[1246] in java-interest
Re: What's needed
daemon@ATHENA.MIT.EDU (Robert Stephen Rodgers)
Sun Aug 27 20:11:47 1995
Date: Sun, 27 Aug 1995 16:26:40 -0400 (EDT)
From: Robert Stephen Rodgers <rsrodger@wam.umd.edu>
To: Mide Services <mideservices@almide.demon.co.uk>
cc: java-interest@java.sun.com, java-interest-digest@java.sun.com
In-Reply-To: <517@almide.demon.co.uk>
> Oberon is not a commercial operating system. Wirth et al are no
> fools.
Who said they were?
I said, another Oberon doesn't do us any good. Writing operating systems
is an outdated way of "proving" a language. Lisp, Smalltalk, Perl,
Prolog, Modula 2, Pascal, Modula 3, Sather, Python, Objective C. What's
to keep Java from becoming yet another notch on the wall of languages
with interesting properties but which were passed over for C and kin?
Answer: Popular use.
And you know what? You'll *never* produce an operating system with any
kind of popular use. It's not even worth the bother.
If Java enthusiasts want a hard task, one with importance stretching into
the future and not into some "curious languages" CDROM, work on something
relevant to the language. Operating systems haven't been relevant to the
success of languages since 1972, or Pascal would dominate both the Mac
and Windows worlds and C++ would be a footnote.
> Oberon is an experiment, an intellectual virus. However, it does
> not FOLLOW that it could never take off.
There are so many reasons why Oberon (proper) wont take off that it's
pointless to even debate them. Not the least of them is that Oberon
isn't much more usable than experimental GUIs from 1983. It's
interesting to note that Oberon isn't really a full OS, though it's
enough of one to keep apps written in/for it from benefitting from their
environment.
> orbit. I would not be surprised if the next revolutionary ideas in
> Software Engineering will be coming out of Zambia or Brasil, or something.
They'll come from wherever expertise meets genuis. That's not restricted
to non-western non-northern-hemisphere countries.
> No, I agree, camp Oberon is full or nerds. But I have to laugh
> though, when you say:
>
> "Do you fellows really think that you can write an operating system on
> the level of Microsoft _Windows 3.1_(+DOS)"
>
> That's got to be a YES!
Let's see you do it. Window clipping, device drivers, and yes, even
performance. OLE. Let's see it.
Let's see it.
There are people with a lot of experience who put *years* into these
kinds of products -- some impressive (PC Geos), some not. But as
primitive as Windows is, it embodies much more attention to detail and
more effort than I have *ever* seen a loose conglomeration of enthusiasts
produce. Linux, Hurd -- Linux is an incredible example, but do you think
a JavaOS project would attract the kind of man hours put behind it (and
they're still using X, what a pity)?
Forget the OS. I despise Windows, but at least I'm not irrationally
estimating that a Java enthusiast group could match it in short order.
It's crap, but even that level of crap takes a lot more work that would
*best be applied elsewhere*.
[...]
> We can succeed *effortlessly* in this, by thinking on a grander scale.
> The Founding Fathers sowed the seeds for you to be able to think like
> this. As the I Ching is always reminding us, draw a bigger circle.
Whatever. I spent a summer translating the I Ching, and the one thing I
remember from the experience is that avoiding futile expenditures of
effort is wiser than pursuing pointless folly.
JavaOS folks, go right ahead. The rest of us will try and turn Java into
something with real-world ramifications, and hopefully, make money doing
so. The more people fall into the silly OS trap, the fewer the rest of
us will have to compete with.
RSR
-
Note to Sun employees: this is an EXTERNAL mailing list!
Info: send 'help' to java-interest-request@java.sun.com
