I expect physicist to have thousands of hypotheses about dark matter.

Until you can design an experiment to distinguish between those that are bollocks and those that are true, you're not going to make any progress. That's the hard part - designing and performing such experiments. This is why beautiful and simple demonstrations like the double-slit experiments are considered the most artistic and wonderful pieces of science.

The problem with dark matter is that we have so little information about the phenomenon that all hypotheses fit and few can be eliminated.

Crap.

Within the M25 (a motorway that's the technical boundary of Greater London before you hit the "green belt" where house building is limited) my workplace - a private school - has spent three years trying to get an Internet connection over and above a consumer DSL offering.

Money - not a problem.
Phone lines - not a problem (we have 4 ISDN's coming in, several lines capable of ADSL/VDSL, we are in a town, etc.)
Speed - fuck off, we can barely get 10Mbps for the ENTIRE site.
Alternatives - two leased line installations, first one having died after 18 months when BT COULD NOT DELIVER. The second one is similarly dragging from the ONLY other alternative provider in existence (and that is a possibility only because a millionaire that lives down the road bought them in to supply his house too).

Sorry, the USA might be less population dense, but that just makes the UK offerings even more shit given that we don't have vast expanses between customers. If you're lucky enough to live in a major city, yeah, you can get 50-100Mbps that you share with your street and 4G. If not, you're fucked. Some places in vast regions of the country (Cornwall, Scotland, Wales) can't get anything usable at all, despite having landline phones and, if you're lucky, basic GSM/GPRS.

And how many providers do we have in the entire country? You have ONE DSL provider - BT. Everyone else rents out the same BT lines using the same BT equipment. And there's ONE cable company (Virgin) that bought up the old bankrupt NTL cable company lines (that went bankrupt putting in all the copper which nobody used!) and pretty much hasn't cabled any new towns since then. Everyone else is a niche market, or deploys in only one town at all, or has expensive specialist equipment (3G links, municipal wifi, satellite, etc.) and is incredibly rare.

Seriously, 20 minutes on a train from Central London and I can't get enough DSL to my workplace to supply my own personal needs, let alone a school with 400 pupils and 50 on-site staff.

More - don't buy an entire house on the expectation of a company delivering a product.

Because selling an entire house just because you couldn't get Internet you were promised some dumbness of inordinate magnitude.

Did they not bother to ask neighbours first? If those neighbours have Internet, can't they piggyback on the wifi or put a microwave connection across at worse?

All I take from the article (twice now!) is "tech guy still trusts in suppliers' promises".

If it's that important you'd sell the house, you didn't do your homework beforehand. If it's REALLY that important, you'd probably consider leased lines, satellite and other more expensive methods.

Hell, just as a bog-standard geek the first thing I did in my house was check the phone lines, check 3G connectivity (now 4G but that wasn't around at the time), and look out in the street for the CATV manholes that UK cable operators dig lines to your house from. The only thing I didn't bother to do was properly check wifi signal propagation because I could already see half-a-dozen of the neighbour's wifi networks from upstairs. And that's in the suburbs. Stick me out somewhere in the sticks and you bet I'd be checking stuff on anything other than supplier's promises.

That's new to me:

VPN's can't have high latency?

Never heard that one.

Pretty sure just increasing a timeout or two at most, unless you're trying to do a ton of stuff in stupendously real-time over the Internet.

Hotel wireless is already a risk anyway.

Let's assume the wireless is open. Then anyone and everyone in an adjoining room can sniff everything you do over it anyway.

Let's assume that you are given the key to join the network. Anyone else who has the same key - same thing. AP isolation doesn't save you against someone recording your traffic and having access to the key used to encrypt it.

Wireless is UNTRUSTED. Even wired is UNTRUSTED. You do not know who's pushing that Facebook DNS entry to you, nor that the Facebook TLS is properly signed if you can't rely on the DNS entry.

When you're not using your own networks, use a VPN. That way you don't even have to care if someone bothered to put even WEP on the connection - the VPN gives you the security for your data. However, be sure that if you're doing this, you have a firewall (you are STUPID if you don't) as anything else can send you traffic in these instances too, no encryption, WEP, WPA, WPA2, it doesn't matter.

Every time someone says "join my wireless", replace it mentally with "just plug this cable that connects to all my local machines and also every guest that's ever had the same offer, into your laptop".

Firewall it. VPN it. Then you don't even need to care that it's an open network. And, shockingly, the same config will work with cabled networks.

And if it doesn't work? You don't want to use that connection. Any hotel that breaks your VPN is one that's almost certainly providing some poor replacement for it.

Or they could just put a toilet (or another access to the toilet) in the cockpit.

The pilot having to walk out of the cockpit mid-flight, wait for the passengers to vacate the area, and leave the cockpit in the hands of one man for something they must do several times a day just seems stupid to me.

How much do they save by you rebuying a teaspoon of ketchup once a month? Ketchup bottles are already refillable.

How much brand name exposure would they get to be the only company with the "unstickable" bottles? Ketchup bottles that are squirty without whacking required (the new plastic-sphincter caps), sit on their lids not their base, etc. have actually increased sales of (technically smaller) bottles in many industries. Hell, in the UK you can buy squirty mint sauce and it costs more than normal mint sauce despite being smaller. People STILL buy it.

This is just supposition, but it's the way I choose to understand it. Note: This is probably not science.

Imagine you're a time traveller but in the classic Hollywood sense where timelines can be broken without the end of the universe, etc. Marty McFly doesn't have to worry about standing next to his former self and breathing in the oxygen he would have originally breathed in, etc.

You can go back in time, steal some cash from yourself, bring it back to a different timeline, use it to make yourself rich. It's all fine. So long as, at some point, you can back and put that money back for you to steal in the first place. This is similar to how particles they borrow energy. So long as nobody notices (in this case, so long as the energy is returned before the "uncertainty" in the uncertainty principle can be resolved), you're golden.

Additionally, you are both "in" the room stealing the cash and "out" of the room simultaneously at the same time because you've been jumping back and forth in time (and maybe even in the room watching yourself stealing the cash in order to put it back once you're gone). In one timeline, in 1956, you were there. In another, at the EXACT SAME TIME, you weren't. So asking "where were you at this exact time in 1956?" doesn't give a simple answer. I was here. I was there. I was not here at all. And I was all of them at the same "time".

Time is just a dimension, so it's one hypothesis that particles may well be doing exactly this - hopping back and forth through other dimensions of space (and thus disappearing from ours and reappearing somewhere else), jumping back and forth in time.

So long as they repay their debts, it all works out and doesn't violate (certain readings of) energy conservation laws. And particles aren't intelligent creatures that decide to do this, they may just be "things" bouncing through dimensions quite ordinary to themselves but "time", "parallel universes", "alternative histories" etc. to us. Following even the simplest of physical rules in those circumstances could look like the weirdest actions ever from certain points of view.

Imagine you're on a 2D universe, you are a piece of paper and cannot perceive things not on your surface. A "ghost-like" car tyre passing through your universe will come from nowhere, grow, change shape, look odd, etc. and then disappear and never have looked like a car tyre. Same kind of thing. If you can't perceive the extra dimensions, this horrible weird-shaped thing just pops into existence, wobbles about a bit as a strange-shaped silhouette, maybe forms a hole in the middle if it fell the right way, then disappears. Or maybe it fell perfectly straight and you ONLY ever perceived a rectangle-like shape coming and going. Same object, same thing happening, tiny change in parameters, totally different outcomes that are very unpredictable for you.

The problem with quantum stuff is that we just don't perceive other dimensions at all, but the maths does.

(x) describes how far along a ruler you are.
(x,y) describes where a pixel is on a 2D screen
(x,y,z) describes where you are in a 3D world.
(x,y,z,t) describes an EXACT point in space and the time you were there (e.g. your birth).
(x,y,z,t,q)? We have no way for you to perceive that. But mathematically it's just another co-ordinate.

Don't expect a layman to understand it. The geniuses don't understand it. They can describe it. They can measure it. They can produce the formulae. But, just taking the knock-on effects and working backwards, they'd have nothing. It's only because the maths comes up with weird outcomes and that we then FIND those weird outcomes in the universe that anything actually looks right. Trying to play it backwards from the weird outcomes to those formulae that you can't understand is never going to help you.

It's like being a blind man and wondering how people can know there's a silent car coming when you can only detect a car's sound. If you can't perceive entire dimensions that - we're pretty sure - are required to exist for quantum mathematics to work, then you're only ever going to see a third of the story (our current best guess is 11 dimensions - we think - as a minimum? So eleven letters in the above example!).

As far as I understand it:

The problem is that it is not in any one state, until observed. Then we just see a snapshot of our particular history that led to that observation. Observation determines the state but also modifies the system forever more, too.

One hypothesis of this leads to the "many worlds" interpretation" - it's in only in one state but until we actually look (and therefore modify the system) we don't (can't) know which particular universe of possibilities we happen(ed) to be in.

Unfortunately, quantum physics gets a lot weirder, which only serves to show us how little we know of it. I get lost in it as it's maths way beyond my capability nowadays (despite a maths degree), but as far as my friends in the research fields explain stuff, you can even get things such as particles "borrowing" energy from their future selves (at least, that's one hypothesis of what they are doing) - they don't have to energy to do X, suddenly they acquire it, then they always have pay it back afterwards. It only works if you consider time as "just another dimension" or if you include other spatial dimensions they could be getting this energy from.

Though we might be able to describe a convergence between classical and quantum mathematically (at some point in the future), the outcome is always going to be the same because we're just 4-dimensional creatures. Weird stuff is going to happen.

Physics is going to get a lot harder for us long before it gets any easier. Breakthroughs are few and far between and we're only now properly confirming stuff that was discovered / hypothesised in the 20's, 30's, 40's, etc. (don't forget, technically quantum mechanics goes back as far as the late 1800's!).

Ctrl-^ does the same.

Ctrl-^ at the start of the block, scroll down with arrow keys. Ctrl-K at the end of the block. Bam, you cut the whole block.

Ctrl-U to paste.

I have to say, I grew up on DOS and then moved to Linux over time.

vi - everyone was talking about it being "equivalent" but I use it ONLY when absolutely, 100% necessary and I can't install anything else.

emacs - can't be bothered. Literally, just no.

pico/nano - lifesavers.

I don't want to change contexts, do line-at-a-time editing, with arcane commands that you have to "man" to find out. Pico/nano lets you navigate with the keyboard, has all the shortcuts clearly listed below, and doesn't play games.

Coming from anything else to use the text editing commands, people must hate them.

The equivalent of "apt-get install nano" is one of the first commands I execute on any Linux machine, no matter how minimal and console-based or fancy and gui-based.

When something says "edit this postfix file" or similar, I just want to run a command on it and start editing with an easy-to-find save and exit.

If you were trying to quit, is this not exactly how you'd start?

No, with diets, NOBODY starts this way. They all jump on calorie counting and weird systems and "don't eat random food group X" junk FIRST.

Eat less if you want to lose weight.
Smoke less if you want to give up smoking.

If you can do neither, you're not going to lose weight, or give up smoking.

Eat less.

Not losing weight?

Eat less.

Still not losing weight.

Eat less.

Granted, you still want to be having a mix of foods and not just less "burgers and only burgers 24/7", but it's a pretty simple rule to follow.

So long as you're eating a mix, you won't veer into malnutrition like this unless you ACTUALLY have a medical problem that requires constant treatment.

Of every person I ever see who diets, or who over-exercises in order to compensate, etc. I'm always just shocked that - rather than follow some faddy diet that's complicated and expensive and has all sorts of problems with it - they don't think to weigh what they eat over the course of a week and eat less the next week.

Not only that, it'd generate thousands of support calls and people would end up just taping it to "on" all the time.

More important and useful (and cheaper and easier) would be a mic indicator light as an option. If you want to see whether the mic is active, like you want to see if the webcam is active, just look at the light.

No disturbance, no unnecessary support calls, and an option to turn it off if it bothers you.

"Hiding" the phones among the IPv6 ranges is just stupid and not "security" at all (literally, security by obscurity!).

Even then, chances are that there's a range of consecutive IP's and just block-scanning through the IP's at random (say, scan every sensible address suffix because most people will start them on something sensible) will narrow it down quite quickly before you'll notice anything's happened. And chances are that most people will split at the usual boundaries, use the same IPv6 range (or the next one up) as their web servers are on, etc.

As stated above, the phones themselves have NO need to be on a public network. Push them through a VPN or similar if you really must but they should be on their own VLAN anyway (so you can QoS them properly and easily) and they shouldn't require direct access to the Internet anyway (the voice gateway is another matter that's separately handled).

But, better, stop buying, producing and selling devices that have debug interfaces that let you do ANYTHING on the device, remotely, without authentication. Because that's so dumb it's orders of magnitude more dumb than trying to hide your IP ranges in a IPv6 haystack.