Joey Hess <[EMAIL PROTECTED]> writes: > Mario Lang wrote: >> * speakup: A co-maintainer on this one would be wonderful. I once bought >> a hardware speech synthesizer for testing speakup, but since >> my primary output medium is still braille, it doesn't get as >> much attention as I'd wish it should. Besides, I always >> had those strange lockups when speakup tries to deliver a lot >> of speech to the serial port, which very badly interacts >> with my other pet interests, namely low-latency audio work, >> so I had to finally stop using the prebuild speakup kernel images >> myself... :( > > FWIW, this kernel varient (kernel-image-speakup-i386) also currently > FTBFS and has dozens of unfixed security bugs (which are fixed in the > general Debian kernel source it builds from), so it's looking quite > likely to be removed if these RC bugs arn't addressed sometime.
Yes, I am aware of this, and I'd like to thank you for your previous NMU joey. > Fixing it is probably pretty easy, it just needs to use an older gcc for > building. I'd rather prefer to use a newer speakup which works properly with gcc 4. The current patch used is pretty old, and adding yet another workaround doesn't feel like moving much forward. Besides, It'd probably make sense to switch to 2.6 for kernel-image-speakup-i386 at some point. This will involve quite a bit of d-i hacking I guess... I'll seriously try to allocate the required time to get this done as soon as possible, however, the offer still stands, if someone wants to work on speakup, I am very happy about this. Again, I am aware that I fucked this kernel package up pretty badly due to nonresponsiveness, and I'd like to publicly say sorry. And thanks for helping out at least a bit joey. -- CYa, Mario
pgp3XkMiGm1dw.pgp
Description: PGP signature