You beat me to it, so all I can contribute is the link :-) http://www.imdb.com/title/tt0088286/quotes?qt0358678

<Offtopic>

You of course know the answer already, since you posted the quiz in your sig, but I though it might be fun to post the quiz in a more explicit form, so that more people have a chance to participate. So here goes:

What is the difference between these two groups?

Group A

• .startdot@example.com
• two..dots@example.com
• twodots@example..com
• enddot.@example.com
• startdash@-subdomain.example.com
• enddash@subdomain-.example.com

Group B

• ipv4@[127.0.0.1]
• ipv6@[IPv6:CAFE::BABE]
• !#$&={|}~?^/@example.com
• user@localhost.localdomain
• "quoted"@example.com

Three point hint: it has to do with RFC 2821.

One point hint: out of the approximately 200 000 people who have implemented a way to pattern match an email address, approximately seven have ever actually bothered to read the RFC.

</Offtopic>

Yup, it's the time-honoured "security through obscenity" approach again.

My bank's site requires three things to authenticate me:

1: a user code, 8 characters of randomness generated by the bank (something I and the bank both know)
2: a password, at least 8 character of not-very-randomness generated by me (something the bank can check without actually having to store it)
3: a four-digit number from a printed wallet-size list of one-time codes generated by the bank (something I have)

The password used to be also generated by the bank, but they came to their senses; now that I get to choose it myself, even the clerk who created my account (and possibly caught a glimpse at my one-time password list in the process) does not know everything that is needed to authenticate as me.

The extra trouble is, of course, the exchange of the one-time code lists. This they do by mailing me a new one when there are ~20 unused codes left in the old one, and then I just need to log in to their web site, give the id of the new list, and confirm the list change by a code from the old list.

Not nearly as high-tech as SecurID, but works like a charm.

Those would be one-time passwords. One time pads are something different.

Since Australia is a member of the UN, the right to free speech is quite explicit there too.

Universal Declaration of Human Rights, Article 19:

Everyone has the right to freedom of opinion and expression; this right includes freedom to hold opinions without interference and to seek, receive and impart information and ideas through any media and regardless of frontiers.

I have a HTC Desire HD. The battery usually lasts almost an entire day, so basically the thing is only "mobile" in the sense that you can carry it from one power socket to another.

Interestingly, the difference in fuel costs for 10 to 11 MPG is even bigger than the difference between 40 to 44 MPG. (proof: assume that the distance is 440 miles. The respective amounts of fuel are 44, 40, 11 and 10 gallons)

But whichever way you place a fraction, the divisor will be the one to throw anyone's intuition off, so it is equally easy to cook up a situation where thinking in liters/100km will mislead you:

If your budget allows you exactly 100 liters of fuel a month, then an improvement from 10 to 9 liters/100km will give you as many extra monthly kilometers as
    a) an improvement from 5 to 4 liters/kilometer, or
    b) an improvement from 5 to 4.5 liters/kilometer?

Solution: Neither. The respective distances are 1000km to 1111km, 2000km to 2500km and 2000 to 2222 km.

Exactly.

It does not really matter. There are three situations where you will ever need the fuel consumption stats, and in each one of them both ways are just as easy to use. I used MPG instead of km/liter below, since the two are conceptually equal, but the former makes for easier reading as it is a commonly used unit.

1. Comparing the fuel economy of cars
        bigger MPG is better
        smaller liters/100km is better

2. Computing the fuel costs for a given trip, when you cannot use the "top up before you leave, top up again at destination" method
        divide the trip length by MPG to get gallons
        multiply the trip length by liters/100km to get liters

3. Estimating whether you can reach a destination without refueling. This is the only case which you'll typically have to do in your head, since you are driving. Therefore you want to avoid division, which is difficult to do this way. However, both systems will cope easily:
      multiply remaining fuel by MPG to get your range, and compare with remaining distance
      multiply remaining distance by liters/100km (division by 100 is easy) to get required fuel, and compare with remaining fuel.

In the third situation both methods involve only a multiplication of two small numbers, so even if you cannot calculate the correct figures, it's easy enough to hit the correct ballpark with either method.

There does not seem to be a real difference between the convenience of the systems in day-to-day use.

It seems slightly silly to me that the content providers have pay for their network access in order to offer their services, which they usually do for free. A contentless net is a useless net, so it would seem reasonable that a content provider get their connection at very cheap prices, at least. Of course this will never happen, but let's say, for argument's sake, that Google started to aggressively renegotiate their peering agreements. It would seem that anyone not willing to peer with them at dictated terms would be left with an unsellable Google-less Internet.

The smaller content providers obviously cannot do anything so straightforward without at least uniting their power first, but if they ever did, I think the ISPs would be the ones to fold first, after all they have to sell _something_ to the home customer.

Naturally, that would mean every web page instantly becoming uselessly heavy with ads and no concern for bandwith usage, so let's hope they never get around to it.

(Full disclosure: I work at a company that hosts several relatively large web services.)