A report from SoCal NEGT

(I wrote this during the workshop a few weeks ago, and just realized that I never actually hit “publish.”  Better late then never, I guess!)

Every year in the fall, all the folks in southern California interested in the intersection of economics and engineering/computer science get together and have a two-day workshop that we call NEGT for “Network Economics and Game Theory.” Hosting duties rotate between USC, UCLA, and Caltech, and this year it was our job.  The workshop is just wrapping up and, thanks to our amazing admin Sydney Garstang, everything went wonderfully!

There were lots of great talks, and the slides will eventually start to show up here.  Of the many highlights, our two external speakers both gave really great talks.  Our first keynote, Tim Roughgarden, gave a great overview of recent results in the area of approximate mechanism design.  This is a direction that many folks in the Algorithmic Game Theory community have been pushing on in a while, but Tim showed some very interesting new results.  Plus, it is always interesting to see how economists react to this direction, which is very different than the traditional viewpoint.  Our second keynote, Markus Mobius, gave a really interesting empirical take on the power of social learning.  He showed results from an experiment involving Harvard undergraduates performing a task that required social learning and was able to test various conjectures for how such learning occurs (as well as the magnitude of social learning that occurs).  Given the huge focus in CS on models where we learn from our friends, it was quite interesting to see that the magnitude of such social learning is actually pretty small, and seems to occur only in vary specific ways.

Continue reading

A good news / bad news week for renewables in the press

These last few weeks, the news has been full of lots of seemingly conflicting messages about renewables, so I figured it was worth talking about things a little bit in a post.

First, the good news.  The old conventional wisdom that solar can never match prices with conventional generation is just plain false at this point. Deutsche Bank recently released a report highlighting that rooftop solar will reach grid parity (i.e., be as cheap, or cheaper, than the average electricity bill in the US) in 47 states in 2016, and that it has already has reached grid parity in states accounting for more than 90% of the US electricity consumption.  Now, 2016 is a pivotal year because those numbers assume the 30% subsidies that are present today for solar, which goes away in 2016.  However, even if these drop to 10% parity will be maintained in 36 states, as this plot from the report shows (the x-axis is the electricity price /kwh – cost of solar/kwh, so positive means savings from solar):


Of course, that still includes a subsidy, so solar costs aren’t matching those of other sources yet.  However, that will happen soon if current trends continue for even a few years.  Here’s one of my favorite plots in that regard (from a World Bank analysis).  I still think it’s crazy how quickly the technology advancements and economics are working in favor of solar.

Now for the bad news. An interesting analysis emerged from some of the folks that were behind Google’s RE<C initiative, that went looking for breakthrough approaches for renewable generation that could make renewable energy cheaper than coal.  Their conclusion: no current forms of renewable energy are enough new approaches to solve climate change, even in best case forecasts.  Thus, in their words, even if RE<C had reached it’s goal, that goal was “not ambitious enough to reverse climate change.”  “To reverse climate change, our society requires something beyond today’s renewable energy technologies.”  So, in some sense, we’re too late.  But, engineers and inventors can do amazing things, so who knows what breakthroughs will come over the next 20 years. For example, one approach that would be a game changer for the models typically used would be a way to pull CO2 from the atmosphere and store it…

Sigmetrics finally has a journal

I’m a little late with the announcement, but I’m definitely excited to pass on the word about a new ACM journal: Transactions on Modeling and Performance Evaluation of Computing Systems (ToMPECS).

Despite its long history, Sigmetrics has never had an ACM journal to call its own — it’s papers are scattered over a long list of journals.  So, I think this is great news! …and it’s thanks to a lot of hard work from a lot of people in the community, especially Carey Williamson and Don Towsley.

Though Sigmetrics folks led the way in the formation, this journal is meant to be much broader than simply Sigmetrics.  From the website, here’s the scope:

It solicits and will publish articles that:

  • Define, develop, and assess new performance evaluation methodologies, including analytical techniques, experimental design, formal methods, instrumentation techniques, mathematical modeling, optimization, queueing theory, reliability analysis, simulation, statistical analysis, stochastic modeling, verification and validation, and workload characterization;
  • Provide new insights on the performance of computing and communication systems; or
  • Introduce new settings within which performance modeling and evaluation can play an important role.

Target areas for these performance evaluation methodologies include traditional areas such as computer architecture, computer networks, database systems, distributed systems, enterprise systems, fault-tolerant systems, file and I/O systems, memory systems, multimedia systems, operating systems, peer-to-peer networks, real-time systems, sensor networks, software systems, storage systems, telecommunication networks, Web-based systems, and wireless networks, as well as up-and-coming areas such as data centers, green computing and communications, energy grid networks, on-line social networks, and networks at large.

The editorial board is quite broad, and strong.  So, it looks like things are off to a good start!