[libdb-develop] Python module - intent

[Terry H. <ha...@an...>]

Hi Morbus (and others),
We spoke earlier this year, I believe, when you just had SQL 
for LibDB (I think we met on Web4lib), but I forget... I 
suggested switching to Python rather than Perl, as you may 
recall, and I think we agreed that I'd probably be able to 
use your SQL schema work as a basis at the very least, but 
that I'd be on my own with the Python. ;-)

Anyway, you recommended reading the FRBR paper, and I 
finally did.  I've also downloaded version 0.0.3 of LibDB 
and have been looking at the code.

I have made an attempt to read the Perl code.  Well, all I 
can say is, I *thought* I had learned Perl, but I seemed to 
have forgotten what I learned.  Your comments and program 
file style are a great relief from the usual hacker fare, 
though -- thank you, I really respect that!

The code doesn't really seem to do that much though, as yet 
(or I'm not looking at the right bits), and I'm wondering 
what it's for in a general way.  I wonder if I need to 
understand it at all.  Do you discuss the object-model 
anywhere?

I also am intrigued by the Javascript code you have for web 
templates -- I wonder if I can make use of that, too, 
provided there is enough similarity of purpose.

My goal is to create a Python cataloging module to be used 
with Zope, which will package an 
object/collection-of-objects  and provide 
FRBR-model-compliant cataloging capabilities.

As a Sourceforge user, you will be familiar with the "Trove" 
categorization system.  This is something like what I'm 
trying to use FRBR for (but I want to do a much better job, 
which I think FRBR compliance can provide).

So there will be self-cataloging, done by the project owner, 
and then catalog searching by end-users trying to find 
software.  I also envision a class of "librarian-users" who 
will have the authority to re-catalog or "flesh-out" the 
catalog information, providing better meta-data and 
"shelving" the results to make them more accessible to the 
end-users.

And of course, I visualize a greater variety of items -- not 
just software packages, but e-books, audio records, 
slide-shows and hardware CAD drawings, etc.  Some of the 
layers in FRBR (such as "Manifestations") will not be 
separate items in the FTP storage server, but rather 
separate ways of processing and delivering the same 
content.

Will I benefit from examining your object model do you 
think, or would I be better off designing code from 
scratch, but sharing the SQL schema?

Cheers,
Terry

--
Terry Hancock ( hancock at anansispaceworks.com )
Anansi Spaceworks  http://www.anansispaceworks.com