Tuesday, November 26, 2013

News Worth Reading

There's been plenty of interesting stuff popping up in the news, so it's time to collect a bit.

This article describes a current patent case in cryptography, where apparently Ron Rivest testified (by deposition) and Whitfield Diffie took the stand.  While I can't help but wonder if the article has dramatized the proceedings a bit, it does give some idea of what patent trials are like.  I found it interesting both legally and technically.  Anyone interested in the legal/expert witness side of technical work should find it worthwhile.

Update:  The case was decided, as Suresh notes.

In other legal news worth noting, Google Books was found to be fair use.  The decision is online (and available at the link) -- I'd definitely recommend reading it if you're interested in fair use and the legal framework for how fair use is currently (at least of this decision) is understood.    

Harvard's CS50 course just made the Boston Globe business section.  I'll quote the conclusion from the new boss, David Parkes:  “There’s a new willingness among the student body to take risks, to not follow what has been the default path of going into medical school or going into finance,” said David Parkes, Harvard’s dean for computer science. “I think part of it is that students are seeing a new way to contribute to society through computer science”

Michael Nielsen put up a chapter of the latest book he's working on, on neural networks. 

I found myself more interested than I expected when reading the article How Academia Resembles a Drug Gang.  The issue of PhD overproduction is already well known, but this brought in a new dimension for me, with the discussion of "dualisation" -- employment systems with an "insider" class with unusually high benefits (supposedly, from the article, tenured professors) and a large "outsider" class of people trying to get on the inside and supporting their lifestyle (from the article, untenured part-time faculty and PhD students).  Probably worth thinking about in terms of trends in our field, but also just generally, I'm now curious about the economics.  Dualisation doesn't seem like it would lead to long-term stable systems -- what's the model?

I'll have to pay more attention to the news myself these days.  I was asked to serve on the Communications of the ACM editorial board, news division, and found myself unable to find a suitable reason to decline. 

Finally, a question.  'Tis the season to purchase some new machinery.  So is retina display for my next laptop a must-have, not worth it, or somewhere in between?  I'd appreciate opinions. 


Michele Filannino said...

I have a Retina Display 13". I have to admit that the day before selecting it, I went to the closest Apple Store and I closely compared Retina and not-Retina: the Terminal on a Retina Display is much much better. :)

I think that Retina Display is worth the money. I have more workspace (more pixels), better images, it's very bright and ... the overall machine's architecture is the state-of-the-art (SSD disk, fast processors, light).

I am quite happy. I think that if I had to buy a new laptop now, I wouldn't have any doubt (money permitting).

Suresh Venkatasubramanian said...

Since you read PDFs, a retina display is a must. And btw newegg just lost the patent suit inspite of the dramatic testimony.

Isn't dualization akin to a two-level Ponzi scheme ? In that the expected reward is out of sync with the true expected rewards

Unknown said...

Nice post with awesome points! Can’t wait for the next one.

Samsung - 13.3" Ultrabook - 4 GB Memory and 256 GB Solid State Drive - Ash Black

Samsung - 13.3" Ultrabook - 8 GB Memory and 128 GB Solid State Drive - Silver

Rob1606 said...

About the long-term stability: it would seem to me that the system is not stable, at least not from the point of view of the individual participants. Outsiders keep dropping out, only to be replaced by new outsiders. But perhaps this is not what you meant.

Anonymous said...

I think "dualization" with "insiders" and "outsiders" does lead to a stable system if there is a enough propaganda to convince the outsiders that they can one day be insiders.

The entire US is a dual system where rich people who make most of their money from capital gains (i.e. not from "earned income"), pay much lower taxes percentage wise and don't pay the high medicare/social security taxes, which are only calculated on *incomes* below approximately $100,000. Why do the outsiders, i.e. wage earners, support this system? Because they believe (through propaganda) that we live in a free, mobile society and if they work hard, they will also be rich and benefit from this system.

In the full information model, I agree that this would not be a stable system.