[Tdwg-tag] Position of GBIF in architecture and centralization of services.

Bob Morris ram at cs.umb.edu
Mon Mar 6 19:09:36 CET 2006


Instead of your "rules out the architecture being based on" wording, I 
would prefer to see "rules out the architecture limited to":

This effectively rules out the architecture being limited to centralized 
data warehouse or indexing service beyond the hosting of standards files 
by TDWG itself - possibly as part of its collaborative infrastructure.

What I really believe is that warehousing and indexing should certainly 
be supported use cases, but so should be the kinds of direct queries 
otherwise discussed. In fact, I suspect that what one will find is that 
the warehousing and indexing requirements will come down to exactly 
those for direct queries PLUS appropriate support for caching support 
(e.g. data validity contracts), provenance and maybe access control. The 
idea of separating those three in
discussion is appealing to me because I suspect they are orthogonal to 
each other and to the other data storage and exchange issues. GBIF's 
experience will be valuable about "what got left out" in expanding the 
original distributed dreams (DiGIR, BioCase) into contemporary 
centralized realities. Central vs. distributed information processing 
and data provision is a perennial debate and in the end always seems to 
come down to current network and server hardware technologies. In 
various contexts, I have seen this question flip-flop many times since 
my first serious programming in 1965 on an IBM 1620. (Very cool CPU 
architecture: BCD arithmetic by table lookup. You loaded the arithmetic 
tables into low memory. The result was that you could make the machine 
do hardware arithmetic in any base up to 10. Excellent for number 
theory...). It will flip again, many times in the future.

Bob



Roger Hyam wrote:

>
> *Centralization of services*
> The TDWG architecture should not rely on any third party providing a 
> particular piece of infrastructure indefinitely. This effectively 
> rules out the architecture being based on any centralized data 
> warehouse or indexing service beyond the hosting of standards files by 
> TDWG itself - possibly as part of its collaborative infrastructure.
>
>
> Roger
>
>
> I
>
>------------------------------------------------------------------------
>
>_______________________________________________
>Tdwg-tag mailing list
>Tdwg-tag at lists.tdwg.org
>http://lists.tdwg.org/mailman/listinfo/tdwg-tag_lists.tdwg.org
>  
>

-- 
Robert A. Morris
Professor of Computer Science
UMASS-Boston
ram at cs.umb.edu
http://www.cs.umb.edu/efg
http://www.cs.umb.edu/~ram
phone (+1)617 287 6466





More information about the tdwg-tag mailing list