Story Reading

I revisited an old story I liked and decided to read it aloud.  The story is "Chicken Hawk's Dream" by Al Young and appeared in Spero, 1,2 (1966:19--21). Here is a version.  Here is my story reading performance.

Fixing IPL Games: Market Design Issues

Indian Premier League (IPL) is an innovation. It takes a classic 5 days long cricket game that had whittled down to its 1 day version sometime in the past, further shaves it down to a few hours, and built a professional league. Now each game is like a long Bollywood movie, and people tell me they are just as popular, even managing to pull in a large female audience thanks to the halftime drama.

This is an interesting market. On the positive side,  IPL has built a fat top pyramid and expanded the opportunities of cricketing talent worldwide, together with all the trickle-down effects. On the negative side, there is the ongoing match fixing scandal, which is interesting for market design:

• They do spot-fixing (target number of runs per over, etc) and not match-fixing. Slicing the full match into nuggets yields a more compelling market. 
• The players use signalling schemes like towels tugged in etc, the bookies signal using WhatsApp and other cross platform messaging systems.
• During a 150sec break, apparently transactions total US$100M. They use software like "Back N Lay Pro" to estimate probabilities and set up bets. 

Btw, this is illegal. 

Announcing SPARC 13

The Coding, Complexity and Sparsity workshop (SPARC 13) will be held in Ann Arbor, MI on Aug 5-7. These will be hour-long lectures designed to give students an introduction to coding theory, complexity theory/pseudo-randomness, and compressive sensing/streaming algorithms. We will have a poster session during the workshop and everyone is welcome to bring a poster but graduate students and postdocs are especially encouraged to give a poster presentation. There is some funding for graduate students and postdocs with preference given to those who will be presenting posters. For registration and other details, please look at the workshop webpage: http://eecs.umich.edu/eecs/SPARC2013/

Lighthearted Symmetry

I was in a conversation with thoughtful and confrontational folks -- technical people --- sometime ago, and they asked me in  a provoking way, "what is the big deal with Symmetry, why do we care?" This was obviously meant to skewer me trying to defend the math, science and beauty of symmetry. Humans make symmetric objects (airplanes, say), nature makes symmetric objects (leaves of trees, say), and symmetry abounds in the particles of the Universe. But the conversation was in a context where I had to reach for something more down to Earth. So, I asked the men in the group, "Say you need to go to the restroom and you hurry, but find the women's. Where will you look for the men's?" They understood the importance of symmetry. :)

ps: When I asked some of the women, "Say you need to go to the restroom and you hurry, but find the men's. Where will you look for the women's?" Some of them said, "I will ask someone." This was smart, and it killed my argument.

Blickets and Streams

I have been reviewing developments in streaming research and wondering if there were any new techniques beyond group testing and projections. In a related and unrelated note, came across blickets -- a game for testing the logic in children, developed at UC Berkeley -- which in its OR version is a simple group testing problem, and is more complicated in its AND variation:
http://www.nytimes.com/video/2013/04/23/science/100000002187112/buckets-of-blickets-children-and-logic.html

On Writing

Someone asked me if I still get time to write fiction. Sentences twist in my mind and float, plots form and fade, and my mind struggles with the puzzles of how to start, continue and end, but the pages still remain empty. Here is the start of a story. Today I am fatigued and I will let this story go, like I have had to let go many other things.  I hope someone, someday pursues the following opening.

• When people smile, you can tell when they learned to smile, was it as a child, fresh and without feedback,  or as a preening teen, practiced in the front of the mirror and as a invitation to be a friend, or as a grown adult, when one has identified and accepted their quirks and strengths, the smile is whichever element won out, and will persist until one is laid down under the ground. She of course had the smile she learned as a child, simple when it appeared, expressing her joy and nothing else. I cared for her and vowed I wont let her smile change, and of course I failed. 

EC 13 PC Kudos

As a PC member, some papers I review are sure accepts, some are sure rejects, it is in the majority of the middle solids I struggle. It is in how this majority is handled that PCs distinguish themselves.  I submitted something to EC 13. I got feedback that was more than the reviewers' notes. The feedback summarized the discussions the PC had, which was technical, found a flaw deep in the proof, discussed how to fix it pretty much how I would have thought about it, and also noted that the PC appreciated the contribution of the novelty of the model.  It felt like a two sided conversation between the authors and the PC. This feedback reinforced my belief and appreciation of the PCs in theory conferences. Thanks and congratulations to PC Chairs, Preston McAfee and Eva Tardos. The paper was not accepted, which was apt.

Indian Hiatus and Linguistics

It has been a while since I blogged, mainly because I have been more quantum than ever, focused on being at India for personal reasons.  I have been carrying many entries in my head, but alas,  if you dont blog them down, they are not real.

Here is humor for the fans of Indian vernacular. I found a way to stay in one of  Chettinad Mansions (some info). These are spectacular, super large multifamily homes, eerily empty, spread out in the South of India. I ate something amazing and asked for the recipe: the staff said, "Put one google". :)

I went to a tourist site and listened to the official audio tapes. And checked out some childrens books. Both these experiences reminded me that what India needs is new vernacular and narrative. The language in these media is so antiquated, they dont give modern Indians a new way to see their past or their present. Finally, I saw a clique of monkeys and tried to feed bananas and mangos to the mothers with  little ones, defying the hierarchy of the two big males. That experiment in breaking social structure did not work out well. 

New Conference on OSNs

A new conf on online social networks, this by ACM, and hope this consolidates a lot of work across communities on OSNs.

ACM Conference on Online Social Networks
October 7-8, 2013 Boston, USA

With well over a billion people as members, today's online social networks (OSN) pervade all aspects of our daily lives. OSNs have grown beyond platforms for social communication and news dissemination, to indispensable tools for professional networking, social recommendations, and online content curation. Their usage has influenced today's societal and cultural issues, and changed the way we see ourselves and communicate with each other.

Not surprisingly, study and research in OSNs is highly interdisciplinary, and participants include researchers from networking and systems, databases and data mining, security and privacy, and modeling and analysis. For a number of years, researchers have published in disparate venues focused in their own areas, and have lacked a common platform to congregate and exchange ideas. This has limited communication between like-minded researchers, and led to repeated and sometimes conflicting results across disjoint venues.

The COSN (Conference on Online Social Networks) is organized with this challenge in mind. Our goal is to provide a premier publication venue that features high quality research from academia and industry across multiple disciplines focused around the study of OSNs. Of particular interest are works that focus on systems, security and privacy, graphs, data management, analysis, and data mining.  We solicit papers in broad areas relevant to the design, analysis and development of OSNs.

Topics of interest include:
    Novel social applications and systems
    Systems and algorithms for social search
    Infrastructure support for social networks and systems
    Social properties in systems design
    Clean-slate designs for social systems
    Measurement and analysis of social and crowdsourcing systems
    Graph analysis and visualization
    Processing and query optimization for large graphs
    Management of social network data
    Modeling Social Networks and behavior
    Streaming algorithms for social data
    Benchmarking, modeling, performance and workload characterization
    Information extraction and diffusion
    Data mining and machine learning in social networks
    Privacy in data collection and management
    Privacy and Security in social systems
    Privacy-preserving mechanisms for social and mobile data analysis
    Information disclosure and its impact on social networks
    Tracking social footprint / identities across social networks
    Trust systems and trustworthiness of social content/media
    Trust and reputations in social systems
    Detection, analysis, prevention of spam, phishing, and misbehavior in social systems
    Social and psychological understanding of these topics

COSN is the result of merger of 6 workshops held over the last few years in various venues: WOSN (Sigcomm/Usenix) SNS (Eurosys) DBSocial (SIGMOD) HotSocial (KDD) PSOSM (WWW) WOSS (VLDB)

Important dates:
        Abstract submission due:  Friday June 21 23:59 GMT
        Paper submission due:     Friday June 28 23:59 GMT      
        Acceptance notification:  July 31
        Camera-ready copy due:    Mon September 2
        Conference                Mon-Tue October 7-8, Boston

COSN 2013 allows two forms of submissions (PDF only):   Full papers (up to 12 pages including references) describing original research in detail.  Short papers (up to 6 pages including references) conveying promising work/high-level vision

All submissions must satisfy the following requirements:
  10-point font, two-column format, letter page size (11 x 8.5 inches)
  Names/affiliations of all authors on title page
  Submission site: http://cosn13.cs.ucsb.edu/
  Use style file at http://cosn13.cs.ucsb.edu/sig-alternate-10pt.cls
  Non-compliant submissions will be rejected without review.

The website for COSN is http://cosn.acm.org and will be populated soon.