Re: prepare_0_5_96

[OKUJI Yoshinori]

From: Alessandro Rubini <address@hidden>
Subject: Re: prepare_0_5_96
Date: Mon, 2 Oct 2000 10:28:22 +0200

> Again, sorry if my words looked harsh.  I should have dropped the "in 2004"
> above. I'm not complaining that development is slow.

  I'm not angry with you at all. I just pointed out the fact. I
believe that if you want someone to do something surely, the only way
is to give the person good value for that. So, for example, Conectiva
hired Andrew Clausen to let him work on Parted. (But note that I don't
insist on that you should hire me, but I just want to explain why I
cannot promise anything I may do.)

> As I stated some time ago, I want to actively help in making grub
> cross-platform, aa I have several platforms to port it to. So I was
> interested in the project schedule before I can jump in.

  I think I see what your point is. But I'd like you to understand
ours, too. For example, you sometimes say, "Don't add anything
dependent on PC!", but I don't want to hear such a thing, since you
don't take it into consideration that we have our own needs, as you
have your own needs. We are working on adding new features into the
Multiboot Specification, because they will be really useful for us,
even if they will be PC-only features. Personally, I don't think it is
bad to add/implement features dependent on a certain architecture,
because this wouldn't make the whole implementation dependent on the
architecture. We just need a way to clealy distinguish which is
architecture-independent and which is
architecture-dependent. Actually, the Multiboot Specification is quite
dependent on PC for now, but I didn't think this was wrong, because
no demand on multi-architecture existed when it was born. Nowadays,
multi-architecture is desired, so I'd also improve it radically some
day. However, now is not the time, since we need minor upgrades for PC
architecture rather than so major changes to achieve a
multi-architecture specification, at the moment. We can't move so
fast.

Okuji

P.S. Do you know how many operating systems have already used the
Multiboot Specification 0.6? I recommend that you survey many other
operating systems than Linux and see how useful the spec is. If you
want to be involved in the development of the spec, you need to have
deep knowledge on various types of operating systems, anyway.