Over 3" is legal if you have a reason to carry them in public. Under 3" folding knife, you don't need to have any reason. De jure there is an additional restriction: you can't have a locking blade. That was not the intent of Parliament, and was established in a court case.

Except for Kelvin, of course, who was Second Wrangler (https://en.wikipedia.org/wiki/Wrangler_(University_of_Cambridge)) in 1845.

You need to be specific about what country you are in. Most don't have such a rule. In any case, if I were to do this, I'd probably wire it to 12V and put in my own control system. It's not rocket science.

Since TFA has a load of speculation such as "any iPhone can ... use Globalstar" without any technical justification, there doesn't seem much need to take Cringely seriously when he talks of Globalstar.

Still going, and you can get a Debian distro. They've got rid of that 2 Gb limit on disk (originally it was linked to the addressable RAM), but it's still 32 bit and things like USB don't seem to be there yet. Most people seem to be running it in emulation rather than on the metal..

We have several security measures, so it is not possible to make a good forgery. However a significant proportion of our £1 coins are conspicuous forgeries and still circulate, so a forgery doesn't need to be good to be viable. I'm pretty sure that starch-free paper with the same handling feel would be enough to pass muster for many purposes, if you could get a photocopier to work with them.

We already have high resolution colour photocopiers, but there is no great problem with them being used to duplicate paper currency, partly because of the EURion constellation printed on the originals. This does rely on the manufacture of the photocopiers being "trusted".

And that terminal velocity is very low for shotgun pellets. You can calculate it (I assume you have physics textbooks handy?), or you can take my word for it as I've been under a light shower of falling pellets - which bounced off like very light rain.

Interesting - thought we'd got past this stage. Still, I found myself working on some K&R C a couple of nights ago.

Could you give a bit more detail on the difficulties? This seems a bit surprising.

Again: I am unaware of any auditing requirements. What auditing do you believe takes place, who is placing the requirements, and what is your source for this information?

In respect of dual signature, the key word is "green" - this would be appropriate for validated domains such as banks, not necessarily for all hosts.

An advantage of a WoT model is that it is possible to give partial trust to different signers, and set a policy to trust a site once there are enough partially trusted supporters for it. This means that the system need not be fragile to a lapse in a single signer. At base though, you can have something exactly equivalent to the current single-signer model by issuing the root public certificates for the current CAs with the operating system.

As far as I know, it is not true to say that CAs are audited, and in fact there are well-known problems with CAs signing stuff that they shouldn't.

An advantage of the web of trust model is that you can incorporate CAs as parties that you trust (exactly as for the current model), but you can also require multiple signatures, which as far as I know is not possible with the current model. You might, for instance, require that two of the current CAs have signed a certificate before it lights up as "green" in a browser URL bar.

I use NoScript. The demo site requires code from googlecode.com to be permitted. While the Javascript provided by Google may be innocuous, I would personally not make this assumption. I don't think that it would be possible for it to get the private key, but I would suspect that it would do datamining which would reveal the email addresses in use.

No, there is no such point of diminishing returns. If you know a little calculus, you can prove this to yourself.

DHCPv6 still exists, and is useful for some purposes. It can do more than hand out IP addresses.