before i get to the meat of my post, first i'll give an amusing quote (the paper is full of them):
In mid-January 2004, the Reuters news service flashed out the headline "No More Text Messages from Jesus" signaling the demise of a distinctive Finnish mobile service. According to the wire story, earlier that month, Ville Nurmi, the Ombudsman for Finland's mobile services and regulatory watchdog organization, shut down a mobile service provider that offered text messages from Jesus Christ. The company, which was not named in the proceedings, promised to answer people's prayer with a text message from Jesus . This service, ruled spam through a complex set of maneuvers that included a determination that Jesus did not own a mobile phone, is but one manifestation of the increasing visible intersections of spiritual practice and technological development world wide.ok, now for my point. i'm not going to comment on this paper, however i did take quite a few notes, and i wanted to store and compare my notes, maybe post them so others could see them.
there is no where you can easily do this now.
and i'm not just talking about posting comments -- then everyone who ever has a comment, even if it has been said, will post it again. rather, i'm thinking of something more along the lines of post and response. so first, you'd compose your notes on the paper, and they'd be stored as your private note in the academic social networking site (let's give it a name.. i'll think of a name for my hypothetical site soon enough). now, if you wanted to make the notes on this paper public, you could, but if you wanted to add them to a conversation about the paper, you'd click the "add to conversation" button. now, what this would NOT do is blindly add your comments to the end of some big list. instead, it would first show you similar comments and any responses to those comments.
why do i think this would be useful? personally, i've been doing some trading on td ameritrade, and they have a fantastic help system. if you have a question and can't find the answer, you can type in your question and have it sent to a representative. but before they send the question, they show you possible answers and ask you, "do these satisfy your needs?" about 75% of the time they do.
in this academic case, if you find a thread that satisfies your comments, you may not post; otherwise, you might follow-up on that thread. or you might just give some sort of props (a la yelp's "cool, helpful, etc.." links) to the original commenter or responder.
this mechanism is especially important for academic papers, where comments and responses are often quite detailed, and are written anyway (whether others have written something or not... though having these comments available and reading them first might generate better discussions). more generally, this is a technique that should be broadly applied online. sometimes when i want to post a comment on a highly-trafficked blog or digg or slashdot i first read through every single comment to find if my comment has already been stated. this is extremely inefficient -- there should be much better comment search (either via search, or automatic collapsing of comments that are similar so i can explore the unique things that have been said more quickly).