ResearchBuzz

I was wondering why it’d been quiet for so long. Gigablast ( http://www.gigablast.com ) announced a relaunch today with a couple of new features. It’s beta, natch.

The front page has been redesigned in a lovely orange and blue that reminds me of the University of Virginia. You can search Web pages, Images, Video, and a Directory. There’s also a new feature called Freshness Dating.

The Freshness Dating feature (which is turned off by default) allows you to search Gigablast only for pages that have been generated or updated in the last day, week, month, or year. How do it know? It uses what Gigablast describes as “patent-pending ‘freshness dating’ algorithms”.

I was lukewarm on this feature — you can get similar at several other search engines — until I saw that it has a custom option. In other words, you can enter a date span and Gigablast will search only for pages generated or last updated in that date span. And you don’t have to bother with Julian dates!

Very cool. I did a search for obama costume in the year 2006, and though it didn’t work perfectly — I found plenty of pages that were published after 2006 on the front page — it did cut down my search results from over 24,000 to about 1,400. (I’m sorry for using a political search, but it was really good at finding current Web pages.) Good deal. I now have a search engine to go to when I want to research something that has a lot of “search noise” due to a recent story, blog post, fad, etc. It’s not clear how far back the custom search goes, however.

What else is new? Gigablast is also offering a new “personal data policy” for privacy. All personal information is erased from Gigablast’s records within sixty days (every sixty days?) Gigablast also doesn’t use cookies or Web beacons/bugs/whatever.

Gigablast is no longer putting up its Web index size, but on its environmentally friendly page, it mentions that it uses less than 500 machines worldwide which are capable of maintaining “over 10B web pages”. The last number I remember Gigablast putting up was 10 billion, so around that would be my guess for its current index size.

I’m going to play with this freshness dating feature…

The New York Times covers the Oscars.

Amazon.com launches a currency converter.

Fly over 5200 golf courses at http://www.golfflyover.com/ .

Blinx hooks up with Encyclopaedia Britannica.

Pakistan knocks out YouTube.

Staples is offering free tax return copying on April 15.

Gutenberg-e (NOT Project Gutenberg) goes open access.

Recently at work we whiled away a coffee break discussing the controversy over writing famous people’s obituaries in advance. While we were having the discussion Anna Nicole Smith came up. Not her name; nobody could remember her name. We were totally stumped. We knew she had passed away shortly after her son, but that’s all we could remember.

Eventually someone went to Google and typed:

that woman who died after her son died

… and Anna Nicole Smith in a CNN story was the second result, and had her Wikipedia page on the first page of results.

I was rather surprised. After all, I do a lot of talking about how people should get as specific as possible. And while the query above presents a couple of facts, it was very generic.

A few days later we were discussing the NFL combine. (I promise, we do get more interesting in our discussions. We had a very spirited argument on whether hosting the Olympics would bring “real capitalism” to China. But nothing funny happened.) I knew there was some kind of intelligence test that NFL players have to take, but I couldn’t remember what it was called. So I told Google:

that test NFL players have to take

… and it pointed me, in the first result, to the Wonderlic test, which is what I was trying to remember.

I thought about it for a while and then went back to Google. “You’re fine at telling me what I can barely remember,” I said to it, “But how are you at naming what I don’t even know?” And with that I busted out a whole bunch of basic-level people searches, wondering which person Google would first place in each of my generic geists.

That man who went to France — Major Clarence Fahnestock
That lady on TV — Mary Birdsong
That group which couldn’t wait — The NPD Group
That guy who quotes Shakespeare — Julian Harris
That guy who played football — Guy Chamberlin
The lady who counted sheep — The Lady of the Lake, OR Elbert Hubbard. Your call.

Of course, you can get too meta for Google. I tried that guy who did this thing this one time and got MySpace Videos by, you guessed it, “The Guy Who Did That Thing That One Time”.

I think I will stick to very specific searching. On the other hand, this is a fun way to get completely 8-ball on Google and see what it identifies with a generic phrase.

Hey! Check Amazon for books on Search Engines.