Managing your shared items
We've gotten a lot of helpful feedback about our new sharing feature. We'd hoped that making it easier to share with the people you chat with often would be useful and interesting, but we underestimated the number of users who were using the Share button to send stories to a limited number of people.
Context & Ripple Effects
This is a course-correction, not a launch. In September 2007 Google rolled out Shared Stuff, and Reader then layered on a sharing feature that automatically surfaced a user's shared items to the people they chat with most often. The assumption was convenience; the feedback, per Google's own post, was that many readers treated the Share button as a private send-to-a-few tool, so items they thought were pointed messages became broadcast to their whole chat circle.
The story traveled fast and critically: on or about December 27 alone it drew pickups from Bits ('Google Thinks It Knows Your Friends'), Fh on earth ('Google Reader shares private data, ruins Christmas'), and mathewingram.com/work, which pushed back that the outrage was overblown. That split — privacy-by-default versus social-graph-by-inference — is exactly what the 'managing your shared items' post is trying to defuse.
First-order effects
- Users who relied on Share for limited distribution have had their audience redefined without consent, and Google now has to ship management controls to let them restrict who sees shared items.
- Google takes a reputational hit at the exact moment it is positioning Reader as a social product, with mainstream outlets framing the feature as Google deciding who your friends are.
Second-order effects
- Rival feed readers and sharing tools gain an opening to market explicit, user-chosen audiences as the differentiator against Google's inferred-contact model.
- Google's broader social push — the same movement behind Shared Stuff — now carries a documented example of defaults backfiring, which raises the bar for how its next sharing features get introduced.
Third-order effects
- If the pattern holds, platform-defined social graphs built from behavioral signals like chat frequency will keep colliding with user intent, forcing the industry toward visible, editable audience controls as table stakes for any sharing feature.
The trend: Feed readers are turning into social sharing layers, and the fight over who defines your audience — the platform's inference or the user's intent — is becoming the central design question.