Twitter takes a break, we're awake, and wondering...
What other basic form of communication goes down for 12 hours at a time?
Context & Ripple Effects
Dave Winer has been making the infrastructure argument all year: back in March 2007 he asked how basic Twitter really is, framing it as a primitive form of communication rather than a novelty. By December the usage data backed him up — he reported Twitter as a top referrer to his blog with over 2000 referrals in 30 days, alongside a visible rise in new users.
That framing is what makes this 12-hour outage more than a gripe post: if Twitter is basic communication, then downtime of this length is the equivalent of the phone network failing, not a website hiccup.
First-order effects
- Users lose their primary short-message channel for half a day, and for publishers like Winer who draw 2000+ monthly referrals from Twitter, the outage directly severs a top traffic source.
Second-order effects
- The third-party layer built on top of Twitter inherits the failure — tools like Twype.exe, which syncs tweets into Skype mood messages, go dark with it, turning Twitter's uptime into a dependency for services outside its control.
Third-order effects
- If a service positioned as basic communication keeps going down for hours at a stretch, reliability stops being a bonus feature and becomes table stakes — either the platform hardens its infrastructure or users reclassify it as an unreliable supplement rather than a primary channel.
The trend: As microblogging shifts from early-adopter toy to daily communication habit, outage tolerance shrinks and uptime becomes a competitive requirement for the platform.