Enlightening and hilarious. Love that the abstracts are now longer than the bodies, and they've retained the LaTeX/PDF submission tradition against all practicality.

My two favorite results both seem relevant to TinyTOCS itself:

* Towards an Emergent Semantic Web "Inductive fuzzy grassroots ontologies form a basis for computing with words. In time they will allow the Web to merge with the Semantic Web." (Edy Portmann UC Berkeley EECS) http://tinytocs.org/vol1/papers/tinytocs-v1-thompson.pdf

* Data Publishing Using Nanopublications "The nanopublication model incentivizes rapid, citable data dissemination, interoperability, semantic reasoning, and knowledge discovery." (Mark Thompson, Erik Schultes, Marco Roos, Barend Mons LUMC) http://tinytocs.org/vol1/papers/tinytocs-v1-thompson.pdf

I'm a tad disappointed by the abstracts associated with each submission; in effect, what's called the body is in fact the abstract and what's called the abstract is in fact the body. Still, an amusing project.

I like how the format forces the papers to be about "big concepts." So many cs papers are about 1% improvements in performance due to esoteric minutiae -- which can't be concisely explained. It biases research towards big picture, heavily-compressible ideas.

I fear I only have one point I can give you for this project. I would just request that there be a search function and web pages instead of PDF files to make the site more user friendly.

The PDF files are part of the joke. They were probably produced with LaTeX too, just like in mainstream academic publications.

Neat concept. Concise summaries really make browsing for interesting articles easier. Although there is a lot of overlap between 140-character "summaries" and plain old titles, which tend to be about the same length.

I think the body should be allowed to be a bit longer (2 pages?) and include figures, links to source code, etc. I was frustrated by the extreme brevity when I wanted to know more about specific articles.

I agree with this. Every time I found myself starting to become more interested in what the article had to say, there was no more. Seemed way too short....

I would find this method useful for some types of college projects when students can spend a few days researching but the conclusions are short(short does not necessary mean poor). I've seen plenty of papers that when you finish reading them you certainly think the author could have written the same in just a couple of sentences. Interesting tool for focusing on a fact and a result.

Amusing concept but still hangs on to the print world by organizing articles into volumes and using PDF for the articles.

I agree--we definitely opted for a more traditional publication format in this first iteration. We've been thinking of how to fix this and were considering something like HackerNews for publicly-accessible articles (e.g., arXiv). Do you have any other ideas?

I would also consider asking authors at the time of submission to explicitly release articles under a Creative Commons license such as CC-BY or CC0. That would enable a lot of fun creative reuses of the works. However some kind of exception may be required for republished works whose copyright has already been transferred to a third party publisher.

If you look at the actual papers, they're all under CC BY-NC-SA 3.0 already.

Why NC? It's overly restrictive and technically prohibits, say, distributing the papers on a website that has advertising revenue.

It would be great if all academic papers had a 140 word tl;dr version. Would save grad students a bunch of time. (Crack joke about how my phd is taking 2 years longer then expected)

TL;DR CS?

that said, I wish twitter was around when I was in school. using twitter has really taught me how to express something very concisely.

Hmm, I'm curious where I could find the originals (if there are any).

Read the citations. Some of the results are self contained, though. (eg the one from http://33bits.org/ )

Some of them cite the real paper.