1/28/09. Downloaded favoritedata.txt from the Metafilter InfoDump (Currently down after the recent SQL Injection attack on the site). This has all the information that I need for the Fav-O-Tron: Every post or comment marked as favorite correlated with with the usernumber of the favoriter and the usernumber of the favoritee. I need to do some text processing with regular expressions to filter other data from this file that I don't need. After that, I can SSH-tunnel into my MySQL instance on my ISP and use the LOAD DATA INFILE MySQL command to get the data into my MySQL database. I know very little about text processing and I need to do some research. I'm hoping that I can use the unix grep and pipe commands to strip away the text that I don't need, but if I can't, I'll need to write a Perl script or something, and it's been a long time since I've used Perl. At any rate, I have a freelance web design project that requires my attention for the next few days. I'm hoping to get back to this next week.

1/27/09. Welcome to the MetaFilter Fav-O-Tron 3000 Stage 1 Development Blog. On the Metafilter community weblog, users can 'favorite' the posts and comments of other users (or their own, though this is frowned upon). For example, here is the list of posts and comments that I have favorited . For Fav-O-Tron Stage 1, I'll be constructing a MySQL database of each favorite, with the 'favoriter' and the 'favoritee'. I'll make a web front end that takes a username or usernumber, and returns a color coded list like the following, supposing that I put in my own username, 'Kwine', or my own usernumber, '41973'...

Kwine's Favoriteees
John(#)
Paul(#)
George(#)
Ringo(#)
Kwine's Favoriters
George(#)
John(#)
Ringo(#)
Paul(#)

...where "Kwine's Favoritees" is a list of those who have favorited my content, "Kwine's Favoriters" is a list of those whose content I have favorited, "John, Paul, " et. al are other Metafilter users, "#" is the number of favorites given or received, and the text color relates to that number: the bigger the number, the "warmer" the color (Perhaps obviously, the actual results will often be a good deal more complicated than) these examples). Stage 2 of the Fav-O-Tron 3000 will follow Stage 1, and will turn these results into a cool visual map of favorite relationships.