freenode/#sicl - IRC Chatlog
Search
8:49:27
heisig
If I can statically determine the list of applicable methods, I could perform effective method inlining.
8:51:04
heisig
The plan is that a programmer can set inlineability of each method. If all applicable methods are inlineable, it is possible to inline the effective method.
8:52:02
heisig
There are two reasons why I wouldn't want to do the keyword parsing in a compiler macro:
8:53:00
heisig
2. Then I would need to write one function with keywords (the default), and one without keywords (the fast path).
8:54:34
heisig
My proposal is to generate the fast path automatically IF the list of applicable methods is statically known and consists of sealed methods only.
10:10:51
beach
I said this in #clim already, but we have a room at Sorell Hotel Seefeld arriving on Friday April 24 (noon-ish) and leaving early on Thursday April 30 (very early).
10:11:30
beach
Google maps says it's a 3 minute walk to the conference site. But we are a bit slow, so maybe a bit more for us.
12:47:34
shka__
i am recently game mastering RPG campaing set in a Space Opera setting, i will note "swiss star california" as an idea
17:08:39
alandipert
beach in a paper, is it proper form to critique previous work with respect to my novel claim? or can the assessment be more general
17:14:28
beach
I don't see any restriction. It is common to critique the previous work, simply because your paper typically claims an improvement on such work.
17:15:08
beach
But, it is perhaps more informative to do a more complete description of previous work. You can praise some aspects of it no problem.
17:23:13
beach
I think if you didn't claim any improvements over existing work, or something that simply has never been done, then the referees would be very confused.
17:30:06
beach
Something very interesting happened to me and my favorite coauthor last year. One referee was rejected the paper and the other one accepted it.
17:30:12
beach
The program chair must have convinced the referees to become known (they are usually anonymous), and we had a long and interesting email discussion about the paper. Finally, we agreed to make the changes that the first referee wanted.
17:30:17
beach
The first referee was one whose worked was improved upon, so he was not pleased with the critique, and thought it to be wrong. Interestingly, we found a bug in his paper that he had not noticed, and that make the criticism more accurate. The result was a much better paper in the end.
17:31:05
beach
There should be more of this kind of exchange in our field, and probably in others as well.
17:32:09
beach
On the other hand, authors should also not take criticism as an accusation of inferior work. They should welcome improvements on their work. I certainly would.
17:37:04
alandipert
was the referee somehow disincentivized to promote criticism, like maybe the work was the basis for a grant or something? when i hear these stories i feel sad, and wonder about the underlying economics at play
17:38:42
alandipert
fortunately in my case i'm combining known techniques in a novel way towards a new set of goals. so my question about how to critique is about where in the paper to critique. case-by-case in my previous work section, or when i rationalize decisions i made as part of the description of my techniques
17:39:04
alandipert
now that i think about it, perhaps i should do both and then revisit if i come out > 8 pages :-)
17:40:49
beach
That's hard to answer i general. I often have a section entitled "Benefits of our technique".
17:41:31
beach
In it, I can compare it to the "previous work" section, and it doesn't have to be case by case for a particular previous work.
17:43:12
alandipert
i think that might make the most sense in my case. especially since i only care to critique previous work as it directly relates to what i'm doing
17:44:00
beach
Dinner is imminent here. And then I will spend time with my (admittedly small) family. So i a few minutes, I'll be off.
17:44:04
alandipert
i started off by being as charitable as possible but i think the paper as a whole might suffer. since it's not a survey