I pointed out in last week’s capstone post on the “search engine indexing of IDX series” that any policy that permits indexing of IDX sites needs to define exactly what kinds of ‘search engines’ are permitted and what search engine uses are acceptable.
In case you were wondering why I’m so concerned about that, I recommend you take a look at the Google maps real estate interface. It was written up in Inman last month. Our friend Matt Cohen noted in a blog post, referring to the Australian incarnation of the Google interface, that this interface makes Google look more like a ‘destination’ site and less like a ‘conduit’ site. In other words, the site is trying capture consumer attention with its content, not just let them search and move off to the source web site. Victor Lund also pointed me to an article about how the interface has been controversial with some data content providers in its Australian incarnation.
(BTW, I wanted to show screen shots with some narrative here, but my web layout/authoring skills suxxors, so instead I created a PDF that you can download if you want to see what the Google site looks like without taking the time to figure it out for yourself.)
So, my concern: for right now, the Google interface in the U.S. is showing only the listings of listing brokers who have uploaded them to Google. But imagine this scenario on the hypothetical site www.Scoobynoogle.com:
Scoobynoogle.com ‘spiders’ the web looking for real estate listings on brokerage (and other) web sites.
Scoobynoogle.com caches all the information and presents it in an interface basically identical to the Google maps/real estate interface, aggregating data from multiple IDX sites based upon the address or other semi-unique identifiers, so there is one “push-pin” per listing.
The site is designed to keep consumers browsing on it as long as possible. It wraps the listing content and site functions with advertisements.
Scoobynoogle.com shows links back to the originating broker IDX sites (perhaps many of them) on the tab where Google currently has a link back to the “original webpage.” The listing broker’s site would not be favored among these links (though it might appear there).
Such a use would be a ‘search engine.’ But its purpose is materially different than the Google that you use to search the web. This hypothetical site is intended as a ‘destination’ – a site where the consumer lingers and executes multiple searches, rather than a ‘conduit’ – a site that gets the consumer to a broker’s IDX site where she can continue her searching on the broker’s interface.
When folks are thinking about how to define what sort of indexing of IDX sites is permissible, and by whom, they need to think about a use like the hypothetical Scubynoogle.com site and decide (1) if they are happy about it and (2) if not, how to define ‘search engine’ uses to include only those they consider benign.
BTW, I have no evidence that Google is even considering such an approach. But someone is. Even if Google is intent on just displaying listings of brokers who have uploaded them, the brokers should be thinking about what use Google is putting their listings to – is it web indexing (a ‘conduit’ use) or Google trying to build its own traffic numbers (a ‘destination’ use)?
-Brian