What's the most recent paper that you read and found interesting?
Mine is from POPL, with the titel "Equality Saturation: A New Approach to Optimization" - http://www.cs.ucsd.edu/~rtate/publications/eqsat.html
-
-
Nobody reading papers?
-
I'm currently reading a Computer Vision paper on object contour tracking, it's taking me a long while though since I'm familiar with everything being discussed.littleguru said:Nobody reading papers?
-
I found this one very interesting:
DryadLINQ: A System for General-Purpose Distributed Data-Parallel Computing Using a High-Level Language Pretty much the current state of the art in my research area. Charles, if you can get an interview with the guys behind Dryad and/or DryadLINQ on C9 I would be eternally grateful.
-
I read the newspaper this morning...littleguru said:Nobody reading papers?
-
You mean you watched the cartoons in the newspaper this morning, don't you?Bas said:
I read the newspaper this morning...littleguru said:*snip* -
TommyCarlier said:
You mean you watched the cartoons in the newspaper this morning, don't you?Bas said:*snip*I confess.

-
None of my newspapers I read have comics in them... they're all too seriousTommyCarlier said:
You mean you watched the cartoons in the newspaper this morning, don't you?Bas said:*snip*
...but then you always get the up-to-date ones online anyway.
-
Will look into this. It's certainly very interesting work and a generally compelling problem space. I'm thinking this would make for an excellent Expert to Expert episode....Sven Groot said:I found this one very interesting:
DryadLINQ: A System for General-Purpose Distributed Data-Parallel Computing Using a High-Level Language Pretty much the current state of the art in my research area. Charles, if you can get an interview with the guys behind Dryad and/or DryadLINQ on C9 I would be eternally grateful.
C -
I would also like to see something on DryadLINQ, if possibleCharles said:
Will look into this. It's certainly very interesting work and a generally compelling problem space. I'm thinking this would make for an excellent Expert to Expert episode....Sven Groot said:*snip*
C
The E2E seems a nice idea... now you only need to grap Erik and get everything set up.
-
Lovely.
I wonder if the Phoenix guys are thinking D'OH! right now. This approach looks quite compelling.
There's been several interesting ideas introduced lately, some new, some not so new. The new TraceMonkey Javascript VM, working on the new idea of trace trees. I believe the researchers behind the idea applied this to Java as well and got a very high performance VM, comparable to the one of Hotspot, which has God knows how many man-hours put into it. Isn't it wonderful when new and simple ideas shatter old and complex ones? Beautiful code.
Then there is SquirrelFish Extreme and V8 Javascript engines which both appear to be using some of the same techniques from the Self implementations - re/compiling to classes because most often, prototypes do have a fixed layout, even if they may be completely dynamic - adding and removing properties and methods at runtime. -
esoteric said:Lovely.
I wonder if the Phoenix guys are thinking D'OH! right now. This approach looks quite compelling.
There's been several interesting ideas introduced lately, some new, some not so new. The new TraceMonkey Javascript VM, working on the new idea of trace trees. I believe the researchers behind the idea applied this to Java as well and got a very high performance VM, comparable to the one of Hotspot, which has God knows how many man-hours put into it. Isn't it wonderful when new and simple ideas shatter old and complex ones? Beautiful code.
Then there is SquirrelFish Extreme and V8 Javascript engines which both appear to be using some of the same techniques from the Self implementations - re/compiling to classes because most often, prototypes do have a fixed layout, even if they may be completely dynamic - adding and removing properties and methods at runtime.I love simple
I'm also interested to know what the Phoenix guys think about these kind of ideas... but they might already know, since this was a paper that has been presented at POPL. -
Every few months I re-trawl the Patterns and Languages Conference papers (here is the list from 2008). I usually find something interesting.
Herbie
-
I've been watching lots of tutorial videos lately. That's kind of like reading technical papers

of course they're all about Reason 4
http://ca.youtube.com/watch?v=TfzmIGn4mf8
Thread Closed
This thread is kinda stale and has been closed but if you'd like to continue the conversation, please create a new thread in our Forums,
or Contact Us and let us know.