louisrosenfeld.com logotype

Home > Bloug Archive

May 10, 2005: Google Acquires nForm

Well, not exactly, at least not yet, but they ought to at least hire nForm to implement this great idea from nForm's Gene Smith:

Last year at work we were kicking around a product idea (codename: Shreveport) that would log users' outbound search terms as well as the results they select. And through some algorithmic mojo these things would create ad hoc communities of interest, where you could find people with similar interests based on what they searched for. It would also enhance recovery, findability and discoverability of information by "attaching" search terms to URL...

Read on; brilliant stuff.

email this entry

Comment: Prentiss Riddle (May 10, 2005)

Cool idea, but I do have two initial somewhat skeptical responses:

(1) Privacy. Danger, Will Robinson.

(2) "It also ties folksonomies seamlessly into the search process. Instead of picking up where search leaves off, they are one and the same." Not exactly. My interest and intention may change completely between the moment I enter the search terms and the moment I click through. I could search for cats and click through on dogs.

I'm not saying this idea couldn't be useful, but I'd see it as closer to, say, PageRank than to tagging, because it repurposes implicit metadata rather than the folksonomic approach of giving users control over the simplest possible *explicit* metadata.

Comment: vanderwal (May 10, 2005)

Nice idea. Something along these lines could be good, but it is largely an extension of Google's Zeitgeist.

When James posted his similar topic there is one important piece missing from this approach, feedback on the returned information being the proper information. The value from a folksonomy is extracted from a person adding a tag that means something to themself, which is the down fall of search and top-down approaches, they are not in the vocabulary of everybody trying to find the information.

I completely agree with Prentiss in that it is far closer to PageRank than anything useful that could come out of tagging. This approach is redundant on where search fails, where it is off mark, and where it needs improvement.

Comment: Gene (May 10, 2005)

A couple of quick responses:

Prentiss: "Privacy. Danger, Will Robinson."

Well put. :) We had thought of a couple of ways to allow private searches, while still getting the benefit of shared search history. Clearly people need to be able to opt out.

Prentiss: "I could search for cats and click through on dogs."

Thomas: "feedback on the returned information being the proper information"

Indeed, one of the problems with Shreveport is that search is often a hunt-and-poke activity. The challenge is trapping the good results, while letting the bad ones slip through the net (the hypothetical algorithmic mojo engine also handles this part).

Prentiss: "it repurposes implicit metadata rather than the folksonomic approach of giving users control over the simplest possible *explicit* metadata"

Well, yes, but that's the point. There is vastly more implicit metadata going out into the ether as query terms--from just about every Internet user--than there will ever be explicit metadata. And Shreveport is all about doing something interesting with *that*--exposing it, making it usable, making it personal/social, and using to enhance findability.

Still, the feedback's appreciated (it is, after all, only a crazy idea). :)

Comment: helge (May 10, 2005)

louis (and gene), i do share your enthusiasm about social search, however i fear the outlined concept would fail. i've explained on my blog why: http://www.helge.at/archives/00000218.php
in short: spam.

Comment: Lyle (May 18, 2005)

Thomas (vanderwal), I think you're missing some of the key pieces I'm reading into Gene's concept.

You call it "largely an extension of Google's Zeitgeist". Zeitgeist (http://www.google.com/press/zeitgeist.html) just shows popular search terms - allowing me to search for those (few) popular terms.

Using Gene's concept (vaporware), I envision being able to search for a bunch of related terms (e.g. UI, "user interface", GUI, usability, "information architecture", "user experience", design, AJAX), and, through his "mojo", being able to located (via tags) related terms/concepts/categories/whatever...BUT also locate *PEOPLE*...something Google doesn't allow me to do today. As we've learned from years of watching intranets develop (wherein the phonebook is always the "killer app"), enabling people to find people is a Powerful Thing(tm).

So often, information system designers try to take people out of the equation...to make the "user" self-sufficient, able to find the task-oriented equivalent to nirvana while only interacting with silcon-based servants. Sometimes, nirvana is found via carbon-based lifeforms, and computers can do a better job of helping us find them.

helge has some good points about spam, but I think it's worth trying anyway. I have faith that people can overcome those hurdles.

Time to get that mojo workin'.

Add a Comment:



URL (optional, but must include http://)

Required: Name, email, and comment.
Want to mention a linked URL? Include http:// before the address.
Want to include bold or italics? Sorry; just use *asterisks* instead.

DAYENU ); } else { // so comments are closed on this entry... print(<<< I_SAID_DAYENU
Comments are now closed for this entry.

Comment spam has forced me to close comment functionality for older entries. However, if you have something vital to add concerning this entry (or its associated comments), please email your sage insights to me (lou [at] louisrosenfeld dot com). I'll make sure your comments are added to the conversation. Sorry for the inconvenience.