Awesome Feature Request

One of the reasons I love Geeklog is that the SID\'s are incremental numbers. For example, an article SID would be 2002072032...etc. which is taken directly from the date and time. I think this method of IDing articles is very good, because if you had incremental numbers of, say, 1, 2, 3, 4, and then deleted article #3, it would, in essence, put a break in the organization of your database.

I\'m an organization freak.

Though the SID\'s are all well and good, the PID\'s (each comment ID) are incremental, so if you deleted a story, although it wouldn\'t disorganize the story id\'s, it would disorganize the comment ID\'s.

What if each comment ID used the date/time method of IDing as well? That way, if at a future date the webmaster decides to delete a story/comment, whatever, there would be no disorganization of numbers in his database since all the ID\'s are date based.

I propose that each comment be date/based as well. It could be something like....

if the story (SID) ID is 20020723193327163

then the comment (PID) ID is 200207231933271631 (notice the one on the end) then if there is a second comment in the article, *that* comment ID would be 200207231933271632 (notice the two on the end)

if you have a new story, the extra number at the end of the SID (which would be the comment\'s id) would restart back to 1 for each new story. this way, each story and comment would have a unique id, that if deleted, wouldn\'t make \"breaks\" that disorganize one\'s database.

For an example of a CMS that implements this feature, check out the Scoop! Engine at http://scoop.kuro5hin.org/

Kuro5hin.org runs scoop, and they have their ID\'s for stories and comments setup in such a similar manner that if they ever deleted a story or comment, there would be no break in the number pattern sequence. It\'s beautiful!

Honestly, I would have used Scoop instead of geeklog for that very feature were it not for the fact that I\'m on a vhost without root level access :(

Edwin Day
http://www.loogy.com