Re: Beta vs. Stable

On Fri, 8 Nov 2002, Aitor Santamaria Merino wrote:

> (1) Years later I have to say that the wishes which at that moment
> believed to be essential and neccessary (maybe sufficient?) for a
> hypothetical FreeDOS "1.0" seem to have been fullfilled already. I just
> remember that these wishes were

Just a few cents here: I remember this discussion from 1.5 years ago I
think.

It all depends what you call "1.0". If you want to judge stability then it
seems to be there -- I haven't seen FAT corruption for a long time and it
doesn't crash more often than other DOSes.

On the other hand we're far from complete if you look at the spec.
If I just look at some projects I'm looking after:
* mem - isn't command-switch compatible with MSDOS mem
* kernel - is missing in config.sys:
(see http://fd-doc.sourceforge.net/spec/config.html)

- buffers=nn[,n] - we have no secondary buffers
- country - quite a useful feature
- driveparm (sp?) - sometimes needed for 720k floppy drives on XTs
- fcbs - this one is stupid in the FreeDOS context; we simply don't need
system FCBs for the FCB functions. It is parsed but ignored.
- install - quite useful and still missing

Furthermore with respect to compatibility the kernel isn't there yet.
The vast majority of DOS programs runs just fine, but there are some
well-known exceptions.

And I must also make all ASM assemble-able on Microsoft MASM :-(
And the "complete" list of kernel interrupts that must be
accepted contains interrupt functions that we should not implement
and misses some functions that we should implement IMHO.

Bart