-----BEGIN PGP SIGNED MESSAGE-----
Hash: SHA1
Hi Pepijn
Some thoughts to discuss about sesame.
Creating/modifying
- -----------------------
- - When getting the keywords from sesame do we still need the search
for keywords or is it possible to get only related keywords and show
them all? I hope it's possible to generate a specific enough query that
a search becomes obsolete.
- - How do we store the chosen keywords in a workunit/subproject/... We
don't have a unique key (at least I haven't found one). Does this mean
that we need to store the rdf tuple we have selected?
Indexing
- ----------
- - We probably will need to index the whole tuple (e.g. keyword) to
make able in the search if its narrower or brother.
Search
- ---------
- - Give the user the possibility to select specific tuples, indicate if
explicit/broader/narrower search shall be done and then execute the
search on our lucene index with the given arguments.
I suggest to start with the creating/modifying task. This will have the
benefit that the stored keywords are consistent with a given ontology
and can be searched via legacy search if the 'ontology-based' search is
not (fully) implemented.
WDOT?
Felix
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2.0.9 (GNU/Linux)
Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org
iEYEARECAAYFAkgHAC4ACgkQ2lZVCB08qHHSyQCdEHmRZwipV8Z+aqPEUC4czFvv
EvcAoN6lchk5YROFPLdlRgnxL1vQSDGc
=B90T
-----END PGP SIGNATURE-----
