Ben Langhinrichs

Photograph of Ben Langhinrichs

E-mail address - Ben Langhinrichs

Recent posts

Mon 26 Oct 2020

Data from Domino 1: Rows ≠ Documents

Wed 7 Oct 2020

Custom Domino Extensions in a Modern AppDev World

Tue 22 Sep 2020

Exciton Boost - Barbarians are at the gate

October, 2020
    01 02 03
04 05 06 07 08 09 10
11 12 13 14 15 16 17
18 19 20 21 22 23 24
25 26 27 28 29 30 31

Search the weblog

Genii Weblog

Better search engines

Mon 5 Jan 2004, 02:12 PM

by Ben Langhinrichs
There is an interesting article on next generation search engines on the CNN site today.  I tried out to see how it compared to Google, and I must say I like the results.  

Inline GIF image

If I search for "rich text" on Google, I get 6,910,000 results!  Nearly seven million hits, which means that the vast majority is useless to me and essentially inaccessible in ten page increments.  On the first page, I get hits as varied as rich text editors, Microsoft RTF specifications, support technotes from MSDN, and a blog about rich text editing in MovableType.  If the first page has that level of variety in just ten entries, how many pages might I have to search before I found what I wanted?  (It is also the second page before Genii Software is mentioned, which also irks me, <grin>)

If I try the same search on Vivisimo, I get tabulated results that are much more useful.  On the left hand side, I get categories with counts such as Text Editor (44), Rich Text Format (42), Class (5), and even Genii Software (3).  Besides the preferable placement of Genii Software among the obvious categories, it is infinitely easier to wade through the results.  Some of those rich text editor entries were more than fifteen pages away on Google!

Trying another search that is near and dear to my heart (feel free to try your own), if I search for "Notes web coexistence" on Google, the first several entries do reference my blog entries or related entries, but the same page also offers a Byte article from 1996.  Vivisimo shows the categories, and has useful collections such as Microsoft, Exchange (22), Coexistence and Migration (17) and others that seemed intuitive and useful.  Of course, my blog entries were still first, but those looking for conference info would have found Session (6) useful, with links to various conferences where this is discussed.

I will certainly keep this in mind.  Let me know if your mileage varies.

Copyright 2004 Genii Software Ltd.

What has been said:

87.1. Stan Rogers
(01/05/2004 02:10 PM)

I don't know if this is any indication of how many people are reading your blog, Ben, but I just get repeated "under heavy load, try again some other day" messages. That's something I haven't seen at Google since forever. I must have wasted nearly a full minute, there. Really. It needs to get a whole lot bigger 'n' faster before it's ready for prime time. The concept sounds good, though.

87.2. Tim Latta
(01/05/2004 02:55 PM)

There was a search engine a number of years ago by Northern Light (out of Cambridge, I just learned) that did very similar results clustering. It WAS a free, consumer search, but it seems the model didn't pan out and now it is a corporate, installed application. I've missed that UI. Thanks for letting us know.

87.3. Ben Langhinrichs
(01/05/2004 03:35 PM)

LOL. Somehow I think the fct that they were reviewed today on CNN might have had a bit more to do with the load. It is true that they would need to ramp up considerably to be able to handle what Google does, but I do like the concept.