It can kill all of *us* though. I don't see it as OK if we all die but some bacteria survive.

Or even, you know, *lots of people*. The Earth has been incredibly inhospitable for long periods of time

Awesome, so the plan is to wait until lots of people are dying and everythings pretty fucked up, *then* start solving the problems. Sounds perfect, can't see a flaw there.

git clone --depth 1

To add to this, last time I used SVN, it seemed to transfer each file individually which was really slow. Git compresses the files and then transfers everything

While a little more work than with svn, you can checkout just part of a repository: http://jasonkarns.com/blog/subdirectory-checkouts-with-git-sparse-checkout/

Although I'd argue that if this is a common problem then you probably have several sub-projects.

Slightly tongue in cheek but a good statistic is that toddlers (3mo - 6y) killed more Americans than terrorists did in the last year. http://www.opposingviews.com/i/society/guns/toddlers-killed-more-americans-terrorists-did-year

How about predicting disease transmission from twitter?

I didn't say it'd scramble the fighter jets, but if someone enters a "I"M BEING COERCED INTO VOTING FOR SOMEONE" signal, that's pretty serious. If I entered it, I'd expect someone to follow up on it (not send around a S.W.A.T team, but have someone follow up over the next week).

This is a side issue though, as there's already the problem of how someone actually enters this duress signal. How is this done without anyone but the person entering it knowing it's the duress signal?

I'm describing just one problem in a theoretically perfectly secure online voting system. In fact, the problem with the duress signal is the delivery of it to a person without anyone else knowing. I was describing a problem even if *that* system was also perfect. Overall, I don't see the point in online voting. Going to a nearby school/church/town hall and ticking a box on a bit of paper behind a curtain is simple, cheap and pretty hard to fuck up. It solves many of the problems with coercion (the wonderful technology of an opaque curtain), fraud is harder because there are physical items to fake/destroy with people around.

Scale. Voting by mail is done in fairly small numbers and importantly is not the standard. You have to go through extra hoops to do it. As it's implemented, it certainly has the problems of coercion, but is probably better than stopping those people voting at all.

Internet voting, however, would be something I'd see as standard. Not a special case for those who can't make it to the polls, but for everyone. And that's where it starts to worry me.

Sounds quite complicated.
It'd need to be unique to each person, and have been delivered to them without anyone else seeing (you don't want the person you're afraid of knowing you're signalling duress). Then, given that someone is likely to turn up at the house (voter fraud is very serious, I'd expect the police to be involved) you could be found out.
If it's a workplace thing, do you want to risk getting fired?
The problem is you need to this to be indistinguishable from a normal vote to any observer, and however you do this cannot be intercepted or detected.

It isn't possible, because you can no longer have any reasonable guarantee that there was no coercion. You need to control the location the vote is cast.

Yes, if only the BBC were more like ITV or Channel 5. Why, then we'd be able to watch reality TV without being interrupted by nature documentaries with Attenborough, quality news or original comedies. What a blissful world that would be!

You're assuming here that if *some* people will still send spam that the problem will be the same. Sure, some people will advertise through email. However, you're now limited to people willing to pay quite large sums for modest exposure.

I'm most concerned about people who have a legitimate reason to send out lots of emails. Anyone running a newsletter for example. You'd need a way of allowing people to send emails without it costing them.

10 times *lighter*.

How about Scholarpedia?

What of the thousands of online copies of peer-reviewed papers?

Why only wikipedia?

A second challenge, try to find academic papers in a library. How many could you not find there?

If you are doing the test properly and mean "*only* non-internet resources" at the end, then I'm certain I'd win. My course was AI, most lecture notes were available online and failing that I could very quickly have the most important papers on the topic on my screen. I haven't taken a single (on topic) book out of the library during my course, have never felt the need to as I could get up-to-date information almost instantly. Those not using the internet would have had an extremely difficult time.

Your test is also somewhat biased, since the idea is to compare the internet an libraries. You've compared the internet to libraries and lectures.