First quick look at Semantic Scholar

[Posted on november 14, 2016]

Since the arXiv has announced today that it is teaming up with Semantic Scholar for a year (a search engine that has received some publicity lately), let’s have a first quick look.

What does it say for Jean-Pierre Serre ? Only two “influential publications” to his name, this is surely ridiculous. Wait, the earliest date of publication is set to 1989, and I can’t find a way to change it…

Let’s pick some recent authors then: Terence Tao. Now the default range is 1978-2016, and the 6th most relevant paper is “Professor Terence Tao Visit – 27 August Canterbury Statistics Open Day” by Jenifer Brown.  Not too convincing either.

One more try: Cedric Villani (to simulate someone typing with a qwerty keyboard without accents). Just one page of results. What difference an accent aigu makes. The suggestions in the toolbar only proposed some Cedrics and no Cédric.

I may have missed some obvious settings, but if not I don’t think it is very useful yet.

***

In other news:

• an interesting paper in PLOS One: The Effect of Gender in the Publication Patterns in Mathematics, by Helena Mihaljević-Brandt, Lucía Santamaría & Marco Tullney (as mentionned by the twitter account of SMF)
• A recent talk, “The role of curvature in algebraic geometry: some history and current issues” by Phillip Griffiths
• not quite mathematical, but puzzling all the same: Yale University Press has just released a high-quality fac-similé of the Voynich Manuscript at an affordable price (see also this news release)

 

AMS data on the backlog of math journals

[Posted on november 11, 2016.]

Since every year in november a new batch of data regarding the backlogs of mathematics research journals is published in Notices of the AMS, let’s look at this a little. Here is the data from the past 6 years : 2015, 2014, 2013, 2012, 2011, 2010, 2009.

Now, while backlogs are one thing, perhaps the most significant column is “Median Time (in months) from Submission to Final Acceptance”. The results for a subjective selection of some of the most well-known titles are as follows (click to enlarge) :

Those median times are thus usually very steady and under 12 months, which is conforting, but with some occasional sudden marked increases for the very select journals that can reach 20 months, which can be a problem for young folks on the job market…

Are there important informations from the whole data worth mentioning beyond this? Comments welcome.

***

In other news, some items noticed recently :

• there will be a Gabberfest next june at IHÉS with an A-list of speakers (there are several anecdotes about Gabber’s aura on the web, whether on MO or blogs)
• next summer at the Newton Institute is due to take place a promising Big Proofs Programme aimed “at the challenges of bringing proof technology into mainstream mathematical practice“
• the Institut Fourier turns 50
• the 5th digit in arXiv identifiers was used fo the first time last month, the counter reaching 10100
• integral calculus was performed for the first time by Leibnitz on this very day, some 341 years ago
• some people are remarkably versatile: Pierre Jalinière’s very recent PhD Thesis deals with “three independent works in cryptography, p-adic Hodge theory and Numerical analysis“!

Heart-shaped devotion by Marneejill on flickr

(which, for some reason, I’m inclined to rename as Two Americas…)