[cap-talk] Capability past and future. CORBA (YURLs?), embedded OS

[Jed Donnelley]

At 12:09 PM 6/12/2007, Jonathan S. Shapiro wrote:
>On Tue, 2007-06-12 at 11:37 -0700, Jed Donnelley wrote:
> > Jonathan S. Shapiro wrote:
> > > But that was the right thing to do 15 years ago. Today, Linux has the
> > > benefit of overwhelming engineering investment performed over those 15
> > > years. It is no longer possible to deliver something "good enough" and
> > > iterate.
> > >
> > I believe the dominance of Unix started during the 1970s when AT&T
> > started giving it away to educational institutions.  We were already
> > using it at LLNL in 1975....
>
>That view is plausible, but it ignores the fragmentation of UNIX in the
>late 80's, and the fumbling and competing efforts to converge it
>(remember how OpenGroup was formed to combat X/Open?). In 1990,
>*businesses* still were not willing to adopt UNIX for back-office
>business operations, and this was the customer base that ultimately
>drove large UNIX revenue. UNIX in 1990 was perceived as too unreliable
>for business server use.

I agree with all the above, but despite the fragmentation and the
concerns, there really was no alternative.  All the major vendors
(except of course Microsoft) were supplying versions of Unix.
This was the environment that Linux was born in.  All the significant
work at universities (as I recall) was being done on free versions
Unix.  Those students go to work and find Linux or BSD available
and you think they would pick up an alternative?  Throw away all
the GNU support code, all the applications that run on the Unix
API, give up their investment in familiarity?  Not likely in my
opinion.

>As to whether reliability was a saleable proposition, HAL Computer
>Systems raised our $240M first round on the strength of two
>propositions:
>
>   1. A 64-bit SPARC-compatible processor attached to an S/370 class
>      system architecture.
>   2. Moving UNIX from "three nines" to "nine nines" reliability.
>
>In the eyes of the investor, who was the number 2 computing company in
>the world at the time, (2) was at least as important as (1).
>
>Absent the dramatic improvement in UNIX and overall system reliability
>between 1990 and 2000, the S/390 would be king of the mountain today.

I believe that dramatic improvement in UNIX reliability was
inevitable.  It is a little more than finding bugs and fixing them,
but not an awful lot more.  The basic design (while a crock in
my opinion) is sound, so I don't think there was anything standing
in the way of that dramatically improved reliability.

Still, at this point I feel like I'm not doing much more than
gum flapping (opinions - which of course can differ).

If there is really an opportunity for OCapCORBA, that sounds like
a worthwhile avenue to pursue to me.

--Jed  http://www.webstart.com/jed-signature.html