Openmoko build / Rebuild vs -dev

[Andy Green]

-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1

Somebody in the thread at some point said:

| Moblin/Ubuntu Mobile is not an alternative right now, despite signs of
| support nobody from Intel stepped up in this thread to make a case for
| their platform and build environment.

The Intel suggestion was a special case to consider where their fat
wallet could remove all thorns from the path of distro change, and we
could imagine to find overall advantage in the medium term.

Without that, there is nothing to change to right now that even starts
to offset the agony of making the change.  So if Intel are just kidding
about it, we can forget the issue of changing distro: except for the
long term (1yr +) we should keep an eye on things like cross Fedora and
the embedded Debian... or even they may come and keep an eye on us if we
sell lots of devices, who knows.  But we don't need to consider it for a
long while.

| OpenEmbedded is not tied to a particular package format, opkg is just
| Openmoko's current choice.
| We have identified two major weaknesses in Openmoko's current use of
| OpenEmbedded, details of which may be connected to weaknesses or missing
| features in OpenEmbedded:
|
| 1. -dev packages
| 2. Poky SDK

Since I am (trying to) start to use our existing tools from zero, I am
going to have some positive proposals about what we can do to make the
step for devs from a normal distro to using this less huge and improve
our own capabilities and performance.

I didn't guess it, but it seems most of what I have in mind is actually
about shuffling OM's use of what is in or will shortly be in OE already,
that's unexpectedly encouraging.

- -Andy
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.9 (GNU/Linux)
Comment: Using GnuPG with Fedora - http://enigmail.mozdev.org

iEYEARECAAYFAkgIRYEACgkQOjLpvpq7dMrs2QCfYpPF9GHjcfO68yVP2OibR4nJ
vjEAniyw4c+GU4EeLlS7P7EJQSZrAp3z
=n4wA
-----END PGP SIGNATURE-----