Typepad goes titsup
Dilbert falls silent, thousands of posts missing — Users of Six Apart's TypePad platform have suffered their worst outage yet, with bloggers left stranded since the early hours. The past week's blogging has also disappeared. — In true "Web 2.0" fashion …
Context & Ripple Effects
This caps a bruising autumn for Six Apart: TypePad was blocked in China in late October, the company itself mused about the ups and downs of running a service growing this fast, and a December 16 status update on ongoing TypePad issues preceded the worst outage yet by two days. The Register reports a week of posts missing and flagship blogs like Dilbert gone silent.
The story has already escaped blog-land — Computerworld frames it as a 'storage glitch' crippling the provider, and inessential.com offers 'moral support', signalling sympathy rather than surprise. The same day, Dave Winer argues for interoperability between blogging tools, giving the outage an immediate structural frame: single-host lock-in means users absorb every failure.
First-order effects
- Thousands of paying TypePad bloggers have no site and a vanished week of archives, with the confirmed cause being services moved between data centers — recovery, not features, is now Six Apart's entire agenda.
- Sympathetic voices like inessential.com are spending goodwill publicly, so each additional hour down converts supporters into critics.
Second-order effects
- Competing hosted platforms gain a reliability talking point at exactly the moment TypePad customers are angriest, turning a capacity failure into a switching decision at renewal time.
- Winer's interop push picks up concrete ammunition: users whose content lives on one host bear all the downtime, so export paths and tool portability become buying criteria rather than developer niceties.
Third-order effects
- If fast-growing hosted services keep failing at the storage and migration layer, uptime track records and data portability will harden into standard procurement questions for Web 2.0 services — mainstream outlets like Computerworld covering a blog outage as an enterprise-style incident is itself evidence of that shift.
The trend: Hosted web services are scaling past their data-center capacity faster than their reputations can absorb, making reliability rather than feature sets the decisive factor for paying users.