Seth Dillingham: "I feel as if Macrobyte was hit by a truck last Thursday, and was in a coma until early this morning. One or two brain cells are finally firing again, but what a nightmarish week it's been." [Scripting News]

Err, I hate to be harsh, but...  I read with sympathy Seth's tale of woe -- until I realized Macrobyte is a hosting provider/application service provider with one Internet connection.  Apparently none of Macrobyte's customers (who were right to be outraged) did their homework, because multiply-redundant Internet connections from multiple vendors is the first thing on your checklist when considering outsouring to an ASP.  What, yer gonna rely on luck and wishful thinking?

Here's a little hint to the Macrobyte guys: you also need to rotate your tape backups offsite, create a disaster recovery plan, deploy redundant switches and NICs, and cluster your servers for high availability.  And all of this must be spelled out in the service level agreement with your customers, so they have some legal recourse when your outage makes them lose money.

One T1?  C'mon!

9:07:19 PM   

Doc Searls: Infrastructure Infrastructure Infrastructure Infrastructure! [Hack the Planet]

Fantastic slides from Doc Searls' talk at OSCon (I wish I'd been there).  Check it out -- he provides insight into why Hollywood is doing what it's doing.  He contrasts Hollywood's view of the relationship between commerce and infrastructure with that of geeks.

If the Entertainment Oligopoly's motives and behavior can be understood (beyond generalizations about greed and inertia), there's hope geeks can formulate a productive response.

9:58:30 AM