Creating My DVD Movie Database

[Christopher Browne]

On Sat, Jul 3, 2010 at 10:36 PM, Stephen <stephen-d-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org> wrote:
> I have about 300 DVD movies.
>
> I would like to create a database to support queries.
>
> Movie Title
> Director
> Actors
> Themes
>
> I will use MySQL.
>
> Probably PHP and a web interface.

I don't particularly love any of those, but "chacun a son gout"
(everyone to his own taste).

> The work I am looking to simplify is the data entry. Specifically the
> One->Many relationship between movie title and actors.
>
> All tables will be normalized, and there will be a relationship table for
> the movie_id>actor_id.

It's well worth looking at library systems, as they have the same sort
of hierarchical relationships between works, as publications can have
multiple (or unidentified!) authors.  Unfortunately, the fact that
names can change and are not unique makes things rather messy.

There are several people that are somewhat more famous that have my
own name, for instance:
- The writer of the "Hagar the Horrible" comic (who inherited the
mantle from his father, Dik Browne)
- If you squint a little at spelling, the R+B singer noted for beating
up Rihanna
- A notable investor that was one of the first to publicly suspect
Lord Conrad of embezzlement

You're targeting pretty fuzzy data :-).

Part of the question will be what you intend to search on.  By the
time you create a UI for doing complex queries against actors,
directors, and such, you may have something *too* complex to be
usable.

The alternate direction I'd suggest would be to look at text search
options, where the notion would be to bundle up a bunch of movie data
together, and do approximate searching, in effect, ala Google.

The example I'd point at would be the Postgres text search subsystem:
<http://www.postgresql.org/docs/current/static/textsearch.html>

Note that you could start with a normalized data model, and then
aggregate data together to generate records for use with text search.
I'd expect that to be more usable than the sort of "complex query"
screen you get in tools like Bugzilla or RT.
-- 
http://linuxfinances.info/info/linuxdistributions.html
--
The Toronto Linux Users Group.      Meetings: http://gtalug.org/
TLUG requests: Linux topics, No HTML, wrap text below 80 columns
How to UNSUBSCRIBE: http://gtalug.org/wiki/Mailing_lists