Wednesday, March 22, 2017

Accepted papers at LICS 2017

The list of accepted papers for LICS 2017 is now available at http://lics.rwth-aachen.de/lics17/accepted.html. Thanks to Joel Ouaknine and his PC for their work in selecting such a mouth-watering list of papers!

If you have not done so already, go ahead and register for the conference now. We strongly encourage prospective participants to register, and to make their travel and accommodation plans ASAP. Iceland gets literally fully booked early during the summer months.

Tuesday, March 21, 2017

Opinions on the ERC after ten years

I am collecting some opinions about the European Research Council (ERC) from researchers who have received funding from it, and from some who have not.

What is your overall opinion on the ERC? Do you think that it is good for European research?

My interest in this matter started because the ERC is ten (and so it might be a good time to draw a preliminary assessment of its impact on the European research environment) and was piqued by the opinions aired by the Italian physicist Sylos Labini who claimed that the ERC has become the main problem in European research funding. He says that there are three main problems with the ERC.
  1. The first is that it uses "research excellence" to mask political choices.
  2. The second is that rewarding today's excellence does nothing to support the excellence of tomorrow. Moreover, one does not reward excellent research by giving money to the top 5% of those who apply. 
  3. The third is that the ERC gives a bad example to national funding agencies in Europe, who also reward excellence. 
See also http://www.sciencemag.org/news/2017/03/10-europe-s-excellence-fund-faces-calls-change, where Sylos Labini makes a cameo appearance in this paragraph:

But some chafe at the singular focus on excellence. Countries in southern Europe have cut their research budgets during the economic crisis, and now ERC is further weakening these countries by essentially redistributing their EU contributions to the research powerhouses in the north, says Francesco Sylos Labini, a physicist at the Enrico Fermi Center in Rome. And it’s not just the money: “The few Italian researchers that get an ERC grant go to Germany or another country to do their research,” he says.

I do not a long piece of text. Just a few lines would do. I'd be grateful if you could contribute to this discussion by posting your comments.

You might also wish to read Helga Nowotny's short piece entitled ERC---the next ten years.

Monday, March 13, 2017

Rogers Hollingsworth on Major Discoveries of the American System of Science


Recently, I have been reading a fairly long study by historian Rogers Hollingsworth on what makes research organizations produce major discoveries. This talk he gave at the NSF summarizes some of the findings in that article. From around minute 50 in the presentation, the speaker presents some thoughtful closing comments on the current state of funding for basic science and on the effects that the involvement of politicians in making decisions on how to allocate that funding, and in monitoring and auditing the work of scientists, can have on science. For the little that it is worth, I found those thoughts worth mulling over.


Friday, March 10, 2017

Has the Feder-Vardi dichotomy conjecture been proved? (Take 2)

In this post, I mentioned that a paper by Arash Rafiey, Jeff Kinne and Tomás Feder claiming a solution to the Feder-Vardi  dichotomy conjecture appeared on the arXiv on the 11th of January. The result in that paper was also covered by Lance Fortnow in this blog post.

We now have another preprint, A dichotomy theorem for nonuniform CSPs (71 pages) by Andrei Bulatov, claiming a solution to this long-standing conjecture. (Thanks to Moshe Vardi for pointing this paper out to me.) Not surprisingly, Andrei's paper  uses "the algebraic approach that associates to every relational structure its (universal) algebra of polymorphisms."

I am not an expert on this topic, but I hope that those who are (and especially the prime movers behind the universal algebraic approach to the problem) will stop what they are doing, read this paper carefully and vet its correctness,

I reiterate what I wrote in my earlier post: "If the technical content of the paper is found to be correct by the community working on CSPs after careful peer review, this is a truly major result."

The Feder-Vardi  dichotomy conjecture seems ripe for a solution. We now have two papers, claiming a solution to that conjecture, that use very different techniques. I hope that they are both right and that the tools developed in those articles will find application in the solution of other problems.





Wednesday, March 01, 2017

Registration for LICS 2017 is now open

We are happy to inform you that registration for LICS 2017 and affiliated workshops is open. You can register for the conference and/or its seven co-located workshops by following this link.

As you will see, due to the pressure of the tourist industry (which has increased enormously since we made the bid for hosting the conference in 2015), we are forced to have Friday, 7 April, as early registration deadline. We strongly encourage prospective participants to make their accommodation and travel plans as early as possible, and the date for the early registration is meant to entice them to do so.

As you will see, the early registration fees for the conference are as follows:
  • Early registration fee (members of ACM, ACM SIGLOG, ASL, EATCS or IEEE): 64,000 ISK
  • Early student registration fee: 46,000 ISK
  • Early registration fee (others): 88,000 ISK
The regular registration fees include the cost of the rooms, technical equipment, four lunches (one per conference day), eight coffee breaks (two per conference day), welcome reception and conference dinner at Kolabrautin (three course meal with three glasses of wine per participant). The student registration fees include all the above items, apart from the conference dinner.

In order to meet ACM regulations, the registration fee for members is 25% cheaper than for non-members. We therefore strongly encourage prospective participants who are not members of the ACM, ASL, EATCS or IEEE to join one of those associations. When choosing which association to join, consider the following things:
  1. The EATCS annual membership fee is of 30€. (For students, that would give a two-year membership.) The EATCS sponsors the Kleene Award at LICS 2017 and its Bulletin of the EATCS is published in an open-access form.
  2. Information on ACM SIGLOG membership is available here. There is no need to have a separate ACM membership in order to join SIGLOG. 
  3. The ACM and the IEEE already take 36% of the registration fees and provide no financial sponsorship for the conference, apart from 100K ISK that were kindly offered to us by our friends in the Icelandic Section of the IEEE. 
  4. ACM professional membership costs 99 USD. 
  5. Membership of the IEEE Computer Society costs 60 USD.
  6. A discounted introductory rate for members of the ASL  is available to new regular members of the Association for the first two consecutive years of membership (US$47 in 2017 and 2018).
The LICS 2017 Organizing Committee hopes that this information helps and looks forward to welcoming you in Reykjavik for LICS 2017 and its seven affiliated workshops.