Unidata - To provide the data services, tools, and cyberinfrastructure leadership that advance Earth system science, enhance educational opportunities, and broaden participation. Unidata
         
  advanced  
 

Re: java <9606181649.AA29958@xxxxxxxxxxxxxxxxxxxx>

> An interface to Java per se would be no more interesting than, say, the
> C++ interface.

Except that there would be no code porting problems; the Java interface and
programs that use it would run without change on Unix systems, Win95,
Windows NT, MacOS, OS/2, VMS, MVS, JavaOS, and any other platforms that
support the Java virtual machine.  The C++ interface and programs that use
it, in contrast, still require porting and testing on lots of different
platforms.

 ...
> The possibly new capability would be to provide data browsing on the
> web. This could be useful if it lowers the (currently high) barriers to
> visualizing other peoples's datasets. I guess that the main technical
> problem (besides the hard work of writing data browsers) is the data
> transfer.

Java might be useful even for local data browsing, because of its "write
once, run anywhere" characteristics.  An investment in developing and
maintaining programs for data visualization may become worthwhile if the
cost of supporting multiple platforms is eliminated.

> At NCAR, we have gazigabytes of data.  We often use rather outdated batch
> mode "visualizers" in order to deal with the large amounts of data. A
> better approach is with a data server that can extract desired subsets of
> the data to visualizer clients. If this is fast enough, then you can
> extract just what you need to draw the next picture, and still get
> interactive browsing.
> 
> Java might be a reasonable language to implement the clientside
> browser. Although the AWT toolkit is immature, without a doubt third
> party GUI builders will become available. It is important to think about
> how the data server side should fit in. I know there is a proposal to
> extend NFS protocols over the Web. It would be interesting to see how
> fast this might be.
> 
> A data server that also provides data compression, data operators,
> derived fields, etc could be a Very Good Thing if it became a standard.
> Then, you'd run your netcdf-server daemon along with your http daemon.
> 
> I know there are lots of ways to skin this cat, and many are thinking
> about this question.  Obviously, the key is standardization.  I'd be
> interested in hearing what others think about web data access.

I think Java may be a reasonable language in which to implement a data
server, not just a client-side browser.  From what I've seen (e.g. the
Jeeves HTTP server completely written in Java), there will soon be more and
better APIs available for servers as well as clients.

--Russ

 
 
  Contact Us     Site Map     Search     Terms and Conditions     Privacy Policy     Participation Policy
 
National Science Foundation (NSF) UCAR Community Programs   Unidata is a member of the UCAR Community Programs, is managed by the University Corporation for Atmospheric Research, and is sponsored by the National Science Foundation.
P.O. Box 3000     Boulder, CO 80307-3000 USA     Tel: 303-497-8643     Fax: 303-497-8690