[
Date Prev][
Date Next][
Thread Prev][
Thread Next][
Date Index][
Thread Index]
[
List Home]
|
Re: [udig-devel] roadmap png (no coffee)
|
Hi Adrian,
>> In the meanwhile I would like to add to the research part of udig the
>> console and scripting engine and the connections to the R statistical
>> package.
>>
>> Ciao
>> Andrea
>
> Hey Andrea,
>
> Do you have a design yet for your connection to R? The CGIAR folk are
> using a connection to RServe; I have been playing with a connection to a
> local R process via JRI. Do you have a plan yet?
No, no real design yet. I gave a look at the JGR stuff, I've used Rserve
in the past. Know what I want but I don't know how to get it... yet.
Soon we will have a developer that will be dedicated to the console
concept and the R integration, hopefully all in one. From that moment on
things will run fast...
> Also, are you going to be constructing Rsp objects? The JRI methods are
> still relatively simple for the construction of R objects (last I looked
> you could pass matrices in/out but not Data.Frame objects). Since more
> and more of the spatial methods in R are using Rsp, it seems it might be
> useful to create an RDataStore or something similar. I'll probably be
> exploring this over the next couple of months,
It would be good to interact in this, if we want almost the same toy.
Please keep me uptodate, I will introduce the new developer to this list
in about 3 weeks or so. At that time we will have design sessions, since
in that concept we want to have also chainable commands (operations?).
Glad if someone wants to partecipate with good ideas :)
Regarding to that I want to ask (so I can already do some prototypes) if
anyone has an idea on how to deal with this. Operations have their
extension point, but we need to implement the openmi interfaces.
Should we create new types of operations or create a new concept of
chainableoperation?
Andrea
