that’s probably because there are a couple of dozen of them and thousands of us They want to spend their time fixing bugs and writing code and not answering emails. They don’t have an army of tech-support monkeys like the Nokias of this world to get back to everyone with a personal reply. Yes this is frustrating when you have a “simple” problem but also understandable if you think about it from their side.
OK, I’ve rambled on for long enough.
I also couldn’t agree more: all points of this blog post. I don’t have my Freerunner yet but I’m already lost with all these distro.
@Kombipom: 1. Since 2007.2 has X what was the problem with porting QT?
2. AFAIU[1] the post is rather about PR/communication then technical problems itself. I may agree that movement to ASU/FSO was needed. But I would rather see an evolutionary model – keeping everything +/- stable andeasy to migrate. Revolutions are not boring. Revolutions may be very good for PR – but may be very bad (see KDE 4.0).
[1] AFAIK s/know/understend/ig;
@ Uzytkownik. I’m not an expert as I said, that’s just the impression I got from the OM guys when the change happened. Also I think the Qtopia phone stack and 2007 phone apps couldn’t co-exist as they had different low level interfaces to the hardware that clashed. I think that there were a lot of issues with the original gsmd and neod daemons which were sidestepped with Qtopia until FSO steps in to replace it properly.
2. I think that you could say that KDE4 has been very bad PR, people don’t like change and obviously we all would prefer stability but sometimes revolution is required in order to deliver the greatest potential for the future this applies to KDE and OM and in both cases everybody was warned that this is just the start and that there is much work to be done. You can’t build a castle on the foundations of a garden shed, you have to start again and dig a big hole and make a mess first.
@ Uzytkownik: 1. I was saying about QT not QT Extended (AFAIK it’s based on fb). Possibly the merge was needed. Possibly SHR was the right direction (I don’t know this distribution).
2. Well – I prefere the Gnome model. By small steps but it changes (gio vs. gnomevfs, remove of gnomeui, libgnome, gail etc.). Of course sometimes it is not possible – but it creates fustration.
s/@ Uzytkownik/@Kombipom/g;
The bugtracker is open for any kind of Openmoko (related) bugs (in the worse case you will have to file the ticket again or it gets closed as duplicate). If in doubt file a new ticket (way better than writing off topic to an existing one), feedback is required and mandatory and the bugtracker is an excellent place for that.
Well said. Openmoko sold a vision, and found an eager audience.
Unfortunately, they didn’t seem to have any idea what to do with that shared vision — as a result, the community gradually dispersed and fragmented.
Like a group of leaderless soldiers, the community has gathered around various rag-tag images and distros, trying to figure out the bits and pieces themselves.
Openmoko seems to have retreated, overwhelmed by an ever-more-restless community, they have hidden themselves behind a bugtracker fortified by draconian rules for valid bug submission. With a few notable exceptions, none of them appear on the IRC channels to speak with their army of recruits, preferring the less-direct means of communicating on the overloaded mailing lists.
–
But all is not lost, by any means. As long as people complain about the situation, there is hope — they still care. Openmoko needs to look very closely at the data that only *they* can see:
Openmoko — how many are subscribed to your email lists? How many of them are active (i.e. have sent an email in the past 6 months)? Now, how many phones did you sell, Openmoko?
Openmoko — where are those thousands of missing phones? Are they in desk drawers? Cardboard shoe boxes in closets? Dust bins? More importantly, where are those thousands of potential contributors, potential ambassadors, your in-the-field sales force for selling both the vision and the phones?
The hardware and software problems are trivial in comparison to the loss of the community mindshare — and ultimately that goes beyond a vision and goes to the bottom line. Openmoko had better hire a community manager, not for the community’s sake (although we’d certainly welcome such a person), but for their own business survival.
Openmoko: The community is your sales force. Act quickly.
I agree with about 99% of your points here. I ranted about it in a long blog post on my site (also syndicated on planet.openmoko.org) at http://www.monochromementality.com/index.php/blog/show/FSO-Milestone-II-Phot-Safari-and–rant.html
I’m not entirely sure I blame Openmoko. I mean, the Openmoko brand is certainly going to be what takes the hit, but I get the sense that there were internal factors (like investors threatening to pull money) that caused several decisions to be made.
I’m ready to turn my back on Openmoko as a vision and a company I spend money with, yet at the same time, I’m holding out hope. I did buy into this knowing it was in development. I don’t hold any illusions that I was given a time frame for a working system. I think I’ll wait until Freerunner gets a “stable” release OR my need for a new phone/device forces me to buy a competing product before the stable release.
I’ve been following the OpenMoko project for a long time now. I bought into the promise of an open and free linux-based smart phone. I was sold. Then I waited. And waited. And waited. And waited. Communication has been very infrequent. And worse, the project doesn’t look like it is making any progress. The hardware looks sound. But the software isn’t even close yet. And with multiple distributions splitting the community’s efforts, there seems to be little chance of the software getting here any time soon. Look, other smart phone manufacturers can produce a new phone in less than 18 months. Hardware and Software from design to customer shipment. We have the hardware. Where is the software? We need to coordinate our software efforts so we can get our first working version released. Then, we can work on various priorities.
Pingback: Discussion 1/3: Openmoko community feelings | Risto H. Kurppa
Pingback: Discussion 2/3: What people want Openmoko to do? | Risto H. Kurppa
Pingback: Discussion 3/3: Openmoko Community Manager position? | Risto H. Kurppa