[GSDI Technical] Towards an open architecture for vector GIS

[michael gould]

Bruce,

Did not mean to get you or other co-authors mixed up in a debate, but of
course your response is justified.
My comment to Kate (cc:egip!) was not exactly on target in the sense that
your paper does in fact talk about the role of SVG as rendering language.

However my point was/is, that the notion of an "SVG-centric GIS
architecture" (citing your conclusion) is quite strange, in the sense that
rendering is not normally a key factor in system architectures....as are
interfaces for example.

In 1999 (5th working draft) we thought that the use of SVG was pretty
wonderful, but then we came to see that it was lacking the details needed
for representing GI; it only provided means for painting it. 
http://www.infoloom.com/gcaconfs/WEB/granada99/gou.HTM
Then GML came along.

Regarding the peer review process, yes, it seems to work when taken as a
whole, however I just completed a review of 10 papers along with 2 other
reviewers, and in the end our 3 opinions did not match all that well. So on
an individual basis, it seems you win some and you lose some. Or rather, we
don't all agree all the time.

Cheers,

-------
Michael Gould
Centro de Visualizaci—n Interactiva  www.cevi.uji.es
Dept. Information Systems (LSI), Universitat Jaume I, 12071 Castell—n, Spain
email: gould (at) lsi.uji.es  //  email2: mgould (at) opengeospatial.org
research group  www.geoinfo.uji.es  //  personal  www.mgould.com
AGILE www.agile-online.org
Vespucci Summer Institute www.vespucci.org
Erasmus Mundus: Master in Geospatial Technologies www.mastergeotech.info
 
 
 
 
 
 

-----Original Message-----
From: "Bruce M. Gittings" [mailto:bruce at ed.ac.uk] 
Sent: jueves, 12 de abril de 2007 18:03
To: technical at lists.gsdi.org
Cc: gould at lsi.uji.es; lancekt at aya.yale.edu
Subject: [GSDI Technical] Towards an open architecture for vector GIS

This thread has just been pointed out to me.  Three comments as an  
author of the paper being discussed:

1. I believe the peer-review process is usually sound

2. The work was done 4 years ago, the problem is getting things into press

3. Read the paper; I think we say rather clearly the SVG is for  
rendering, GML is for
data transfer and (slow) data storage.

Regards

Bruce Gittings
bruce at geo.ed.ac.uk


--

Kate Lance klance_remote at yahoo.com
Mon Mar 5 00:46:03 PST 2007

well then, doesn't this bring into question the peer review process?   
which reviewers
gave the green light on a paper with questionable approach?



----- Original Message ----
From: michael gould <gould at lsi.uji.es>
To: Kate Lance <lancekt at aya.yale.edu>; SDI technical <technical at  
lists.gsdi.org>
Sent: Monday, March 5, 2007 9:22:40 AM
Subject: Re: [GSDI Technical] Towards an open architecture for vector GIS


Hmm, yes. I suppose that Ron Lake and a few other OGC members might  
have interesting comments on this. The authors seem not to realize the  
importance of the fact that SVG is rendering-oriented. (Same might be  
said of KML today
)

We realized this (shortcoming) also too late, after having published  
on the wonders of SVG, 7 years ago.

_______________________________________________
Technical mailing list
Technical at lists.gsdi.org
http://lists.gsdi.org/mailman/listinfo/technical