Comments on scripts, modeling and Grass

[Bernhard Reiter]

On Thu, Oct 28, 1999 at 06:15:34AM -0700, Rich Shepard wrote:
> On Thu, 28 Oct 1999, Agustin Lobo wrote:

> > The question here is to decide whether we should select a language to
> > become the standard for "gluing" grass commands into more complex
> > processes. 

>   My opinion on establishing a "standard" meta-language for GRASS is to not
> do it.

I second that.
There would be advantages in doing so, but staying independent is even
better in the long run.

>   However, there is value in providing a set of common hooks (which I
> believe already exists within GRASS) to pipe input and output not only
> among GRASS modules but also with external programs. As long as this API is
> well documented we can use whatever "glue" language we prefer.

Correct.
If the hooks were some distributed object protocol, that would make
life easier for "glueing". Standards for that are CORBA and XML-RPC.

Did you look at www.swig.org, which actually is technology to provide
such hooks to higher level programming languages.

>   Personally, I like C, and I always marvel at the proliferation of
> scripting languages. 
Well there is a reson for that.
But let me add that it is not a scripting/interpreted versus compiled
languages issue. It is about modern programming paradigms, like
object-orientated and dynamic. And you can program better in them for
most, but not all tasks.

> New ones show up quite frequently on freshmeat 
perl, python and scheme are around a long long time. :)

	Bernhard
-- 
Research Assistant, Geog Dept UM-Milwaukee, USA.    (www.uwm.edu/~bernhard)
Free Software Projects and Consulting 			    (intevation.de)  
Association for a Free Informational Infrastructure              (ffii.org)
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 288 bytes
Desc: not available
Url : http://lists.osgeo.org/pipermail/grass-user/attachments/19991028/ba6698fe/attachment.bin