Re: VisAD for general data visualization purposes for general users?

S F Palu wrote:

Please ignore this kind of  stupid question and criticism of  VisAD from
NEWBIE as M.Huang .

I don't think we should ignore criticsm.  We were all newbies
once and probably (at least I did) have the same criticism.

Obviously, this is an insult to Bill and the VisAD

I've said the same things before, but didn't intend to insult
and I don't think maohai intends it that way either.  m's pointing
out valid deficiencies in the documentation.  But all the
Javadocs in the world don't substitute for a good tutorial.
Ugo's done a fine job in his attempt.  He did this to fill
a void.  If that doesn't fill the void, then someone should
suggest changes or write a new one.  Sun's Java tutorial is
excellent at describing how to use the java classes, much
better than a set of javadocs.  That doesn't mean javadocs
aren't needed for developers.  But if I write a system on
top of Java, I don't expect that users of that system will
need to look at the javadocs for JTable to use my package.

If M.Huang cannot understand VisAD, then use JFreeChart or SGT. VisAD is
for serious stuff (or at least scientific stuff) . Scientific computing is
different from
 simple High School type plotting that is available in JFreeChart. If you do
High School plot
then VisAD is not for you. Try some commercial Java graphics APIs or
any open source ones. VisAD is the best is there is for Java. I rate VisAD
as the same
par as MatLab graphics or in some functionalities it has more built-in
graphics manipulation
functions than MatLab. MatLab is almost the standard Tool of every
Engineering School, Industrial Research (eg, NASA), Physics and Maths
around the world. So if VisAD is  equivalent to or almost superior than
MatLab graphics, then
 this means that M. Huang, need to join the mail-group for JFreeChart or a
newbie Java graphics
mail-group and demand the functionalities that she/he needed to be

I agree that VisAD is a powerful library that is not intended
for the weak of heart. ;-)  And simpler interfaces could be
constructed on top of VisAD, or perhaps one of the easier to
use tools like JFreeChart or SGT can be used as the rendering
engine for VisAD Data objects.  While the Display model in
VisAD requires you use the VisAD Data model, the Data model
could be used with other Display packages.  But someone would
have to convert between VisAD Data objects and JFreeChart/SGT/...
data models.

Perhaps we should take Andrew Wilson's suggestion and
get a list of the specific tasks needed to make VisAD a
suitable candidate for the Herschel IA group.  If no one
wants to take on these tasks, then they are free to choose
another software package or develop one themselves.  But they
need to weigh the trade-offs in personnel time between
re-inventing the wheel or documenting an already built one.
In the end, someone has to fund the development or do it
on their spare time and contribute it back.  I've got plenty
of people telling me what the IDV needs to allow them to
use it.  It goes into a list and waits for funding or
time to implement.

Anyway, for now I have to sign off this discussion.

Don Murray                               UCAR Unidata Program
dmurray@xxxxxxxxxxxxxxxx                        P.O. Box 3000
(303) 497-8628                              Boulder, CO 80307

  • 2002 messages navigation, sorted by:
    1. Thread
    2. Subject
    3. Author
    4. Date
    5. ↑ Table Of Contents
  • Search the visad archives: