Thunderstone Software: First to spider JavaScript hyperlinks

Ohio firm's Texis search software makes history?

[Winooski]

[library.northernlight.com ]

CLEVELAND, Jul 30, 2002 (BUSINESS WIRE) -- The industry's first search product to index JavaScript hyperlinks and JavaScript dynamic content was announced by Thunderstone Software. The new capability is bundled with Thunderstone's Texis search software, and is available immediately.

So is it true? Is Texis the first spider to be able to travel through and index JavaScript?

[Knowles]

That would be a nice advancement I am sure but I am not sure how useful it will be. It will allow people to use more dynamic JS for menus and such and still be indexed which only reduses usablity for people with disabilities. Also there is currently no advanctage to stuffing keywords in JS would this be something that suddenly becomes possible?

[ScottM]

Well there goes my "anti-spam javascript e-mail addresses"!

[grnidone]

From: [thunderstone.com...]

Our customers use Texis in a wide range of applications such as: auctions & classifieds, automated categorization, litigation support, competitive intelligence, help desk, document & multimedia management, Internet publishing, real-time message handling, and web content searching.

As I understand it, Texis by Thunderstone is geared to intranets, isn't it? Is there a major search engine out there that uses Texis as the search mechanism?

[mbauser2]

No major sites, but it does show up in some meta-search engines like Search.com (fed from the Texis demo site, the the Thunderstone Web Catalog [search.thunderstone.com]) and Dogpile (as the Dogpile Web Catalog [dpcatalog.dogpile.com]).

[GoogleGuy]

I'm surprised people haven't mentioned Texis much around here. Texis is geared more for intranets though, or at least that's my perception. Their documentation lets you write some simple scripts to do interesting things, like access SQL databases. It's almost along the lines of PHP or Python.

[Josk]

spidering javascript is not *that* hard... just parse the pages, including the links, parsing javascript as needed. there are well known, pre-written j-script engines that can be used. the only tricky thing is to simulate a page coming from a user with data that can be parsed by javascript sub to give a valid page.

The only real reason not to parse j-script is processing speed, and that any web site on the net using j-script for its links is either commercial, or badly designed, or both.