Fortnightly Mailing

Anand Rajaraman on how Google measures search quality

Anand Rajaraman is another of those clued up computer scientists/entrepreneurs who takes the time to share their insights in a candid and accessible way. Today's How Google Measures Search Quality is a good example. Excerpt:

"Let me try to explain the latter point. There are two broad classes of queries search engines deal with:

• Navigational queries, where the user is looking for a specific uber-authoritative website. e.g., "stanford university". In such cases, the user can very quickly tell the best result from the others -- and it's usually the first result on major search engines.
• Informational queries, where the user has a broader topic. e.g.,
  "diabetes pregnancy". In this case, there is no single right answer.
  Suppose there's a really fantastic result on page 4, that provides
  better information any of the results on the first three pages. Most
  users will not even know this result exists! Therefore, their usage
  behavior does not actually provide the best feedback on the rankings."

Anarchy. The hidden cost of open access.

"Ah, sweet irony. If this article had undergone "peer review", or some other accuracy or quality checking criteria, then it would never had seen the light of day..."

writes John Kirriemuir amongst the many others deriding Philip Altbach's recent Hidden cost of open access in the 5 June 2008 Times Higher, for asserting this in particular:

"Profit, competition and excess have spawned the open-access movement. Academics, librarians and administrators think it is the answer to monopolistic journals. But there are several problems with it. Chief among them is that peer review is eliminated - all knowledge becomes equal. There is no quality control on the internet, and a Wikipedia article has the same value as an essay by a distinguished researcher. Open access may also offer greater benefit to those already at the top of the knowledge tree. A less well-known institution in a developing country would likely gain less attention than Harvard. While traditional journals also tend to privilege scholars working at top institutions, at least the peer-review system allows some opportunity for publication in recognised journals.

Essentially, open access means there is no objective way of measuring research quality. If the traditional journals and their peer-review systems are no longer operating, anarchy rules. Researchers will have no accurate way of assessing quality in a scholarly publication."

Learning and technology in China. Long piece from Donald Clark.

This long Learning and Technology in China post by Donald Clark, who has recently returned from time spent during May doing e-learning work for the Chinese Government through the World Bank, is worth reading. I bet Donald would welcome feedback from Chinese readers, and I'd happily pass these on to him.

XP on the OLPC. Proprietary software meets open source hardware.

Normally it is the other way round.

In the continuation post below is a 5 minute video of Microsoft's Bohdan Rakiborski explaining somewhat thinly how Microsoft made Windows XP and Office run on the OLPC. The video serves also as an impressive demonstration of the OLPC laptop itself. There is no mention of OLPC's mesh networking capability - key to its exploitation as an educational device - and which is presumably inoperative when the OLPC runs Windows XP; and of the suite of educational software that ships with OLPC there is no sign. Because it is gone.....

There is a telling look on Rakiborski's face when he says "It starts up about four times faster than the original operating system that ships on the XO laptop" (is this mainly a result of the addition of a 2 GB SD memory card?), but Rakiborski is pretty obviously taken with the underlying design of the laptop - low power consumption, sunlight readable screen, e-book mode etc*.

* At the 2006 World Economic Forum Bill Gates reportedly said, "If you are going to go have people share the computer, get a broadband connection and have somebody there who can help support the user. Geez, get a decent computer where you can actually read the text and you're not sitting there cranking the thing while you're trying to type."

Debategraph: the debate processor. Guest Contribution by David Price.

Question: What do you get if you cross a wiki, a forum, a blog, instant messaging, and social bookmarking with an argument map?

Answer: Debategraph.org

I started to collaborate on Debategraph with, my co-founder, Peter Baldwin in 2005. It’s one of those delightfully unlikely collaborations that the Internet makes possible: I’m based in Somerset in the UK and Peter is based in the Blue Mountains in Australia. We discovered each other, across the net, because we had two things in common: a shared perception of a problem and a shared idea for a solution.

When the cloud bursts - following an electrical fire in a data centre in Texas

Over the weekend a large data-centre in Texas suffered an electrical fault, resulting in a fire and in several thousand servers providing services to several thousand customers (including me, indirectly, as my Statcounter site statistics service is hosted there), being taken off line - with some data-loss. A fascinating chronology of how Planet, the company running the data-centre, handled the emergency, is (currently) visible on the company's bulletin board.

Ultra-mobile PCs in abundance - all are potential platforms for Sugar

Walter Bender writes in the 2/6/2008 Sugar Digest (see this earlier post for more on Sugar):

"The success of the OLPC XO and the ASUS Eee PC seems to have attracted the attention of the industry: it has been a busy week in the world of ultra-mobile PCs. Dell (see http://www.engadget.com/2008/05/29/dells-mini-inspiron-eee-pc-killer-revealed/), Acer (see
http://www.engadget.com/2008/05/29/first-pics-of-acers-aspire-one-eee-pc-twin/),  Wizbook (see http://www.theinquirer.net/gb/inquirer/news/2008/05/28/noon-eee-pc-wizbook-hits), Elonex (see http://www.desktoplinux.com/news/NS8294433279.html), and Kanguru (see http://pc.kanguru.pt/Home/) were all making headlines. Each of these machines represents yet another potential platform for running Sugar."

Machines with consciousness? A spectrum from Kurzweil (yes) to Penrose (no).

Will machines develop consciousness? Published by the IEEE in a special report on the "singularity" (a.k.a. the technological creation of smarter-than-human intelligence), this  handy and quite entertaining chart by Paul Wallich colour codes 15 thinkers on the subject (all men) according to whether they are true believers, deniers, or somewhere in between.

Scan and Release: Digitizing the Boston Public Library

St. Oswold's Church in Grasmere, England. Fore-edge painting on 1832 edition of Volume 4 of the Poetical Works of William Wordsworth. Digitised by Boston Public Library. Creative Commons License.

Enjoyable piece by David Weinberger about the process of digitising some (?) of the contents of Boston Public Library. Excerpt:

"Of this abundance, the digital group has so far scanned about 24,000 objects. When I point out to Maura Marx, the group’s head, that, given the library’s estimate that it has maybe 23 million objects, she’s looking at a 2,000 year project, she tells me that they’re just getting started. They’re going to bulk up, maybe do some offsite digitizing, and begin to make some serious progress. When I ask Thomas Blake, who does the actual digitizing, how he decides which stuff to do, he laughs a little and says, 'What I think is cool.'"

Richard Elliot's eLearn Watch

New Zealand's Richard Elliot produces eLearn Watch a sort of monthly NZ version of Fortnightly Mailing, with plenty of useful links and insights.