Comments on Chapter 1

June 24th, 2009 by marti | 2 Comments »

Comments on and discussion of Chapter 1 should be placed here.


  1. I have a minor point, which to me illustrates the way that Google is the supernova, blotting out the other search engines.

    You say: “Sometime around 1997, Google introduced conjunctive (AND-based) query analysis for Web search ranking, meaning that every word in the query must be present in the document in order for that document to be shown. ”

    I’m pretty sure it was HotBot (Inktomi) that pioneered the default AND. I have vivid memories of that because I was surprised by liking it, and often used it by preference for the default to matching all words.

    I’ve been working on an article about search engine internet operators (+-“”) so have been looking back a lot lately, and trying to find out where they came from and when, so I have the materials pretty close to hand.

    the SearchEngineShowdown review of HotBot is undated but says that it does default AND, The SearchEngineShowdown Google review has it also doing default AND, and mentions that it was in alpha test until February 1999 (confirmed by the Google History page)

    The book, “Search Engines for the World Wide Web,” by Alfred and Emily Glossbrenner, Peachpit Press (1999), describes HotBot as having a default AND. It doesn’t even mention Google. (I happen to own it).

    I also think that Google did not invent their snippets and match term highlighting, and I wish they would give proper credit to Hans Peter Luhn for it.

    sorry I’m being a grumpy librarian here.

  2. Marti says:

    Avi, thanks for that comment. Great Hotbot reference. I still have time to make that correction in the hardcopy (proofs due this weekend) and will do so.

    However, I don’t believe I said that Google invented snippets; I have some very old citations for KWIC in section 5.2 including a cite to Luhn.

    Rather, I believe they were the first to used kwic snippets in web search. If I’m wrong I’d love to know it.

    As a meta-point, I actually tried to keep Google from dominating the book as much as possible, for reasons I stated in an earlier comment. But the fact of the matter is they often adopted interface ideas that stayed around and were copied by other major search engines. Some examples of ideas described in chapter 1 have been adopted by other engines since the time of writing, but they didn’t do it in time for me to adopt them as an example.

    Finally, there were a few cases in which I wanted to use a different engine as an example of more recent developments but the owners of that engine would not allow me to use screenshots in the book, so those are not illustrated as they would otherwise be.

Leave a Reply