Yow! Are we having fun yet?

My goodness, I was away for a few days and lots has happened.  This is 
excellent.

It will take a few days for me to get caught up, but I should say a few 
things quickly before I run off to a client for the day.

First, thanks very much to everyone who has contributed code and 
drivers!  Broader hardware support is a good thing, and we appreciate 
your efforts.

Next, I am pleased to announce that Michael Brown has volunteered to be 
in charge of i386 core and bus code.  He is currently hard at work on 
what will be 5.5, and says he'll be populating 5.5 CVS when he has 
something a bit more stable.

Also, Timothy Legge has generously agreed to continue being in charge 
of drivers, and driver additions and changes should generally go 
through him before being CVS'ed. (Besides the fact he is one of two 
people who currently knows how to do a release :)

Oh yes, I'm looking for someone who has an interest in helping maintain 
rom-o-matic.net.  It's PHP code on an Apache server.  Please send me 
mail if you have the interest, time, and desire to help.

Looks like we'll be needing a 5.4.1 pretty soon.

I realize I haven't responded to YhLu's excellent questions about 
LinuxBIOS support in Etherboot going forward.  I have not forgotten 
about that, or about Memtest not working.

In general my feeling about supporting architectures is that if someone 
is willing to be a part of the development process and maintain the 
code in a cooperative fashion, then it will be great to have it a part 
of Etherboot.  This goes for all architectures and features.

I would really like to get the HTTP and PCMCIA code working well in the 
next production release, for example.  Those are features that would 
really make Etherboot desirable in more situations. We can arrange for 
hardware donations and other support as needed.

I think there has to be an incentive for people to put forth the effort 
to maintain code, and I think the main incentive should be that we 
simply won't put non-working, non-tested, or buggy code in our 
production releases.  If we keep carrying over code that doesn't work, 
we just make the distribution bigger and more complex, without a lot of 
benefit.  People will also lose confidence that things work, if it's 
hit-or-miss.

So, by the next US Linuxworld in a few months (San Francisco, August 
2005), I'd like to us to do another production release composed of 
stuff we know works, and hopefully with a group of testers that have 
beaton on the code pretty well for interesting PXE cases, buses, and 
architectures. (and anything else people care about)

Gotta run.  I'll try to catch up on specifics later today or tomorrow 
morning.

Thanks again to everyone who's helping out.  It makes me feel like I 
wasn't crazy to think I could do some good when I volunteered :)

Marty



-------------------------------------------------------
This SF.Net email is sponsored by: New Crystal Reports XI.
Version 11 adds new functionality designed to reduce time involved in
creating, integrating, and deploying reporting solutions. Free runtime info,
new features, or free trial, at: http://www.businessobjects.com/devxi/728