On Monday, 3 January 2022 13:06:39 CET Dr. Michael Lauer wrote: > Thanks for churning up these interesting comments, Paul.
I appreciate the feedback, particularly from someone who has been at the heart of the action. I was reading through your retrospective on Openmoko yesterday, and it still makes for interesting and pertinent reading: https://www.vanille.de/blog/openmoko-10-years-after-mickeys-story/ This is relevant to something else I wrote recently: "One of the OpenMoko special features was the so-called phase 0, in which we sent out a dozen of Neo1973 pre-versions (for free, without any obligations) to well-known people with a history in open source. The idea was to get some early feedback, perhaps even some contributions, and have those people spread the news about it, hence act as some kind of multiplicator." "On 25th of February 2007, we shipped the Phase 0 developer devices – and even though we didn't get as much feedback as we would have liked, we felt it was a good practice to do this." I have to wonder how many of these devices were initially or subsequently seen as trophies or collectables and never got any real use. It is certainly an awkward calculation to make: how to get those with expertise interested and involved whilst not forcing them to pay to help the initiative. Some might receive a device and feel an obligation to do something, regardless of whether this was intended, whereas others might feel that they didn't really ask to be sent anything, or that it was all very speculative and that they need not feel any such obligation, anyway. Meanwhile, there is also an argument that there are people who will buy a device to improve it, and in doing so they will have made a conscious choice to invest in its future. Getting people to put down even a token sum might deter anyone looking to just collect some more gadgets. > It’s extremely sad that after all these years with all our technological > advancements and choice of free software projects, getting the very > basic task – solid and reliable phone calls with tolerable > audio quality and battery life – right, still seems to be unachieved. Yes, we're back to the featurephone here! Maybe it needs doing after all. > As much as I love hacking middleware and user interfaces, I never got the > hang out of kernel and/or hardware programming. Even back in the > Openmoko days, I always had the feeling that the kernel and userland people > did not work together close enough. There were certainly people missing who > had a holistic approach, trying to work across the stack to provide an > optimal user experience. It seems that the realisation that a stable foundation was necessary only arrived fairly late on in the Openmoko lifespan, meaning that things were starting to look a bit more productive, and then the funding was pulled and everyone was left to keep the initiative going by themselves. I would have hoped that other initiatives might have chosen to build on that work, but efforts within Nokia like Maemo, for example, were already committed to doing their own thing. > I always thought this would mainly be a problem with our chaotic > community-based self-organized structures, but not with corporate ones… I know that Qtopia, as QtMoko, got used on the GTA04, but that framework was another initiative that was abandoned by its corporate parent. And the whole Maemo, MeeGo progression is a good illustration of corporate chaos. > That said, if the hardware and the kernel does not provide a certain > quality, you can only do so much in userland. And it looks to me like the > problem is still mainly that. I don't really follow the work that is done these days. As I noted in another message, I don't think that the way the Linux kernel is developed and architected is particularly conducive to sustainable or focused development, but the fact that lots of people spend lots of time and effort pushing it along masks the inefficiencies involved. I concede that Linux contains a lot of functionality that is widely deployed and that does offer a quick route to market, which tends to involve vendor kernels and people hoping that they got a snapshot of functionality that works for them. A lot of that functionality could arguably be more portable, however, meaning that it would not be glued into a monolithic kernel-specific framework. But I don't there is much appetite to do things differently. Personally, I think that device-specific software architectures could be interesting, but even in the microkernel realm it seems that a lot of people are happy to put Linux on top of a microkernel in some kind of virtualised form and to arrive at a supposedly fully functional system that way. Paul _______________________________________________ Community mailing list [email protected] http://lists.goldelico.com/mailman/listinfo.cgi/community http://www.tinkerphones.org
