Wirt, in his typically beautiful prose, wrote:
 
> I consider Posix to be an aberration of design that can have no real future.
> It is a bastardization that must ultimately disappear from the HP3000 or it
> will kill it.
 
Gees, Wirt.  Tell us how you really feel about this ;->
 
> If you really wanted a Unix system, then you should have bought
> Unix. Unix cannot be slowly blended into MPE in a series of half measures
> taken over a period of years.
 
I must disagree.  UNIX, which in and of itself is composed of a loosely
connected series of half measures, survives and grows.  It may be undesirable
to inflict MPE with POSIX, and hazardous to do so over time, but it can be
done.  Given limited lab resources, it may be the only way to do so.
 
> It is a design path that will, in the end,
> satisfy no one -- and it will very likely ultimately kill the HP3000. The
> continued enhancement of Posix-like features at the expense of the
> simplicity, elegance of design, and traditional ease of use of the HP3000
> will defeat the very nature and values that attracted so many of the people
> on this list to the machine 10 to 15 years ago.
 
I agree that I would prefer efforts at enhancements be made to improve MPE,
and not make a UNIX clone of it.  But the POSIX interface is resurrecting
MPE as a viable alternative to UNIX (as viewed within my limited set of
circumstances).
 
> Posix is not Unix. It can never be Unix. It can only be a bundle of
> significant compromises, a poor substitute, as all emulators are, for the
> original. And the closer the HP3000 becomes to being Unix-like, the more
> danger it is in.
 
POSIX is indeed not UNIX, and for this we can be thankful . . . in most
instances.  The promise of a common interface for proprietary systems
excites me, even though POSIX is not what I would prefer.
 
A crazed maniac, during a brief moment of lucidity, wrote in a recent
issue of Interact about how common languages have facilitated growth in
humanity and in the computer industry.  POSIX, detestible as it may be,
is the lowest common denominator, and the common language needed to
facilitate a new spurt of design growth for proprietary systems.  Thus,
I think by providing the common language/interface, and by taking away
the "need for commonality" argument used by the hell-bent-for-unix and
management-by-magazine crowd, it may be more the salvation of MPE, and
not it's demise.
 
And I also believe that the technical or aesthetic condition of POSIX
may be irrelevant.  If simplicity and utility were the only criteria,
then we would all speak Esperanto.
 
[colorful discussion on early television deleted]
 
> The same thing will inevitably occur in data processing -- or at least in the
> form of data processing conducted among businesses -- if any of this activity
> is ever going to pay for itself. The organization that makes this happen
> first will be the one that will prosper (and in the case of color
> televisions, it wasn't RCA).
 
No doubt.  This is what I find interesting about NT.  Uniform application
across multiple platforms, with a Judy approved management interface.
 
> Large data processing staffs will almost certainly go the way of the TV
> repairman. To a degree, you can already see the movement in that direction
> with the disappearance of the HP SE. Data processing managers,  SEs, or TV
> repairmen all represent an enormous cost that gets invisibly bundled into the
> price of any machine. But the simpler, the more automatic, and the more
> reliable a machine can be made, the more inexpensively it can be sold -- and
> the agency that can create such a device first will be the one that will
> ultimately prosper.
 
The complexity issues are interesting, and like the evolution of the
industry in general, are manifest in where the human resources are
being applied.  IS cost, and staff payrolls, are not descreasing.
In fact, IS appears to be the one organizational entity that survives
(more or less) the axe of the business re-engineering experts.
 
Application staffs can be cut due to 5th generation tools that take most
of the drudgery out of application design (the staffs won't get cut, but
may finally catch up on that multi-year backlog they have maintained).
Operations staffs may get trimmed, but not where there are systems from
multiple vendors on site (same applies for system adminstration teams).
Network engineering staffs will grow as will the new caste of DBAs.
 
In short, I think the situation will get worse as diversity grows.  Job
security through complexity and confussion :-o
 
> Unix is no model for such a path -- and no one should be fooled into
> believing that it is.
 
See above.
 
To summarize, I have to disagree about the effect POSIX will have on MPE,
while agreeing that it would be desirable for HP to apply more effort and
resources to the design of the underlying operating system, and not the
latest shell on top.  But that shell must be completed to assure the
short term survival of MPE before it's long term fucntionality can be
significantly expanded.
 
 =======================================================================
Guy Smith                                Voice:  804-527-4000 ext 6664
Circuit City Stores, Inc.                  FAX:  804-527-4008
9950 Mayland Drive                      E-Mail:  [log in to unmask]
Richmond, VA 23233-1464         Private E-Mail:  [log in to unmask]
 
The thoughts expressed herein are mine and do not reflect those of my
employer, or anyone with common sense.