Perhaps just now it’s unfashionable to grumble about patents, given that Sun and IBM are showering the world with bucketfuls of ’em (and a fine thing, too), but if you want a loud, ugly, wake-up call, start at VarChars, which is offered by someone who apparently wishes to be known only as “rayg”. Samples: IBM Receives Patent for Using Regular Expressions to Extract Information from Documents, Google Receives Patent for Highlighting of Search Results, and Microsoft Attempts to Patent Object Persistence. Read on for some predictably-negative commentary and—gasp!—a look inside some incredibly-advanced Google technology.

First the conclusion: I’ve already written that the idea of a patent on software is not inherently broken, but rayg’s site is still more evidence that the system in use in the U.S. for granting them is, perhaps beyond repair.

Enough generalities, let’s look at specifics; specifically, U.S. Patent 6,839,702, granted to Google on January 4, 2005, for highlighting matches to search terms in retrieved documents. The idea is nicely summarized by this figure from the application:

US Patent 6,839,702, Figure 6

Gosh, I seem to recall having this same conversation in 1987, back working on the search engine at the New Oxford English Dictionary project. “So you see where we highlight the match in the result summary?” “Yeah.” “Well, it would be nice if we could do that in the actual doc too when they fetch it.” “I suppose.” “Not hard to do, I wouldn’t think?” “Nah, a bit tedious, but we have all the info.”

I’m going to try to put this as moderately as I can... THIS IS NOT AN INVENTION! THIS IS A MINOR PRODUCT FEATURE THAT ANY HALF-ASS COMPETENT SOFTWARE ENGINEER COULD IMPLEMENT WITHOUT MENTAL STRESS! THIS PATENT IS COMPLETE UNADULTERATED HORSE-SHIT! Yep, that was as moderate as I could be.

To give an idea of the whimsical fluffery that constitutes the U.S. Patent process, I’ve enclosed a couple of more figures from that Google patent. The “Systems and methods for highlighting search results” are designed to work on something called a “computer”, illustrated in the figure below.

US Patent 6,839,702, Figure 5

Well, to be fair, they do go a little deeper than that, they explain pictorially how things are organized inside that computer.

US Patent 6,839,702, Figure 3

Gotta love that “miscellaneous data”. Some of mine is very miscellaneous.

I’ll close with yet another tip o’ the hat to our colleagues from Poland who are trying to keep Europe from heading down the same Trail of Tears.


author · Dad · software · colophon · rights

January 26, 2005
· Technology (77 fragments)
· · Intellectual Property (10 more)
· Business (106 fragments)
· · Intellectual Property (48 more)

By .

I am an employee
of Amazon.com, but
the opinions expressed here
are my own, and no other party
necessarily agrees with them.

A full disclosure of my
professional interests is
on the author page.