Stardate 20010520.1250 (On Screen): I just wanted to provide some closure on the story of Kaycee Nicole. The "mother" (Debbie) has confessed to fabricating parts of the thing, though she claims it was mostly based on reality. The man she took in claims that it was based on reality and that there really was a Kaycee (thought that was not her name) who really did suffer from leukemia and really did die of an aneurysm, but at this point we only have the word of the woman who confessed to fabricating part of the story that the rest of it is true. This whole thing just took a major leap downward on the credibility scale.

As with the guy on AnandTech's forum, I think this is simply a case where someone started an innocent fake and it got out of hand.

By the way, my conjecture about Audra Lea was incorrect. To that end, I am removing her pictures.

Stardate 20010520.1240 (Crew, this is the Captain): How do we really know who we're talking to online? I think we can't.

Update: The article "Trust but Verify" was written before I learned that the Kaycee story had been confessed to be a fraud.

Stardate 20010520.1150 (On Screen): I don't think it should have surprised anyone that the "network appliance" has been a commercial flop. For $600 you got a standalone device (a "thin client") which gave you web and email ability but little else. Why would anyone want such a thing? For only slightly more money you could get a complete PC, with a hard disk and a preloaded office suite and the ability to run (badly) PC software from the store.

The problem with an internet appliance as a concept is that it doesn't have a manufacturing cost advantage. While its hardware may be somewhat cheaper (no hard disk) it more than loses that back on economy of scale (when low end PCs are being made in fifty times the quantities). And the full PC has much more perceived value because of software, which costs nothing to manufacture. So it's always going to be perceived as providing more value for the money -- because it does.

There was, in some circles, much enthusiasm over the last four years for the entire concept of "thin clients". The theory was to make the client cheap and to centralize the compute power on big servers. (This was being pushed by, coincidentally, companies which made big servers. Imagine that.) But in fact it never made sense.

"Thin Clients" do make sense in a different area: portables. You don't want a big powerful PDA or cell phone because it will require a huge battery and weigh a lot. There and only there it makes sense to centralize the computing resource. But on a desk top, where power is available in unlimited quantities (except in California) and weight doesn't matter, distributed computing simply makes more sense. It's easier to scale and it costs less. (People muttering "Total Cost of Operation" have never made a convincing case that centralization actually does in the long run make things cheaper, and that particular TLA has largely dropped out of the lexicon.)

The long term trend in computing is decentralization and increasing local compute power. "Thin clients" on the desktop have always been a leap into the past. (I first got a local computer on my desk in the 8th year of my career. Believe me: we don't want to go back to centralized computing. We do not.)

Stardate 20010520.0635 (On Screen): Oh, dear. Hundreds of thousands of cute baby seals are going to starve because of a terrible change in the weather! Only it turns out that it isn't going to happen. The scientific agency making the announcement was in error. So why did they do it?

Hunting of young seals for their pelts is big business in that area, but the harvest has been reduced for conservation reasons. But if they're going to die anyway, you might as well relax the conservation limits and hunt them.

Oh.

Stardate 20010520.0630 (On Screen): I think that this means the Secure Digital Music Initiative is dead (and good riddance). Their words about a housekeeping meeting in September is a face-saving gesture, but what this really means is that they've given up. It's really rather amazing that it took this long; indeed, it's amazing that they were foolish enough to try in the first place.

Stardate 20010520.0620 (On Screen): I don't think there will ever be a precise defini