Notices of the AMS is killing it
I am somewhat surprised to hear myself say this, but this month’s Notices of the AMS is killing it. Generally speaking I think of it as rather narrowly focused but things seem to be expanding and picking up. Scanning the list of editors, they do seem to have quite a few people that want to address wider public issues that touch and are touched by mathematicians.
First, there’s an article about how the h-rank of an author is basically just the square root of the number of citations for that author. It’s called Critique of Hirsch’s Citation Index: A Combinatorial Fermi Problem and it’s written by Alexander Yong. Doesn’t surprised me too much, but there you go, people often fall in love with new fancy metrics that turn out to be simple transformations of old discarded metrics.
Second, and even more interesting to me, there’s an article that explains the mathematical vapidness of a widely cited social science paper. It’s called Does Diversity Trump Ability? An Example of the Misuse of Mathematics in the Social Sciences and it’s written by Abby Thompson. My favorite part of paper:
Oh, and here’s another excellent take-down of a part of that paper:
Let me just take this moment to say, right on, Notices of the AMS! And of course, right on Alexander Yong and Abby Thompson!
mathbabe 
Yes, I am pretty excited too at some of the responses an
LikeLike
Yes, I am pretty excited too at some of the initiatives and responses that have come out of the ICM/AMS/IMU sphere. I think people in those spheres have a good sense of the general mood of their community, and are thoughtfully approaching issues of interactions with society at large.
LikeLike
I’m reading the “ability trumps diversity” article, and I’m bothered by the phrase “code available upon request”. Argh. This is 2014; don’t we know that this is not a good way to do things?
Should it not be “code available on github”? Or be hosted by the AMS — since they host the PDFs of all these articles, it can hardly be a burden to keep a copy of their code around.
(It does seem like making the code available is not critical to their argument, but…still.)
LikeLike
I liked this:
“In this section the authors find that one collection of algorithms outperforms a second collection. They assign the label “diversity” to the first collection and the label “ability” to the second, and conclude that this is evidence that “diversity trumps ability.”
😉
LikeLike
I was pleasantly surprised by this month’s issue, as you were, and for pretty much the same reasons. I was really liked Thompson’s article. I wonder: were any responses attempted at the journal that originally published the original paper? As much as I enjoyed the article in the Notices, I wonder if it will reach the audience it needs to reach.
As a complete aside: any chance you will comment on the secret recordings of the NY Fed that came out on “This American Life”?
LikeLike
I haven’t looked at that issue of Notices yet, but I am puzzled by your title. What does “killing it” mean? When I first saw the title, I assumed it was negative – so negative, in fact, as to amount to a denunciation of Notices and all that it stands for. But then I read your article, and it sort of looks the opposite. I suppose “killing it” is some weird American slang that is beyond me. Like “slam dunk”, another phrase I have never understood.
Sorry to be so language obsessed, but I really am confused by this.
LikeLike
“Killing it”: performing at the highest level. http://www.urbandictionary.com/define.php?term=Killing%20It
or
http://www.phrasemix.com/phrases/someone-is-killing-it
LikeLike
That too!! 🙂
LikeLike
Sorry! Yes it is positive. Super positive. It basically means they are rocking out. Hope that helps.
LikeLike
Regarding the H-index, the authors point out that for those individuals where their rule of thumb 0.54*sqrt(N_Citations)~h_index fails (apparently common in e.g biology), removing the most highly cited papers makes the rule of thumb correct. This is more than a bit disingenious, since the whole *point* of the h-index is to have a measure that is not skewed by outliers. It seems to me that their paper actually proves the value of the h-index over N_citations, in particular for other disciplins than math.
LikeLike