I'll volunteer 10-15 hours a week to build a recommendation engine like this! I too have wanted a search-engine/feed-reader, no-holds-barred web-experiment tha

I'll volunteer 10-15 hours a week to build a recommendation engine like this! I too have wanted a search-engine/feed-reader, no-holds-barred web-experiment that does exactly what you describe. I'm not an academic, but I have fairly deep & broad reading interests across mathematics, physics, economics, history, etc. Ok, make that broad interests in basically every subject that is truly interesting. I'm the guy who pretty much reads journal articles for fun because newspapers, magazines, blogs, and most books suck and just aren't smart enough. But it's more work for me to find reading material I like and want to learn from than it is for me to read it. I could certainly hack around on something like this, as I too have been wanting to scratch this itch for a many years. I've got the time, since I'm currently un(der)-employed(ahh, gotta love contracting) and bored, but my first two roadblocks would be: 1) instant non-starter: I no longer have any university library proxy account to use to login and scrape the journals, abstracts, authors, citations, etc. I personally could care less about copyrights on academic papers. I justify my position via jury nullification using a jury of one—me. And I don't give a second thought to doing whatever dirty work it takes for building a system of scraping scripts to pull all info about every paper from every journal out there. 2) If the beta site gets even the smallest amount of traffic, I will be crushed by EC2 server and bandwidth bills. I realize a bajillion other folks probably want to build the exact same thing, and I realize it's stupid to build your own when someone else might already be well along the way and just needs some help. Anyone need a Linux/Ruby/SQL/Sysadmin/Wear-All-Hats-Get-Shit-Done developer? I'm in LA and on #freenode. Just say the word!