Re: Hurd state

[Alfred M. Szmidt]

   I know this will require deep changes on the Hurd, and that it will
   take time. But i am unable to find proper documentation/guidelines
   for the task.

See the hurd-l4 module in the Hurd CVS repo.

   - There are some official (commonly accepted) strategy for the Hurd
     development?

Fix what you consider broken, add features where you think they are
missing.  Is what I have always followed, and has always worked.

   What is the problem with the Hurd? Lack of proper direction? Lack
   of hackers? Lack of documentation? Lack of ...? What is the
   problem, really? We need to know it in order to fix it.

There is no problem, but the lack of imagination from those who wish
to hack I think. I never really considered it a problem to know what
to hack on in any free software project, you fix what is broken, and
you implement what is missing.

   I am ready to hack, but i should know what to hack.

Fix ftpfs so it works correctly and well.  Use GNU/Hurd as your main
system, report bugs, fix them.  Fixing nfs (mmap support) and nfsd (so
it doesn't crash if you do something) would be nice.  Write
documentation, update the Hurd reference manual.

And that is just the top of my head, if you want more complex things
you can implement libchannel (libstore for streams), and add support
for sound, USB or whatever.  Or you could help out with Hurd/L4.