[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [MiNT] old KGMD or new NMD?



> The way I see this, we have 2 choices:
> 
> 1) Keep old KGMD
> 
>    a) people still complain that it won't install from HD, for some 
>       reason, and some config files have to be fixed by hand, which
>       is not well documented.  
> 
>       Result:  newbies give up and move to MagiC or Windooze.....

This is not the fault of the KGMD filesystem structure, but a fault of
KGMD installer. Frank is preparing a new installer AFAIK. Refreshed KGMD
would be (sorry) much more appreciated than a totally different
distribution.
 
>    b) we have to replace every part of the package with more recent ports,
>       since just about everything in KGMD is outdated or deprecated, so...
>       why bother with it in the first place?

This has been mostly done already.
 
>    c) everything we port results in HUGE diff files, because we have never
>       upgraded our paths and libs to support /sbin instead of /usr/etc and
>       have all PID files written in /var/run .... and many more things.

Almost any port for any distribution for any Unix results with diff files.
Don't think you'll get rid of them.
 
>    b) start with a distribution that is up-to-date and fully functional.

But non-standard. And you now exspect (I already told you that once) that
(for example) I will wipe my perfectly working, upgraded and configured
KGMD setup just to install a new distribution, which won't work better for
sure (it may at least work as good as my current setup). Sorry, but this
assumption (that I will reinstall everything from scratch without any
reson to) is not reasonable. I think that several people may think same
way. So, if we loose compatibility between MiNT setups, a mess will be
made without any rational reason (because, as I did explain my point,
changing FS structure just because someone calls it "obsolete", is not
rational).
 
> Obviously, this will require recompiling some of the existing software,
> but this easily done;  just a few paths to change in MiNTLIB.

Yeah. In worst case, this will require recompiling almost all the existing
software. And look at the FTP servers: PIWO has over 200 MB of MiNT stuff
in tar.gz files (guess about 0.5 gig after unpacking). Will you recompile
all that? And so, will FTP servers keep same stuff twice, just because
"pppd for KGMD" and "pppd for NMD" has differ with 1 hardcoded pathname?
This is pure waste of time and disk space.
 
> This might surprise a few people, but the biggest work I'm doing is on
> finding ways to easily upgrade from KGMD and documenting the procedure,
> so that everybody can follow.

I don't doubt that (less or more) everybody could follow even without
detailed docs sooner or later, but I don't see the point to do it. I'd
like to hear an independent opinion though, maybe I am totally missing
something...
 
--
Konrad M.Kokoszkiewicz
|mail: draco@mi.com.pl                  | Atari Falcon030/TT030/65XE |
|http://www.obta.uw.edu.pl/~draco/
|http://draco.atari.org

** Ea natura multitudinis est,
** aut servit humiliter, aut superbe dominatur (Liv. XXIV,25)
*************************************************************
** U pospolstwa normalne jest, ze albo sluzy ono unizenie,
** albo bezczelnie sie panoszy.