June 22nd, 2004

  • fg

ljArchive 0.9.2

ljArchive now supports downloading/browsing/searching comments.  In a nutshell:

  • Added support for downloading/browsing/searching/exporting comments.

  • Added a new plugin that provides comment analysis, both how many comments received from each user and given to each user

  • Entries and comments both displayed in a single view, via the updated HTML template system.

  • Updated loading/saving code - new file format is about 35% smaller and loads approximately 6x faster than the old format.

  • Various bug-fixes from user feedback.
Here are some screenshots of the new functionality:The web site for downloading ljArchive is here.  Those of you with monster journals with 10,000 comments or more, are especially welcome to try this new version, and please let me know how it performs     :)
  • bryant

Birthday Scraping

I'm interested in getting a list of user X's friends with birthdays. The context is a web page, not a client per se. I can see a few ways to do this:

a) Get X's FOAF data, parse the friends, then get the FOAF data for each of them. Advantages: uses the bot interface, no need to log in as X. Disadvantages: lots of HTTP requests.
b) Scrape X's userinfo, parse out the friends, then scrape their userinfo. Advantages: no need to log in as X. Disadvantages: strictly inferior to method a.
c) Require X to enter her password, and use getfriends. Advantages: uses the client API, single HTTP request. Disadvantages: ew, requiring passwords.

No matter which method is used, there will be caching of data.

I am not so emotionally committed to this project that I will be offended if anyone says "gee, that project is a bad idea." Someone said "Oh, it would be neat if" and I got interested. I think if the three options above are the only possibilities, I'll probably just skip it. However, I wanted to ask before I punted.