[vtkusers] I think this is the problem......
Joey Mukherjee
joey at swri.edu
Fri Mar 19 11:43:11 EST 2004
On Mar 18, 2004, at 7:55 PM, Bill Oliver wrote:
> On Thu, 18 Mar 2004, Joey Mukherjee wrote:
>
>> On Mar 18, 2004, at 8:16 AM, James C. Robinson wrote:
>>
>>> designing the look and feel of the interface):- graphics is something
>>> that I
>>> want to be able to take for granted - this is why I have opted for a
>>> 3rd
>>> party solution.
>>
>> We all want something for nothing!
>>
>
> I have a bit of a problem with interpreting this as wanting
> something for nothing.
What I inferred from James message was that he opted for a 3rd party
solution and was upset that it didn't have enough documentation for his
particular and very specific needs and wanted the user community at
large to help him out because he needed it. I don't remember him
offering to do anything. He wants something for nothing. He also
claims that FE analysis is very common and incorporates a large part of
the VTK community. Is that true? I don't do any of that.
> The request I've been seeing is not to get something for nothing,
> it's to find solutions to work from rather than starting from
> square one all the time.
Have you looked at the examples? These are all working solutions.
Every last one of them. They all take an input and produce an output.
Learning how these works is normally left as an exercise to the user.
Sometimes they have everything you need, but more often than not, they
don't and you build solutions exactly as you did in grad school. I
don't think Kitware has ever refused more examples if people have them
to offer. Plus, Kitware has training classes (or used to) which are a
way people can support Kitware and get solutions.
> Take a look at how the few remaining commercial viz programs
> have survived -- like IDL, AVS, and Matlab. They survive in
> large part because they have managed to create user communities
> that share applications and build on each other.
I agree, but you mentioned the keyword: commercial. We use IDL here
and it ain't cheap. Open source never has worked like that, and if
James is willing to get a support contract to fund the documentation
that he needs, more power to him. Actually, wasn't that the way the
annotation of plots got added to VTK? Someone funded Kitware and
donated the changes back. I want to say the MPI stuff is coming along
the same way...
> You don't have to be on your own. And you don't have to write them all
> down. It would be useful, however, to have a repository that shows how
> different people solved common problems in different ways. Providing
> that
> would a) cut down on the repetitive questions on the mailing list, b)
> provide a learning tool that would increase the user community base,
> and c) provide a resource that even experienced VTK programmers could
> make use of. Once again, look at the AVS5 experience -- AVS5 comes
> with
> what, 100 modules? And how many user-community modules are at
> iavsc.org?
> 600? 1000?
I am all for it, and the wiki concept is a good one for this. I'd be
happy to put some of the stuff I've done into something like this.
However, the search on the Kitware site doesn't work and I think that
would make more of difference since people could actually search the
mailing list for all same questions and hits might appear. I'd opt for
spending resources on that than adding a wiki that few may use.
I understand James' frustration since we've all been there, but he did
ask for comments and I gave him mine. I really did not know the answer
to his other questions.
Joey
More information about the vtkusers
mailing list