alexr_rwx: (my fandom writes your software)
Alex R ([personal profile] alexr_rwx) wrote2006-04-28 01:07 am

Question for you!

Academic and studently sorts! What do you do with papers that other people wrote? You get this fresh shiny article, full of really important information that you're going to use or whatever...

Do you have them nicely organized in folders, by topic and author? Do you just delete them? Are they PDFs? postscript? Word docs?

What if you had something that was kind of iTunes-like, but for papers and slides and stuff, and you could sort and search and organize and have a central place to stash all of them? What if I wrote something like that, and it was cross-platform and easy to use and exciting and awesome?

[identity profile] gtv42.livejournal.com 2006-04-28 10:45 am (UTC)(link)
Are they PDFs? postscript? Word docs?

Yes.

Following the iTunes analogy, I'd like to have the ability to create a sort of "on the go playlist". While I'm viewing the documents in your program, if I could highlight sections and have them flagged/collected in another document (perhaps in the form of links to the original), that would be nifty. I typically print out all the papers I use for something and take a highlighter to the portions I want, but if there were an easy way to do this without involving dead trees, man...

Would it be possible for the program to identify unique language in each paper and scan your document for that language as a sort of "citation check"? Like, it also checks for common citation formats nearby, but if they're not there it says "This paragraph 70% match for 'Boron Carbide Applications' by Speyer, Robert. Add citation?" and gives you the chance to break out of the check cycle for a moment and add a citation.

I'm just thinking "outloud". Something to collect and organize papers would be wonderful in whatever form it takes.
ext_110843: (coffee)

[identity profile] oniugnip.livejournal.com 2006-04-28 02:39 pm (UTC)(link)
Would it be possible for the program to identify unique language in each paper and scan your document for that language as a sort of "citation check"?

Probably :) I would do that like this (http://en.wikipedia.org/wiki/Vector_space_model), and probably like this (http://en.wikipedia.org/wiki/Latent_semantic_analysis) too, which put together is just like "compare things using vectors and linear algebra and the blood of newts".

... would you want that if you had to do it as a separate step, though? Anyway, I think I might hold off on this feature, but it's a really interesting idea!

[identity profile] gtv42.livejournal.com 2006-04-28 04:15 pm (UTC)(link)
It was a seperate step in my mind. And the same sort of functionality might be useful in identifying connections between source documents; if you have a pool of papers that you've collected over time, useful connections between them might go unnoticed. If the program looks at all the documents dogeared for an active paper, it could see what they have in common (on a superficial level, at least) and look for that in other places. Wouldn't often yield tremendous insight, but it could be handy.