googlism

[Glenn Scheper]

> http://www.googlism.com/index.htm?ism=Thomas+Pynchon&type=1

Thanks Jasper!

Exciting resource. I forwarded that URL to everyone at work,
leaving Thomas Pynchon as the query. Eyebrows will raise.

But you know, in my infinite patience, I am also working on
such a program idea. It would make no assumptions, but try
IP addresses at random to find hosts, then follow all links
to find HTML forms, then make submissions at random to find
search engines, and note which page scraping heuristics are
good for which engines, and keep performance metrics. When
a battery of such facilities are established, the user can
query, and results would have sentence boundary detection,
and eliminate all non-unique ideas, and present best ideas
first. Ranking would consider vocabulary and structure and
past queries. The user clicking on emoticons would affect
element ranking in any of these layers. KeyWord-In-Context
is the best way to display results, and clicking on a KWIC
result would lead user to read an instance of that context.

But I'm crucified on the bristling armament of the Windows
programming API: document-view architecture, and tree-view
controls and object-orientation generally, because I want
to make a pleasant program for the public, not for hackers
like me.

Yours truly,
Glenn Scheper
http://home.earthlink.net/~glenn_scheper/
glenn_scheper + at + earthlink.net
Copyleft(!) Forward freely.