Google Base Was Sort of Live
Several people report Google Base (as predicted yesterday) went live, or at least, its login-screen. I can't reach it at the moment as it seems Google took it down again, but Dirson posted a screenshot to Flickr. So what is it? Quoting from Dirson's screenshot of the login screen:
Context & Ripple Effects
This is the accidental first sighting of Google Base, caught only because users reached a login screen Google hadn't meant to expose yet — evidence preserved by [[a:none|Dirson]]'s Flickr screenshot before the product vanished again. Two days later, John Battelle framed what Base really was: not an open 'freeweb' of structured data, but a Google-controlled vertical search play, in his analysis of Googlebase as structured/vertical search. The stakes became clear within a month when a glitch flooded the service with porn, forcing Google into cleanup mode on a product still finding its footing.
First-order effects
- Google has effectively confirmed a new structured-submission and listings product ahead of any official announcement — the leak itself is the news, giving competitors and would-be content partners their first concrete look at Google's ambitions beyond crawl-and-index.
- Sites whose business is listing structured information (classifieds, databases, vertical directories) now know Google intends to host that data natively rather than just link to it.
Second-order effects
- Content owners and vertical search players face a forced choice: feed their structured data directly into Google's database on Google's terms, or risk being disintermediated from their own query traffic once Base matures.
Third-order effects
- If the pattern holds, Google becomes the default registry for structured world-knowledge — but the longer arc shows these community-data bets are fragile: Google ultimately shut down its ambitious Freebase knowledge graph, migrating data to Wikidata in 2015, a caution for anyone building on proprietary Google data layers.
- The episode foreshadows today's structural tension over who owns the data layer beneath search — the same question resurfacing as Google experiments with AI-generated interfaces built on user context, like its Disco GenTabs experiment turning browsing history into apps.
The trend: Google's recurring attempt to move beyond indexing the open web toward owning structured, user- and partner-submitted data layers that sit underneath search itself.