Tim Abell <[email protected]> writes:
> Hello to all potential and existing maintainers, shr-u devs,
> administrators, package maintainers, application programmers etc,
>
> I really want to breath life back into shr-testing.
Hooray, that could be what I really want to do too - but it depends what
you mean.
If you mean forking from a recent shr-t release, and cherry-picking
known-good fixes and enhancements to it conservatively - yes, that's
exactly what I'm looking for.
If you mean taking a snapshot of shr-u every now and then when it's in a
good state, then I'm less interested. I don't like the Illume 2
interface, and I doubt that that approach will ever deliver stability.
I'm aware that that may sound like a bit of a downer on the approach of
the main SHR team, so let me try to counter that in two ways.
Firstly, I've realized over the last few weeks how much I really _love_
my current shr-t. For me it's beautiful, stable, works reliably as a
phone, fun, and interestingly different (and freer of course) from what
all the Iphone and Android fans have. I'm regularly using phone
function, sgt-puzzles, ventura, wireless, GPS, GPRS, podboy, cellhunter,
and it all works. I like the SHR settings, especially the Quick
Settings for disabling dimming when solving puzzles. I like the Illume
left and right arrows for cycling through open apps.
There are just two inconveniences (the SMS focus bug, and iliwi not
working first time), but I'm used to them now and they're easy to work
around.
If I could have just small incremental improvements from this point,
that would be perfect. First fixing those bugs; a faster kernel if it can
be dropped in without having to change everything else; gradual
improvements to the messages UI maybe.
So, a huge thank you to the team for already providing something so
close to perfection.
Secondly, over the last couple of weeks I've been playing myself with
the OpenEmbedded system, and I can see now that the whole system is set
up so as to make taking a conservative approach practically impossible!
There's of the order of 2000 recipes there, and most of the ones that
I've looked at are written so that they always download from the latest
VCS tip.
In other words, it's bleeding-edge bandwagon city, and it's very
difficult to get off the bandwagon.
Or so it seems, anyway. Maybe I'm missing something - in which case
please let me know, because I've trying for a while to recreate a build
to build an image similar to what I have on my phone, and haven't yet
succeeded.
Probably, following the bleeding edge was the right thing to do before,
when fewer things were already working, and E and Illume and Vala and
FSO2 were still being worked out. But now it seems to me that the
disadvantages outweigh the advantages. Perhaps the core SHR developers
have become so used to this sysem, that they don't see that it has now
become a case of the tail wagging the dog?
In summary, two points:
- I strongly suggest that the right thing for SHR now is to jump off the
OpenEmbedded bandwagon, and switch to a more conservative mode of
development
- I'm personally interested in taking that approach myself, but so far
without success - so if anyone can give me any clues as to how to do
it, I'd very much appreciate that!
Many thanks, and best wishes,
Neil
_______________________________________________
Shr-User mailing list
[email protected]
http://lists.shr-project.org/mailman/listinfo/shr-user
