MathGroup Archive 1999

[Date Index] [Thread Index] [Author Index]

Search the Archive

incompatibilities between releases of Mathematica (was: Mathematica Link for Excel and Excel 2000)

  • To: mathgroup at smc.vnet.net
  • Subject: [mg19293] incompatibilities between releases of Mathematica (was: Mathematica Link for Excel and Excel 2000)
  • From: "P.J. Hinton" <paulh at wolfram.com>
  • Date: Thu, 12 Aug 1999 22:34:47 -0400
  • Organization: "Wolfram Research, Inc."
  • References: <7or5mr$153@smc.vnet.net> <innclip7ouvmd$osf$5@dragonfly.wolfram.com>
  • Sender: owner-wri-mathgroup at wolfram.com

On 12 Aug 1999, Joseph Yoon wrote:

> While we're on the subject, I've seen no explanations from WRI as to
> why the changes that were made from v3 to v4 cause so many
> applications and notebooks to "break" and the reasoning behind them.

Perhaps you could help us out by describing in precise terms exactly what
you mean by "broken."  The likelihood that you will get your code working
to your satisfcation will increase if you can post specific, minimal
examples that reproduce the problems you've been running into.

> Getting info about bugs and "incompatibilities" from WRI has never
> been easy.

Diagnosing compatibility problems is not as simple as looking at a general
list of known issues and saying, "yup, that's the problem."  Your
effectiveness at resolving these types of difficulties hinges greatly upon
your approach.  If you send a notebook with 1000 lines of uncommented code
to Technical Support or MathGroup and say something like "This doesn't
work anymore. Why?" your're less likely to achieve a resolution to your
problem in a reasonable amount of time.  It is not out of negligence or
malice on anyone's behalf that this happens, it is because the scope of
the problem has not been sufficiently narrowed.

There may be experts on the Mathematica language both at WRI and on
MathGroup, but they may not necessarily be well versed in your choice of
computational techniques or areas of research.  If the example you send
does not sufficiently isolate out the behavior, then the task of problem
solving quickly becomes an exercise in soothsaying, which may lead nowhere
rather quickly.

Also, there are several instances where a user will discover a feature of
the language that may be desirable from his or her standpoint, but it is
undocumented.  In subsequent releases, the user finds that code which
relies on the feature may not function in the same manner.  These types of
incompatibilities cannot be predicted a priori, and the introduction to
the Reference Guide in _The Mathematica Book_ cautions the user be
vigilant in this respect.

--
P.J. Hinton
Mathematica Programming Group           paulh at wolfram.com
Wolfram Research, Inc.
Disclaimer: Opinions expressed herein are those of the author alone.



  • Prev by Date: Re: Short Doesn't Work
  • Next by Date: Re: Re: Subscripts, Doh!!!
  • Previous by thread: What I try to ask is ...
  • Next by thread: Re: Re: incompatibilities between releases of Mathematica (was: Mathematica Link for Excel and Excel 2000)