ICSE 2006 paper accepted

Our paper about PACE got accepted into the main research track at ICSE 2006. Only 36 papers were accepted into the main track out of 395 submissions (9% acceptance rate). With perhaps three software architecture slots open (ICSE is a general software engineering conference, so the competition in any particular focus is brutal), this is very good news.

We now figure that our group is producing an ICSE paper every two years since 2000 (and many more before that too):

• Roy with REST in 2000
• Eric with xADL in 2002
• Rohit with DECENT/ARRESTED in 2004
• Girish and myself with PACE in 2006

You can read Dick Taylor's full DBLP entry for more. Nice company.

Combined with our recent IEEE Internet Computing PACE article, we should be able to produce some nice buzz about PACE.

Since I'm the webmaster for this conference and I'm under a grant, I was already planning to go. Unfortunately, Girish says he won't be able to make it to China due to visa issues; barring any change of heart on his part (he's the first author), this means that I'll be presenting the paper. Along the lines of his earlier comments, Girish's first remark after telling me to stop teasing him was along the lines of, "Darn. I can't make this paper a tech report now." Hah!

Regardless of how scary that presentation will be to give, this certainly made my week.

I'll likely have the full list of papers from the program chairs next week and will get them on the ICSE website shortly thereafter. I'll be curious to see who else has their papers accepted.

[/ucirvine/papers] permanent link

Ashamed by the UC Press's approach to scholarship

Two things came together this weekend:

• Saw the beginning of the debate between Alan Dershowitz and Noam Chomsky on BookTV on CSPAN2. I don't often find myself agreeing with Dershowitz, but I think he's exactly right on his views of Israel. Chomsky was just rambling on incoherently and the captioning wasn't working right on CSPAN, so I got bored and flipped back to football highlights.

• Watched Paper Clips, a phenomenal story about a group of middle-school students in Tennessee trying to visualize just how large six million is. I didn't know much about this project going into it, but about 15 minutes in, I went, "woah", and put away the laptop.

These two items reminded me of a link I saw a few months ago about how an "academic" (I use that term very loosely), Norman G. Finkelstein, who claims that the Holocaust is merely being exploited by American Jews for their own personal benefit. And, more recently, he published a new book called "Beyond Chutzpah: On The Misuse of Anti-Semitism and the Abuse of History", from the UC Press.

A sizable portion of his new book published under the UC Press imprint are scurrilous attacks on Alan Dershowitz. Dershowitz is more than capable of defending himself, so I won't go into it here.

Yet, what does this have to do with UC Press? It's this - and one of the points that I feel often gets lost in discussions of academic freedom: there is no obligation for a well-respected university to promote heresy or unfounded rantings under the cloak of so-called scholarship.

UC Press tries to shield themselves from criticism under the guise of academic freedom, and yet promotes the controversy in order to increase book sales. The University does indeed have an obligation to the public to present legitimate research. Yet, how is providing a platform for personal attacks research? Do personal feuds now override any requirement of legitimate scholarship? They are selling all of our souls for a hack who isn't worth the time of day.

The imprimatur of the University of California should have exceedingly high standards - one in which, this instance, a prima facie case for excellence can not be made. Let the tabloids or commercial publishing houses print it if they desire, but please don't force legitimate, honest scholarship to share any relationship with this drivel.

Campus Watch has it exactly right when they say:

But all this indicates a far more serious problem, the inability (or
unwillingness) to recognize bad or non-existent scholarship, and worse, to
defend it as merely "controversial." No one in this country is restricting
Norman Finkelstein's free speech, but many people exercise their own free
speech to criticize him, and to criticize the poor judgment shown by
institutions like Georgetown (and University of California Press) which give
him a podium. No one...has a right to speak at a university just because they
have a bunch of degrees. And in fact, no one has a right not to be disinvited.
Universities are supposed to be about judgment, not simply about controversy,
especially cheap controversy.

With their disregard towards what can be considered worthwhile academic research and scholarship, UC Press demonstrates little to merit a continued association with the people of California or the University of California.

[/ucirvine] permanent link

Risky or scared?

The UC Irvine librarian blog claimed that Google Print is risky echoing the sentiments of the Chronicle of Higher Education (paid sub. reqd).

The article states: "It pains me (author of article – j) to declare this: Google's Library Project is a risky deal for libraries, researchers, academics, and the public in general. However, it's actually not a bad deal for publishers and authors, despite their protestations."

As with a lot of things with Google Print, I think this is characterized by fear, uncertainty, and doubt (FUD).

Has Google and other commercial search engines hindered the spread of knowledge? I would find it very hard-pressed to argue that - I'd go so far as to claim it does the exact opposite. Endeavors like CiteSeer have done more to advance to academic knowledge than any innovations that the libraries have introduced in the last 20 years.

Through search engines, people are more aware of what other people are doing. This is unequivocally positive.

As Girish has been constantly pointing out to me, all we have to do is to publish an ISR tech report on the website, wait a few months, go to Google, and we'll find a number of unscrupulous characters plagiarizing our work. My claim is that Girish is just trying to make this argument so we don't have to submit to conferences. Yet, in all seriousness, there's a real point here.

Girish used to post a collection of the papers that 'stole' from his papers outside his office, but he's now given up. He was worried at first that he would be accused of plagiarizing from some undergrad in Russia. In all of the papers we've written together, we agonize and argue over phrasings of the silliest things for hours. We probably spend more time on one sentence than the plagiarizers do on the entire cut-and-paste and submission.

The key to all of this is in the oft-talked about 'long tail'. It's no longer about visibility for your papers online - if people are copying it, they can clearly find it. Having effective search engines is a positive thing for legitimate researchers and one of the most positive research impacts of the World Wide Web. It should work both ways: you encourage dissemination of your work and then learn about others work through the Internet. As Kuhn argues so well, most research is evolutionary not revolutionary.

The public libraries, like UC, are constantly under space pressure to remove books from circulation. How is limiting and removing books a good thing? How is restricting information good? I am of the definite opinion that libraries will only be helped in the long run by letting people know what they have. They are a public service with the express intention to share knowledge. How does opening the content and knowledge within those books go against this policy? It doesn't. Academics consume knowledge to synthesize new ideas. Searching will allow us to be able to identify sources far quicker.

The next argument is whether Google should be the steward. We live in a free-market society. Why can't Microsoft do this? Why can't Yahoo do this? Clearly they can and probably already have plans to do so. And, if they do it better than Google does, the marketplace will follow. Connecting people with the information that they want (or don't even realize that they actually want!) is the goal of any search engine.

Now, in this discussion, I am clearly sidestepping the right of the copyright holder in restricting the proliferation of the materials. But, doesn't the library operate in the same way that the search engines does: they acquire a physical copy of a copyrighted work and then allow anyone to view it in person. How fundamentally different is the concept of a library from a search engine? Is it just accuracy and relevancy? How is improving that a risky thing?

Or, is it just a fear of commercialism? If these search companies can make a buck off it selling ads, fine by me. All I want is the information I'm looking for. If the ads are somehow relevant to what I'm looking for, I'll go click on the ads. Let the marketplace decide - not some scared people who fear what they likely can't understand.

[/ucirvine] permanent link

What's in a title?

Speaking of UC lawyers fighting lawsuits, I had a conversation today with a colleague to change 'resource-sharing' to 'file-sharing' in the title of a paper we submitted today. The paper was aimed at an audience who probably wouldn't understand what 'resource' is, but would understand 'file' - so I thought it'd be best to change it.

Part of our research into decentralized architectures can (and does) apply to file-sharing applications. I don't view file-sharing as a terribly interesting decentralized domain, but it's one that connects with a lot of the audience. Therefore, we inevitably mention it in our papers. In this particular case, we had a group of undergraduates working on a project to create a file sharing application using PACE. It turned out rather well - all things considered.

What's interesting is that the campus forbids any use of file-sharing software automatically. All BitTorrent traffic is automatically blocked and if you do use it, your machine gets black-holed by the campus firewall. We had one of our servers in the office blocked because we downloaded Knoppix via BT. Lovely. As more and more legitimate uses are found for BT, I'm continually locked out due to the wide blanket NACS casts over the campus network.

We can write the software, write papers about the software, but we're forbidden to actually take advantage of it. Isn't the ivory tower nice?

[/ucirvine] permanent link

Go, go, go into the blinding light...

Now that I've been given permission, I can share that I've accepted an offer to go to Google (as an intern) for three months starting in January to work on Subversion. I can also share that my main goal while there at Google will be to integrate Subversion with Serf.

This isn't the first time I've been offered a position at Google, but, for a number of reasons, this offer was better for me both personally and professionally.

I'm excited because this will grant me the opportunity to work on this problem that I've been talking about for years - and talk about it publicly at the same time! It also helps that I have a lot of former and current colleagues there already (curse them and their boat load of options!).

I still need to find a place to live up there... Grr.

Plus, I've still got to advance to candidacy before I go, so there's a lot of work to be done before I take a hiatus from the ivory tower. (Actually, hiatus is too strong a word - there's a lot of overlap here with my eventual research plan.)

[/ucirvine] permanent link