🚀 go-pugleaf

In article <komarimf.703431575@craft.camp.clarkson.edu> komarimf@craft.camp.clarkson.edu (Mark Komarinski) writes:
>
>I think Russ's point is that people will stop buying DOS in favor of a
>real operating system, be it OS/2, UNIX, or NT.  I would not be so 
>optimistic about DOS's demise tho.  Maybe 2-3 years and most development
>for DOS-based platforms will stop.

Hardly any people buy operating systems.  The only one I've ever bought was
the DOS 5 upgrade; in every other case, the OS has come with the machine,
and usually isn't worth the trouble to change.  (The DOS 5 change was only 
worthwhile because it cost about $25, as part of a site license.)

This makes it seem very unlikely that people will stop buying DOS:  that would
require the clone makers to start distributing a different OS on their
machines, and nobody but IBM is brash (stupid?) enough to do that.

It's also true that people hardly ever buy new software.  I bought new
packages for the first couple of years of owning a PC, but since 1987,
have only bought one or two more.  The rest of my software
spending has been on upgrades.  Software vendors know they can keep making
lots of money by selling incremental upgrades to their packages every year;
they'll port to other OS's if there's enough demand (e.g. for something
like WordPerfect), but they'll never forget their big installed base
of DOS users who can be milked every year for another upgrade.

That still leaves the possibility that all the development will happen on
a different OS, and DOS will just get the ports.  I'd say that's possible,
but quite unlikely.  With so many more DOS machines and DOS programmers
around than for any competing OS, it doesn't make economic sense.
Furthermore, for anything but niche development, there's a wider variety of 
development tools available for DOS than for any other OS, and they're much 
cheaper.

Duncan Murdoch