It only takes a few morons to screw things up the world over

You imply that the people making these systems were aware of the rammifications of what they were doing, and decided to do it anyway. I fear it is worse still: they simply didn't think -- or even know to think -- about the rammifications of what they were planning on doing before doing it.

Doing a "point of attack" or "single point of failure" analysis of something appears to be a very rarely taught skill which is most unfortunate in many areas of life, not just technology. (Law being the other area obvious to me since I spent a while studying it -- lots of laws are passed apparently without anyone considering "if we pass this what will people do to (legally) get around it" leading to counterproductive results when passed. Sometimes they get it right the second time, sometimes they don't.)

As to how to solve the problem, I'm not sure. Making everyone as cynical as I am about such things has other disadvantages.

FWIW, as security risks go, I'd take "access content they haven't paid for" over "multi-year distributed denial of service attack" (email worms, viruses, spam, all largely (now) being launched from insecure end-user systems). It's at least contained, and has a built incentive for the people causing the problem to fix it.

Ewen

While it would be wise, I don't expect the people implementing these systems to necessarily think about possible adverse effects. What would be sensible is for the people who are going to be held responsible for the results -- the people hiring the people who implement these systems -- to think through the ramifications of the design.

Certainly, access to content someone hasn't paid for is not a big deal. Indeed, the only loser there is the hotel responsible for the system that allows it. Access to lists of who is in the hotel, what is on their bill and what they are doing online are all rather insidious. From a personal privacy and security point of view, they are worse than DDoS: They victims have no idea it is happening and no way to find out who is accessing the data.

But the people hiring the people who implement these systems don't know anything about "points of attack" or "failure modes". All they see is that they're getting the ticklist features that they want (viz, an automated hotel).

About the only way to change that would be to impose some fairly heavy penalties for leaking private information as part of a general set of privacy laws. Then those people (doing the hiring) would have an active reason to ask "this won't leak private information will it?". It's possible that New Zealand's Privacy Act might be usable for such a purpose (although IIRC there's not much in the way of penalties, at least being actively applied), but the US is notably lacking in such things.

Finally while I agree that leaking private information is worse than a DDoS (although after 5 years of DDoS I'm starting to think that N years of DDoS is equal to leaking private information), it's important to remember that along with the same holes that made this DDoS possible is that possibility of private information leaking (and indeed some of the worms did precisely that).

Ewen

I don't know, but page 2 of that Wired article had something that disagreed with Firefox.

Odd. It looks fine to me in Mozilla.

The first time I tried going to page 2 of the article, Firefox crashed. The second time it worked fine.