> e.g., a la XVT which is a package used to port software between Macs,
> PCs, and UNIX boxes. I have heard some contradictory statements over
I have been following the recent conversations with much interest, because
I am in the early development stages of a new project which needs to support
a wide variety of user platforms. XVT has been used at Ford for internal
packages, but I am not entirely convinced that it is bullet-proof enough for
commercial software (see BYTE's cross-platform issue a few months ago) or
that it will be worth the time spent getting up to speed on the rather cryptic
XVT toolkits. A whole other issue is that I am not entirely convinced that
the user interface is going to be what the users want: things seem to get
very GUI with XVT, and I have had a great deal of positive feedback for my
current character based full-screen PC product. They want something like
FMEAplus, but if we go with a cross-platform toolkit, we have to use the
interfaces designed into that toolkit, which by definition are going to be
a least common denominator of all the platforms supported.
It isn't easy to get "Customer Delight" with a software tool, but that's what
we all want, something that makes the job EASIER for the engineer or
scientist. I am afraid no software initiative alone is going to do that, in
my experience you have to take it to the users, get them fully involved in
the development process. I don't know whether the government or private
software development firms are doing that in this case. If they aren't, we
aren't likely to see the fantastic software possibilities here realized at all.
--
Chad Childers Home: chad at deep.mi.org
FMEAplus Project Leader Work: chad at quality.ta.ford.com
