My Internet Search ideas on 9/13/1994

Gil Press posted a short article on Forbes on 8/13/2017 that included this section:

August 15, 1994
Microsoft engineer Ben Slivka sends an email message suggesting to his colleagues that a World Wide Web browser should be bundled with the Windows 95 operating system. Released on August 16, 1995, Internet Explorer reached peak browser market share (95%) in 2003.

In trying to find that email, I found the following email with my ideas for an Internet search service.

In September, 1994, the fastest common dial-up modems were 14.4Kbps, Windows 95 was still 11 months away from commercial release, and the minimum hardware to run Windows 95 was an 80386DX chip running at 20MHz with 4Mb of RAM:

• Cairo was the code name for an ambitious Microsoft project to implement Bill Gates’ Information at Your Fingertips vision, which he presented at COMDEX on 11/12/1990.
• “content indexing” was the term used by the Cairo project to describe the process of analyzing a (potentially very large) set of files and producing an index that would allow the Cairo OS to instantaneously answer any query about the files.
  Many companies (Excite, Yahoo, AltaVista, Lycos, Google, …) arose starting as early as 1993 to offer such a service for the entire Internet.

—–Original Message—–
From: bens
Sent: Tuesday, September 13, 1994 10:15 AM
To: stevem; markz
Cc: johnlu
Subject: cairo content indexing and search UI

Who is the best person to talk to about this stuff?

I’m working for JohnLu on Internet access, and the problem I’ve bitten off is to (eventually) content index the entire Internet and provide a flexible UI for searching this mass of data to hone in on interesting stuff.

I don’t want to reinvent the wheel, so I definitely want to learn about Cairo (and experiment with it).

My roadmap looks something like this:

Stage 1:
Basic search UI
A content index of an interesting (and small) subset of the internet
(stored at microsoft.com)

Stage 2:
Fancier search UI (wizards?)
A bigger subset of the Internet content-indexed
Distributed content indexes from third party providers

Stage 3:
Add natural language to search UI
Distributed content indexes of entire Internet

thanks, bens