Re: [OpenBIOS] OpenBIOS: reality check

[Florian Wunderlich]

> FW = Florian Wunderlich 
> FW> Filesystem code in a _BASIC_ input/output system is not a good idea
> IMHO.
>
> The firmware on IBM-PCs can no longer be correctly termed solely an
> input/output system.

The emphasis was on BASIC, as shown with upper case letters, and maybe we
should remind that fact.
What I wanted to express is that FS code does not belong to the BIOS or M/B
firmware or whatever you want to call it. It depends on the operating
system, not on the hardware, and in consequence, it should be implemented
in the os.
> OS> Note, I did not say I think it should be part of the BIOS code, just
> that it
> OS> should reside in ROM.
>
> With all due respect, this is a technical mailing list, not an English
> lesson.  :-)  We are not computers.  We can (hopefully :) figure out
> what other human beings mean without needing literal terms.

Huh.
May I repeat that? :-)

---
OpenBIOS -- http://www.linkscape.net/openbios/
[e-mail]   Body: un/subscribe
Problems?  [e-mail]

OpenBIOS: reality check

[Benjamin Scott]

FW = Florian Wunderlich 
BS = Ben Scott  (me)

FW> Filesystem code in a _BASIC_ input/output system is not a good idea
FW> IMHO.

BS> The firmware on IBM-PCs can no longer be correctly termed solely an
BS> input/output system.

FW> The emphasis was on BASIC, as shown with upper case letters, and
maybe we
FW> should remind that fact.

Why should we limit ourselves to something which only basically
accomplishes our needs?  Let's make it an advanced input/output system,
then.  <grin>

FW> What I wanted to express is that FS code does not belong to the BIOS
or M/B
FW> firmware or whatever you want to call it. It depends on the
operating
FW> system, not on the hardware, and in consequence, it should be
implemented
FW> in the os.

I agree, in general, except that it would make booting oh-so-much
easier.  Wouldn't it be nice to be able to boot a Linux kernel simply by
specifiying device, partition, and kernel path and name?  Sure beats
saving physical disk locations to the MBR every time you recompile.  :-)

Some other platforms include such basic real-filesystem support in the
boot firmware.  I think we should, too.  My opinion, of course, but
others have expressed a liking for it, too.

OS> Note, I did not say I think it should be part of the BIOS code, just
OS> that it should reside in ROM.

BS> With all due respect, this is a technical mailing list, not an
English
BS> lesson.  :-)  We are not computers.  We can (hopefully :) figure out
BS> what other human beings mean without needing literal terms.

FW> Huh.  May I repeat that? :-)

Just that the person was making a big deal out of the fact that BIOS
stands for "Basic Input/Output System", and thus if we have code that
isn't strictly for basic input/output, he claims it's not part of the
BIOS.  And then went on to complain about calling "flash ROM" ROM
because it's not read-only.

I would hope that we're all smart enough to adjust to the "BIOS" being
more then a basic input/output system, and that modern "ROM" isn't
read-only.  :-)

We're here to improve things, not argue semantics, no?  :-)

-- Ben

---
OpenBIOS -- http://www.linkscape.net/openbios/
[e-mail]   Body: un/subscribe
Problems?  [e-mail]