Directory Names for Movies, Albums, etc.

hmmmm depends how the data is going to be shown

if there are going to be alot on a page use a unique id for each
or if its just a page with a link somthing small use number 3

keep everything lowercase
wouldnt recomend using word seperators or spaces
the completeness would help because you wont have to remember what the move is. it will tell you

if your going to have alot of movies i would sort them buy date yes, probably by year would be yes

give each film a unique id, will help for comments, browsing etc etc

yep a unique id is good and it keeps everything in good order and not look messy

Thanks for the replies!

Two more questions...


QUESTION #1
What happens if you have albums and songs?

You now have to choose whether to have a song be "within" an album, *or* have a song be a sibling of an album. Since a song can exist on multiple albums (compilations, live CDs, etc.), you don't have a clean tree (1-to-many) relationship but instead a web (many-to-many). To me, this is an argument for keeping songs at the same level as albums, but I'd like to know what you think:

Here are some options: (using "Beat It" on Michael Jackson's "Thriller")

1) /albums/thriller/beatit/
2) /albums/thriller/songs/beatit/
3) /songs/beatit/
4) /songs/b/beatit/ (a folder for each letter)

#3 will result in a TON of directories, but if you start adding dates or album names to the directory structure (e.g., /songs/thriller/beatit/ or /songs/1985/beatit/, you are basically forcing a 1-to-many relationship, which is imperfect in this case).


QUESTION #2
What about versions?

Let's say MJ has 10 different versions of "Beat It", some live, some studio, some recorded on different dates, some on the same date... How would you organize each of these versions, in terms of directories?

Let's say there's a version (the third) recorded live on Christmas 1987:

1) /songs/beatit/ (all versions discussed on one page)
2) /songs/beatit/version3.html (exposes file extension)
3) /songs/beatit/version3/ (subdir for each version)
4) /songs/beatit_version3/ (each version treated as a standalone song)
5) /songs/beatit_03/ (shorter version id)
6) /songs/beatit/03/
7) /songs/beatit/live19871225/ (verbose version name)

Which of these options is most likely to be usable and still around 20 years from now? Which makes it easier to add versions w/o shifting around page contents?

Thanks!

Travis

There should be a consistent algorithm that turns a title into a directory name.

1) It should be "mindless" to apply; we should never have to make judgement calls. Given an input (a title), there should be only a single viable output (directory name). It should be automatable.

2) It should be flexible enough to handle "realistic" input in the current problem domain, but not much more. Flexibility is expensive. We shouldn't start thinking about movies and books if all we need to handle is music.

3) The algorithm should not influence the inputs. If we decide to map "Beat It" to "beatit", that might discourageMichael Jackson from releasing a song called "Be a Tit". Not cool. (either the title or the notion of the algorithm influencing the input) "beat_it" or "beat-it" would be better. "Beat It" would be perfect, except for the space and the capitals... *as long as* Michael doesn't release two songs by the same name. What about versions???

Travis

Wow, this board is quiet! Either that, or I asked a hard (or boring) question!

Is there a more appropriate forum for this kind of topic?

Travis

Just keep the directory structure logical and simple. That should work best!

Thanks for the reply! Unfortunately, "logical" and "simple" are exactly what I'm trying to define!

Well, I was sort of pointing towards not using underscores or dashes...