First 2020 notes
22 Jan 2020The first set of notes of 2020.
A picture from January 2020. Magellanic penguins.
Conference model: outcome of the PODC survey and remote attendance
As reported here in November, the community of the theory of distributed computing is in the process of changing some aspects of its conference model. A survey has been proposed that has received many answers. The email summarizing these answers is here. It is fairly short, and I don’t have much to say, so I’ll just list the topics discussed:
- changing to several deadlines a year, to avoid a large gap between important deadlines,
- a transition to a system with a journal,
- collocation of the two main conferences.
A topic mentioned in a previous mail was that this kind of survey is more useful than business meetings: in a business meeting, most people are afraid to talk, some people are very vocal and do not let others talk, and more importantly, it’s late, and everybody wants to leave the room.
Another related text is by Moshe Vardi in the Communication of the ACM, see here. He basically says that conferences have a big environmental impact, and that we should allow people to participate via video. The usual answer to this is that you would loose a lot of informal interaction between participants. He says he thinks it’s not as bad as it looks.
Selfish routing and traffic lights
A recent paper on the arxiv, consider the classic model of routing but with traffic lights. As discussed recently on this blog (see here) a classic problem in algorithmic game theory is to evaluate how good is the traffic on a network, if you allow each car to chose selfishly the best route. A surprising phenomenon in this model is that sometimes opening a new street can slow down the traffic (Braess’s paradox). The authors show that this does not happen when the network is equipped with some traffic lights.
Sunset geometry
- Which of the two pictures below looks more like a sunset over a very calm lake?
If the Earth is flat then it’s the one on the left. If it’s spheric, it’s the one on the right, and you can even compute the radius of the planet from the picture!
See this for an explanation using trigonometry, and that for one using geometric algebra.
[I learned about this on Eppstein’s blog.]
Other notes
-
A bit more of geometry if you read French or Spanish: the former flag of Chile has a very nice geometric construction. See here for the article in French, and here in Spanish.
-
DBLP now keeps track of the citations of the papers. You can access them by clicking on the page symbol close to the colored square. Not all the citations are present because they use an open access database, and of course, not everything is open.
-
I used to use latexdraw to generate pictures that can be included in latex. That is you draw, like on a software like Paint, and it generates the code of this picture. Unfortunately the output is only pstricks code, and one would sometimes like a tikz code. I discovered yesterday mathcha.io which is an online latex editor, and does the same but generates tikz code. [Thanks to Pedro Montealegre for showing me this.]
-
We mentioned robust statistics on this blog some time ago; if you’re interested, there is a survey on the recent advances on this topic here.