[elephant-devel] Representational Question

[Robert L. Read]

On Wed, 2008-03-05 at 21:50 -0800, Paul Legato wrote:
> I think the thing is not that we need professional consulting, but we
> are trying to become the professional consultants. :) Could you more
> knowledgable ODBMS coders recommend some good books or papers on the
> topic of ODBMS care and feeding that we should read to get a better
> handle on all this?
> 
> Cheers,
> Paul 

Thanks for the input, Paul.

I'm afraid I can't recommend a book on ODBMS, but for me a starting
point is this:

http://www.ibm.com/developerworks/library/wa-objprev/

I have personally always used Elephant in "prevalence" style where (Ian
doesn't do this, for example, because his database is too big.)

I personally think this mad would be clarified if someone (like me or
this community) would write a short guide to "doing SQL like things"
with LISP analogs.  This would have the advantage of showing how simple
it is to do the SQL operations in LISP.  

To me, the fact that Elepahnt allows you to use LISP as your query
language rather than creating yet another mismatch between the
programming language and the query language is a major advantage.

I will make a 30-second beginning for such a document:

*) We have a way to map over every instance of a class.  A simple filter
performs the equivalent of a "select" in SQL, but with much wider
computational power.

*)  The "aggregate" operations in SQL (SUM, MAX, MIN, AVG) can be easily
expressed in terms of the the "reduce" operation in LISP.

*)  Any sort of "graph" problem (topological sort, find-all-children,
find-all-descendants, etc.) is easier to program in LISP than in SQL.