Posts

Data Privacy and its Consequences

This opinion piece appeared in the EPFL Forum Magazine (p. 22), but since most people will never see it there, I thought I'd post it. [I hope the student editors of the magazine read Martin Vetterli's interview on open science and apply the same spirit to this reposting.]



Data Privacy and its Consequences 

Prof. James Larus
School of Computer and Communications Sciences, EPFL

Public perception is slowly changing in regards to data privacy. From fighting terrorism to getting targeted advertisement instead of random ads, many people feel as if giving up a bit of privacy may not be such a bad thing. But is it? James Larus, Professor and Dean of the School of Computer and Communication Sciences at EPFL, sheds light on this complex issue. 

What is your privacy worth? Probably more than you expect.

Most people unknowingly trade away detailed, specific personal information for a “handful of trinkets” and, in doing so, have made companies like Google, Facebook, Baidu, and Tencent immens…

Congratulations Dave and John!

Image
Dave Patterson and John Hennessey won the ACM's Turing Award this year for their outstanding research contributions to RISC computers and a methodology for quantitative computer architecture. A video of their acceptance lecture is online.

This is a well-deserved award to two great researchers and teachers who have had a profound influence on industry and generations of students. I've known Dave since I arrived at UC Berkeley in 1981, and I want to thank Dave for his profound influence on my career.

In fact, if it wasn't for him, I would be doing something else, in some other field. When I arrived at UCB in 1981, I loved computing but wasn't too excited about computer science. The undergraduate computer science program at Harvard (which was then hidden in the Applied Math department) was theoretical and rather dry. When I was accepted to graduate school at UCB, I wasn't sure I wanted a PhD, but I was sure I wanted to live in California, so I accepted their offer wit…

Gender Diversity and CS Conferences, cont'd

Addendum (Nov. 5, 2017):

I brought up the idea from the previous post at a couple of larges CS conferences (OOPSLA and SOSP), and even male faculty members saw the value of it. Faculty with small children report they travel like I used to, by making the most minimal trips possible (typically only spending 2 days at a conference with the tightest possible flight connections). No one particularly enjoyed this, but it is widely seen as necessary to balance professional advancement and family responsibilities.

In addition, I was approached by Philippe Vollichard, the person at EPFL responsible for sustainability, who saw this blog post and pointed me to an ETHZ initiative to reduce university travel because of its large carbon footprint. EPFL has a similar report (which doesn't appear to be online??), with the conclusion that the carbon footprint of airplane travel at EPFL is as large as the daily commute to and from campus and larger than energy consumption on campus.

I'm…

Gender Diversity and CS Conferences

Last week in Barcelona, I had an interesting conversation with Koen De Bosschere, the editor of ACM Transactions on Architecture, Compilers, and Optimization (TACO), about the computer science publication process. His contention is that the CS community's publication process, which treats conferences as the primary and certainly the most prestigious publication venues, is a major cause of gender inequality in the CS academic and research community. Because women are primarily responsible for child care, they quickly realize that regular travel to conferences is incompatible with a family but is necessary for career advancement. Hence many women leave the research field. In addition, conferences put researchers at non-US schools at a disadvantage because of the cost in money and time of attending conference that primarily occur in the US, and because of the airplane travel, is not sustainable.

Koen argued for making fast-turnaround journal publication (as he is doing with TACO) the…

A Letter to the Economist

The April 8, 2017 edition of the Economist had two articles whose juxtaposition amused me. The first was an excellent story “Computers security is broken from top to bottom” and the second “How hospitals could be rebuilt, better than before” described the virtues of increased computer usage in hospitals. I sent the Economist the following letter:
F. Scott Fitzgerald wrote that a first rate intelligence is the ability to hold two opposing ideas in the mind at the same time and still retain the ability to function. I must commend the Economist for publishing “How hospitals could be rebuilt, better than before” and “Computer security is broken from top to bottom” in the same newspaper. More computers in healthcare; what could go wrong? (see article 2)
More seriously, several crucial segments of the world economy — finance, communication, and transportation — can no longer function without computers. In a few years, other important industries automobile and healthcare most prominently will …

Developers Should Write Papers

Many – but not all –software developers would probably rather learn COBOL than capture and analyze in a research-type paper the systems they have constructed. But, contrary to popular belief, research papers are one of the best way to capture the important aspects of a system in a concise and usable form. All developers should write (or be made to write) a paper when they do something that is worth sharing with others. [A paper does not need to be an academic research paper. It could be a detailed blog post like Joe Duffy's Blog, which contains the equivalent information in a less formal format.] It cannot be a system architecture description – hundreds of pages of detailed, hierarchical description that no one can (or does) read.
Let me tell you a cautionary tale how Microsoft threw away $100 million, with little or no return, because developers said, “writing papers is not part of my job”. In Microsoft Research, Galen Hunt and I ran the Singularity project to write a modern oper…

Jim Gray's Advice for Authors of Rejected Conference Papers

Jim Gray was a pioneer in database systems and Turing award winner, but he was also a truly nice guy who went out of his way to support other researchers. When I was working at Microsoft Research, I wrote him email complaining about a paper being rejected, and got this reply from him:
From: Jim Gray
Sent: Tuesday, July 04, 2000 1:51 AM
To: Jim Larus
Cc: Michael Parkes Subject:
RE: Visit?

well, the <omitted> paper is in good company (and for the same reason).
The B-tree paper was rejected at first.
The Transaction paper was rejected at first.
The data cube paper was rejected at first.
The five minute rule paper was rejected at first.
But linear extensions of previous work get accepted.
So, resubmit! PLEASE!!!

We did, and the paper was published. But, I value Jim's response more highly than the paper, and since then I have added to his list. My favorite is that Tim Berners-Lee's paper on the World-Wide Web was rejected as a full paper at Hypermedia '91 and presented as a poster.…

Rebooting Computer Security

The NY Times asked the wrong question about the Obama administration’s response to Russian hacking of the November US election ("U.S. Reacting at Analog Pace to a Rising Digital Risk, Hacking Report Shows"). The question is not why did it took 16 months to develop a response, but what could the US have done to prevent it? The disturbing answer is nothing.
Computers are fundamentally insecure, and this sad situation is not going to change quickly. As someone who has spent his entire career computer science research, it pains me greatly to admit that Donald Trump is right when he told reporters “You know, if you have something really important, write itout and have it delivered by courier, the old-fashioned way. Because I'll tell you what: No computer is safe." 
Computers’ original sin is that they run software that is written by humans. People make mistakes at a predictable rate – roughly 10-20 defects per thousand lines of code. Testing can find and eliminate many …