Dear David
I just tried the version you uploaded, very nippy, and personally I don't
think either a 1MB index or using javascript are problems. The ability to
search the docs far outweighs the hassle of upgrading the browser if
necessary.
It was unclear what the ranking scheme refers to: "hit relevance" or
"number of hits". Clearly relevance would be preferred. If the rank is the
hits per page/document I think it should be clear that this is the case.
Lastly, if the search phrase is the name of a module, returning that
modules' page first would be nice.
Peter
At 08:35 13/02/2002 -0700, you wrote:
I have developed a search engine for the HTML documentation that doesn't
require a server. This would allow a user to search the documentation on
their host machine without having to have a network connection or worry
about a server being on-line. It requires a fairly large index file (~
1MB) and a Javascript enabled browser. I'd appreciate people testing it
and giving me their opinions. To be warned, when you open the page, please
wait until the search entry box appears--it will take some time to
download the index file (this would not be a problem with a local index
file). For those that want to complain that this is written in Javascript,
I only want to hear from you if you have a better idea that can support
client side searching. You will not be able to click on the results as
this is not in a valid dx/html directory--but you get the idea.
You can reach this page by selecting Search from New Design - 2 at
<http://www.opendx.org/variations>.
--
.............................................................................
David L. Thompson Visualization and Imagery Solutions, Inc.
mailto:[EMAIL PROTECTED] 5515 Skyway Drive, Missoula, MT 59804
Phone : (406)257-8530
