ISINDEX

In our crusade to recreate the world's first browser the question of how search came up.

It's an interesting problem because the context for those days, 1989-1990, was that: the web had just been invented/proposed, the first http daemon had been written (and published) and the first browser had been written: WorldWideWeb.

For search to work, there would have to be an agreement between all three: the servers, HTML and the browser parsing the HTML.

So <ISINDEX> was used. This post looks at how it worked, where it faded out, and how, today, we actually have a very close descendant that is alive and well.

Notes

• Many indexes and mention of search in history of W3
• Summary of http 0.9 - agree how servers should handle isindex requests
• Early hint of dropping ISINDEX and introduction of INPUT
• HTML 2.0 Forms specification '94
• <queryform> successor to ISINDEX and precursor to FORM
• INPUT was pre-web?
• How did ISINDEX work?
• prompt and action properties
• Example on Lynx and WorldWideWeb
• Included in the announcement of the first draft of the HTML5 Parsing spec in 2006.
• Dropped in Chrome 35
• Could also be used to trigger XSS attacks

The rendered DOM in browsers resembled this - an injected form, two horizontal rules, label and an input:

Browser support

• Chrome 34: 2014-04-09
• Firefox 55: 2017-08-08
• Opera 21: 2014-05-06
• IE11: 2013-10-17
• Android Browser 4.4
• Lynx 2.8.8rel.2

To read

https://web.archive.org/web/19970120135440/http://www.eit.com/goodies/lists/www.lists/www-talk.1993q1/0079.html

Drafts may be incomplete or entirely abandoned, so please forgive me. If you find an issue with a draft, or would like to see me write about something specifically, please try raising an issue.