From: dreier@jaffna.berkeley.edu (Roland Dreier) Subject: Re: ANNOUNCEMENT: Alpha release Linux/GNU/X unix clone on CDROM for PCs Date: 29 Nov 1992 18:20:27
In article <id.TJEV.31K@ferranti.com> peter@ferranti.com (Peter da Silva) writes:
In article <1992Nov29.180008.3128@news.nd.edu> dmatiske@virgil.helios.nd.edu (david matiskella) writes:
> I keep hearing how great Herd is going to be but when was the idea first
> proposed?
I was running Unix on a PC/XT in 1984, and it was faster than MS-DOS.
Minix has been out for at least 6 years.
CP/M was better in 1978 than MS-DOS in 1982.
Cromemco had a UNIX-lookalike running on multiple Z-80s in 1980.
AmigaOS first shipped in 1985, and it was already better than any
PC operating system that was to ship until OS/2 2.0 seven years later.
I have been following the threads lately about GPLed vs. Public Domain
vs. Commercial software lately, and the above exchange struck me as
rather interesting. All of those OSes you cite in response to the
Hurd query are commercial. In March 1985, the GNU Manifesto was
published in Dr. Dobbs. In the 1986 Byte Interview, Stallman was
promising the kernel in a year or two. It is now almost 1993, and the
GNU operating system, now Hurd, is shaping up into vaporware of
proportions that would make Microsoft jealous.
It seems rather interesting that we have commercial operating systems
with versions like Nextstep 3.0 or Solaris 2.0, and even OSF/1 can be
had, all before Hurd gets out. Linux is almost up to 1.0, but Linux
does not have interesting new technology. Even Windows/NT, which will
probably be shipping before Hurd, offers new stuff like
object-oriented device drivers and so on. I'm not sure exactly what
the delay has been with Hurd, either; it seems like all that is been
done is taking the Mach 3.0 stuff written at CMU and crossing it with
the BSD stuff written at UCB, and tacking on MIT's X windows.
If we were truly living in the "post-scarcity" world the FSF dreams
about, there would be no commercial operating systems. Would we all
be running Hurd alphas? I reread the GNU Manifesto recently.
Stallman's vision for the future seems like a immensely bad idea.
Complain all you want about Gates and Microsoft, but the FSF wants to
eliminate all commercial software and fund development by a "software
tax" on computer sales. It seems to me that this would change an
industry that is one of the few that the US seems to do well into a
god-awful cross between the IRS and the current system of funding
research. The LPF can terrorize you with nightmares about software
patents; but how would you like to have to write grant proposals to
write software?
OK, the asbestos undies are on :)