| Re: RSI response to CONTOUR problems/questions. [message #1399] |
Tue, 05 October 1993 16:48 |
rfinch
Messages: 51
Registered: March 1991
|
Member |
|
|
>>>> > On 5 Oct 1993 18:38:25 GMT, turet@carson.u.washington.edu (Philip Turet) said:
Philip> NNTP-Posting-Host: carson.u.washington.edu
Philip> Try to go to Sun or the
Philip> ANSI committee if you want a new whistle put in Fortran.
Non-sequiter. Fortran is in the public domain (or equivalent), and
produced by tens of developers, IDL is not, and produced by only one
company.
Philip> Some other developers might get into more commercial
Philip> pandering
I wouldn't call helping users with problems, instead of letting them
flounder, 'commercial pandering', but I guess support, like beauty, is
in the eye of the beholder.
Philip> know about you but I'd rather not have to deal with another Mathworks.
Here is an example of what I was referring to in the
comp.soft-sys.matlab group. Seems to me Mathworks' candor is
refreshing, what did you find wrong with them?
----------
From damian@mathworks.com Tue Oct 5 15:40:23 1993
Newsgroups: comp.soft-sys.matlab
From: Damian T. Packer <damian@mathworks.com>
Subject: Re: RBBOX question.
X-Xxdate: Tue, 5 Oct 93 15:16:51 GMT
Organization: The MathWorks
X-Useragent: Nuntius v1.1.1d16
Date: Tue, 5 Oct 1993 19:16:51 GMT
Craig Jones,
craigj@irus.rri.uwo.ca writes:
> Can anyone give me some sort of idea of how to use the RBBOX command?
> The help on RBBOX is not helping me very much.
>
[stuff deleted]
Since this might be of more general interest, I am posting to the group
as well as e-mailing directly to Craig.
The HELP entry for RBBOX is optimistic, but wrong.
[Rest of response deleted].
----------
It was certainly a surprise to me to see such honesty by a vendor.
--
Ralph Finch 916-653-8268 voice
rfinch@water.ca.gov 916-653-6077 fax
Any opinions expressed are my own; they do not represent the DWR
|
|
|
|
| Re: RSI response to CONTOUR problems/questions. [message #1452 is a reply to message #1399] |
Tue, 05 October 1993 11:38  |
turet
Messages: 7
Registered: January 1993
|
Junior Member |
|
|
RE: RSI's responses to the newsgroup. Sorry, but I agree with RSI, they
should stay off the airwaves routinely. Most of the issues that are not
directly bugs are best (most creatively) worked out by users, in fact
there's a tug-of-war between the applications that people want built in
and what is best done with the language itself. Try to go to Sun or the
ANSI committee if you want a new whistle put in Fortran.
Some other developers might get into more commercial pandering, I don't
know about you but I'd rather not have to deal with another Mathworks.
--Phil Turet
|
|
|
|
| Re: RSI response to CONTOUR problems/questions. [message #1473 is a reply to message #1452] |
Wed, 29 September 1993 09:37 |
rfinch
Messages: 51
Registered: March 1991
|
Member |
|
|
>>>> > On Mon, 27 Sep 1993 16:31:20 GMT, info@rsinc.com (Research Systems Inc.) said:
RSI> Nntp-Posting-Host: 192.160.108.6
RSI> Although we do not
RSI> normally respond to postings on the news group
Please reconsider this. I have been very impressed the last few days
with the comp.soft-sys.matlab group because MathWorks, a maker of a
commercial Matlab, has several of their people responding to questions
posted to that group.
RSI> In response to Patrick Ryan's (NASA/GSFC) requests:
> 1) to "Get rid of /MAX_VALUE. put in code to do real handling of
> missing data."
RSI> Ralph Finch (Calif. Dept of Water), and Bill Thompson (NASA/GSFC) also
RSI> expressed this opinion. Part of the IDL philosophy is that since IDL
RSI> is a full feature language, it is better to solve such problems using
RSI> IDL, rather than jam in special purpose features.
Excuse me. Missing values are not a special feature in real data. We
have 10 million + data points, and probably 10%-20% are missing. They
are so common that they should be handled cleanly and easily. That
includes not only plotting them properly, but all routines where
possible should have the MISS_VALUE= keyword and deal with it:
statistical routines, fourier transforms, etc.
The 'trivial' way to deal with missing values, as given in the
example, certainly works but is annoying to have to do once; a real
time-waster to have to do 20 times in 13,000 lines of code (our
situation).
Anyway, how is MISS_VALUE any more a special feature than MAX_VALUE?
--
Ralph Finch 916-653-8268 voice
rfinch@water.ca.gov 916-653-6077 fax
Any opinions expressed are my own; they do not represent the DWR
|
|
|
|
comp.lang.idl-pvwave archive
Members
Search
Help
Login
Home
