It seems that the inherent database design in the relationship is flawed
after having to put up with numerous emails about the problem from the list,
I decided to put my two cents (American) in. ;)
The flawed relationship is the one to many relationship between topics and
articles.
The assumption (incorrect) is that an article pertains to one and only one
topic.
The correct assumption is that topics contain many articles, and that
articles can pertain to many topics. As we are all aware the proper way to
do this is to have a join table.
topic -----< topicarticles >----- article
(id,etc.) (unique, (id, etc.,
topicid, w/o topicid)
articleid,
probably also
score as I assume
that score
indicates the
articles weight
vs. the topic)
Codd's normalization rules at least to the third form, are imperative for
developing
expandable relational databases. As you are aware, by making sure that data
is
only directly associated to the key, and that there is no need for redundant
fields
(extra1, extra2, extra3) etc., expanding the data model does not impact the
existing
structures.
I'm not saying this to criticize, simply to point the fact that if Midgard
is to
survive and thrive, a good model should probably be laid down before 2.0.
--
This is The Midgard Project's mailing list. For more information,
please visit the project's web site at http://www.midgard-project.org
To unsubscribe the list, send an empty email message to address
[EMAIL PROTECTED]
