IANAL. IANA American. However, my understanding is that: copyrights are decided in federal courts. This decision does not establish any precedent, although of course people in other trials may reference it as a useful example of something clearly argued. If appealed, it goes to the 9th circuit, which covers a few states including California. If that circuit makes a firm ruling on the copyrightability of the SSO of APIs, then that is precedent-setting for all lower courts within the 9th circuit area. Courts in other areas are not bound by it, but may of view it as a useful example. If further appealed to SCOTUS, then any decision there binds the entire country.

See Talk origins; in particular 29+ Evidences for Macroevolution

Did you mention hydrogen bombs to him?

and similarly, if the jury decide that that Google *did* copy the API (i.e. it wasn't fair use etc), then if after appeal it's decided that APIs are protectable, it avoids having to send the case back with a new jury to decide whether Google violated that copyright.

Its not a leak, Someone has already commented in the ticket that if you repeatedly create and destroy perl interpeters, then you need to set PL_destruct_level, because otherwise, (for efficiency), perl doesn't completely free the old interpreter, on the asumption that you're about to call exit(). So, it's just that no one got round to marking the ticket as rejected.

No, the "ALTER DATABASE BEGIN BACKUP" was just how the issue was first discovered. The issue is that someone with *low* privileges on an obscure, minor DB server can bump the SCN on that server, which if it happens to be linked to any other servers (like your big important one), causes the SCN to get bumped on those servers. So you can DoS all the other servers.

Well, the A380 was the first civil airliner to really use composites. It has >20%, whereas prior to that it was 10%. See "Advanced materials" on the A380 wikipedia page.

In case anyone was thinking that £12B is a lot for a database, what that money *actually* represents is providing a large chunk of the NHS's IT infrastructure for the next 12 years; or £1B per year for the IT needs of an organisation with 1.4 million employees and an annual turnover of £100B. The central data spine (which is the bit suffering the biggest problems and delays) is only one aspect of the system: it includes everything from making X-rays digitally available, to providing the network connectivity to individual GPs. Much of this is already in place and happily working.

Because unless you know what fried the primary controller, there's a good chance you'll just fry the back up controller too when turning it on.

Well, the Colossus wasn't "based on bomba technology", and was never used to crack Engima.

Say a cup of water is 0.25L, and its temperature is being raised from 20C to 100C. That requires 4200 x 0.25 x 80 = 84 kJ

Now lets be really pessimistic on the Google front. Suppose my search takes Google 1 second, and the search is distributed over ten 500W servers. That's 5 kJ expended. Lets double that to allow for the costs of spidering and indexing, and double again since the article mentions two searches per cup. Thats 20 kJ. Assume I spend a minute on my 30W laptop viewing the search results; thats another 2 kJ.

So We have 84 kJ verses 22 kJ.

That should be $3.5M. A lot cheaper than students.