Re: DoD and NIST undermining commercial CASE industry

[ogicse!flop.ENGR.ORST.EDU!gaia.ucs.orst.edu!umn.edu!csus.edu!netcom.com!m cgregor@uunet.uu.net (Scott Mcgregor)]

In article <1992Dec3.093920.19673@sei.cmu.edu> firth@sei.cmu.edu (Robert Firth)
 writes:

>However, of the competitors, PCTE has the most diverse support in
>the global market; if you accept Emeraude as a realistic "proof of
>concept", then I can't see any potential competitor displacing it
>on technical grounds.

As Robert points out (with his QUERTY example)  it rarely is technical
grounds that causes displacement.  It is more likely because another
solution becomes a de facto solution.  That's likely to be because of
an important vendor making it standard (e.g. Sun's NFS), the move to a
new major platform with extant solutions (e.g. the moves from
VMS to Unix changed the importance of DECnet vs. TCP/IP.  And lastly,
sometimes these things are end runs. The engines don't matter until
you can buy an off the lot car that has one standard--then what
matters is that it is the standard engine not that there are better
alternative engines for which there are no cars. This is the grave
risk to PCTE *AND* ATIS until there are more end user environments in
place that rely on them.

>Again I agree - but this seems to me the critical issue.  The key problem
>is not the lack of end user solutions, but the lack of the "enabling
>technologies" that will open the door to end user solutions - solutions
>built by numerous third parties, competitively, but all with a high
>degree of compatibility and interoperability.
>
>As a poor analogy, consider PostScript (TM).  This is an enabling technology
>that alows you to buy any of a dozen machines, drawing programs, printers
>and display units, plug them together, and do useful work.  That's the
>kind of rationale behind PCTE, and I think it's the critical step in
>allowing us similarly to "mix and match" our software development tools.

Consider that there WERE and ARE today other alternatives to
PostScript.  It is an enablying technology. But it is not that it
*theoretically* can allow you to connect any of a dozen machines,
printers, programs, etc. The alternatives share this theoretical
possibility. What PostScript has going for it is that it works this
way *in practice*--i.e. that there are end user solutions to allow
useful work.  The real  risk with PCTE and ATIS is that they don't
have such an installed base of end users benefiting yet. This leaves
both vulnerable to the next solution, even one less technically
correct, but which in practices solves real end user problems for many
people.

>Well, is PCTE at the point the VHS de-facto standard reached, where any
>attempt to compete will go the way of Betamax?  In my view, yes, but as
>always feel free to disagree.

I don't think so. But primarily because PCTE and ATIS are at the level
the scarcity that VHS and Betamax were BEFORE home VCRs were common.
Note too that VHS is not the clear winner for all time--with the
growth of the handicam market the smaller formats have beat out the
larger formats. One would think that VHS-C would have an advantage due
to compatability, but Sony's 8mm is doing better (so far). When HDTV
comes around all these things could be lost in the dust.  Is the
future in computing any less uncertain? What will be the implication
of the new OO O/S's like NeXTstep, Taligent Pink, et. al.? What role
with Windows/NT successors play?  What are the implications of
massively parallel processors, palm tops, and another factor of four
increase in price performance in the next 3 years, a factor of sixteen
in six (application of Moore's Law).  It is easy to displace
technologies with installed bases in the hundreds and thousands. Much
harder to displace technologies that millions of end users rely on.

I don't mean to imply that PCTE and ATIS can't win, just that the odds
are similar to weather prediction.  I still remember when it was
ASSUMED that OS/2 would be dominant over MS-DOS by now. But those
predictions were made when OS/2 was had as few end users as PCTE and
ATIS. I think this is why CASE vendors and hardware vendors are
playing it cautiosly: verbally committing to the standards "when they
mature" (which means when they become pervasive in the market and they
can't sell their product without them), but not actively making their
products crucially dependent on either quite yet.

Markets are fascinating things, aren't they?



-- 

Scott L. McGregor		mcgregor@netcom.com
President			tel: 408-985-1824
Prescient Software, Inc.	fax: 408-985-1936
3494 Yuba Avenue
San Jose, CA 95117-2967

Prescient Software sells Merge Ahead, the tool for Merging Text or Code and
offers consulting  & training in project management and design for usability.