minority?

One of that alleged minority is Linus.

However, in the Linux world you can pick and choose between many desktop UIs. My grouse about Gnome 3 is not that it sucks. Rather, that they developed a radically diffrerent UI without forking the source, so that if you have Gnome 3 installed on your system you can't also install or use Gnome 2.

Anyway, for now XFCE will have to do, or the Scientific Linux 6 distro (which will be sticking with Gnome 2 for many years to come). Then there's Cinnamon, which looks promising (a classic UI running atop the Gnome 3 libraries). This is what Gnome devs should have done in the first place, and only then started playing around with tablet-style UIs layered on the same library, as optional alternatives to the classic UI.

Invisible start button?

But it's got a "hot corner" that works like a start button? How does this differ from a start button, apart from you have to know it's there because you can't see it?

Maybe you don't have to click it either? Oh great, so there's a corner of the screen that you have to be careful to avoid with the mouse because all sorts of annoying or even dangerous things happen if you don't.

Then someone will work out how to paint it with an icon, give it an option to require a click to un-mask the stuff underneath, and they'll call it an app and sell it to you.

Taking the piss

No, that would be the water-cooling patent with the pump downstream of the heat source.

Search FB & TW

You mean, use Google? (There can't be any bigger data miner, not even in .secret.gov)

Not mine

The physics of chemistry is universal. There's no other element that has even half the versatility for building complex molecules that Carbon does. Single bonds, double bonds, rings with de-localised electrons, stereochemistry where one carbon is bonded to four different groups or the cis/trans arrangement around a double bond or the sequence of groups around a ring ....

Alien biochemistry may look nothing like ours, but I'll eat my hat if it's not based on Carbon, with Hydrogen, Nitrogen and Oxygen in important supporting roles. I also think it highly likely it'll have proteins (loosely defined as long chains of amino acids, CO-NH bonded).

No hinge would be even better

It should be a two-part design. Tablet. Keyboard+mouse. Bluetooth link. Software that adjusts the user interface so that you don't have to touch the vertical screen to accomplish anything when the keyboard/mouse unit is present. In "enjoyment mode" just take the tablet and leave the keyboard behind.

OK, it's probably trivial to make it so that the two parts clip together for transport and for keyboard use on the move - you might call this "netbook mode". But at home there would be a plastic stand for mounting the tablet vertically above the level of one's desk, and put keyboard wherever the user finds most comfortable.

Geostationay - a bit slow.

Geostationary orbit is fine for high-latency emergency communications between meatware systems. It's also OK for transferring gigabytes of data using large buffers.

But it takes light about a quarter of a second to get up to geostationary orbit and back again (i.e. a network packet round-trip time of half a second). Skynet will need low-altitude satellites, or more probably will just hijack our fibre-optic cables here on the ground.

As for aliens, they must have FTL technology or they wouldn't be here!

Diagonal versus width

The size of a screen is its visible diagonal measurement.

Not sure whether there is a standard for the "size" of a laptop, but I've always assumed they were referring to its width.

Therefore it's quite easy for a screen to be "bigger" than the machine it's in. The maximum "size" of a screen is gigen by Pythagoras: sqrt( width^2 + height^2 ). Thes subtract the parts of the diagonal that aren't display.

Maxtor?!

> "I used to like Maxtor until seagate got their grubby hands on them".

Maxtor was the company whose drives gave me so much grief that I came to avoid them, albeit with a full understanding that the probabilities favoured the hypothesis that I'd just been unlucky. My worry was the inverse of yours!

Seriously, you'd need an absolutely huge number of drives to start drawing conclusions about manufacturers in general. What's clear is that every manufacturer has on occasion shipped bad batches, caused usually by a component supplier shipping substandard components . Also that some models turn out to age less well than the manufacturer hoped. This is inevitable, given that the tech moves so fast that by the time a drive is known to be reliable in service, it's also obsolete.

Maybe not so boring

Can it be gamma rays *all* the way up? At some energy you'd have a photon with more energy than the entire observable universe. Maybe the entire universe, if it's finite. This isn't pedantry at all, if the whole universe did start very small in the big bang.

Some (exotic, unproved) theories suggest that photons of sufficient energy may travel slower than low-energy light. Slower-thn-light photons, sort of a counterpart to faster-than-light neutrinos? Observations of the next supernova may throw some , er, light on this.

Lots of unexplained observations remain.

We already know lots of stuff outside the standard model.

Gravity, for a starter. One of the most stunning numbers in physics is 10^38, the ratio of the strength of the Electromagnetic force to the Gravitational one. You don't really understand anything about matter, until you appreciate that the electromagnetic force between an electron and a proton is that many times stronger than the gravitational force between the same.

Then there's dark matter, inferred from the observed nature of galaxies and the impossibility of binding them gravitationally if what we see is all there is. (There are good theoretical reasons why it can't be lumps of non-luminous ordinary matter, sized somewhere between marbles and Jupiters).

And "dark energy", needed to explain why the universe appears to be not only expanding but accellerating.

We might get lucky and spot a particle or three of dark matter in the CERN detectors. (Oddly that may be more likely while the LHC is down than when it's up). Or careful observations of the things we know it can manipulate may help pin down a better theory that in turn will guide our observations. (It's easier to find a needle in a haystack if you come to suspect it may be magnetic).

And if those neutrinos really are going faster than light, it's time to tear up all the theories and start again.

Python?

It'll run Linux and Python. In my opinion if someone doesn't take to Python like a duck to water, s/he'll never be a programmer. What more could you want?

(Java? Perl? Ruby? Occam? Fortran? C++? Not my choices for a first language, but those as well, and more).

$$$ because they are thinking globally. Outside the EC most people probably don't have the faintest idea what a £ sign denotes (assuming it even displays right!), and little more idea what the GBPUSD or GBP/local exchange rate is. Global trade works in US$. Even the BBC World Service converts prices to US$.

Brick it?

Why "Stlll ...". They've thought to boot off a socketed SD card rather than anything soldered down or hard to reload. So download the image if you didn't make a backup, find a PC, borrow a camera-card USB thingy if you don't have one, and re-load the boot media.

Or if you mean burn out the hardware with a 13.6 volts 1000A down a digital IO line or something like that, it's part of the learning process. At £35 one can afford a mistake or two. (What do beam tetrodes cost these days ... never mind).

Neat idea, but ...

Are you really telling me that if they offered, you wouldn't take the $100M cheque from Google?

Apparent equilibrium is rarely simple.

>All it can show are the limitations of the mathematical model and data collection...

There are two sorts of equilibrium conditions. The intrinsic, such as a passenger jet in straight level flight. And the dynamic, such as a modern fighter jet doing the same.

The former has swept-back wings with wingtips higher than their point of attachment to the fuselage (when the plane is level, obviously). A moments thought shows this means if it gets tilted down on one side, the wing on that side generates more lift than the other one and the plane levels itself. If it rotates about the vertical axis, the wing on the outside of the turn generates more drag and straight-line flight is resumed. This is intrinsic stability.

A modern fighter jet has wings that sweep forwards and downwards. It couldn't fly, were it not dynamically stabilized by the millisecond, a computer cancelling out every random twitch before its intrinsic instability can send it tumbling out of the sky. A complex feedback system, instead of a simple one.

Why? The passenger jet's stability means that it cannot change direction so easily or quickly. A change of direction means fighting its own tendency to maintain straight level flight. A fighter's manouverability is a matter of life and death, and its intrinsic instability gives it a huge edge.

I've deliberately used a simple and man-made example. Nature has evolved the same solution in many contexts and at many levels. Nature's apparent equilibrium conditions are almost never static. Instead they are chaotic with feedback, or "edge of chaos" self-optimisation to the prevailing conditions prone to avalanche-like changes if those conditions change.

Try to model a system on the edge of chaos and you won't get deterministic answers. You may get statisticaly significant ones if you run for long enough or do enough runs with different small random perturbations to each. You can forecast tomorrow's weather. You may be able to forecast the climate of the next century given changes to the atmosphere (or maybe not). What you can't ever do is forecast the weather a year hence.

Model the behaviour of the component parts even slightly incorrectly and your overall results may be dramatically "wrong". But in another sense, they aren't. The real world is unstable. Fighter pilots know not to push the envelope to the limit, until there is hostile incoming. We'd do well not to push ecosystems to the (unknown!) limit of apparent stability, because we don't and can't know what happens when the apparent equilibrium is punctured. And the system starts with the biochemistry and genetics of bacteria, and mixes levels with a vengeance, like the sort of code you absolutely never want to be in charge of maintaining.

Like a brain does.

Gaia?

Thistledown / Eon

And then as your technology advances, you make the inside bigger than the outside. MUCH bigger.

I wonder if Greg Bear got that idea from Dr. Who? It's like a Tardis, but ... bigger.

A better idea

Agreed. I don't think they've quite thought this through.

Suppose something completely trashes our civilisation. Might be nuclear war, a small asteroid, a bad mistake in a bio lab, blind chance / mother nature / God (take your pick) unleashing the mother of all human flu virii. Human genetic diversity is unusually low - we went through a genetic pinch-point ~80k years ago. We're far more vulnerable to plagues than (say) chimps.

After the few survivors have climbed back to agriculture, iron and cities after 500, 5000 or 50,000 years, they'll have lost just about everything we know today and will have to discover it all the hard way. Paper won't last that long exposed to the elements, digital media neither (anyway they won't have readers). How can we help, just in case it happens? At little cost, unless a very pessimistic billionaire gets involved?

Ceramics are far less destructible. We have the technology to print a lot of small detail onto plates etc. for little extra cost compared to humdrum decoration. (We're using it to print marble-effect tiles, no two alike! ). What about an encyclopaedia ceramica? Sell as many as possible as novelty items, and in the far future if someone digs them up they will still be legible. Plates and tiles might even last long enough to reach the next intelligent species, if this one goes extinct.

Trouble is, they'll also have lost English as a written language, and cracking an extinct language without a Rosetta stone has never been done. So another interesting interesting part is to devise a code that will let them read the stuff we're trying to leave them with.

Hopefully, it'll never be needed. However, it has the makings of a fun project - what to write and how to code it, to save the future 10,000 years of pre-electrical civilisation? What to attach it to, that will be produced in large numbers and be noticeable by the naked eye after that sort of time buried? And so on.

As you say you CAN do it with XP!

"Yes, you can do this with Windows XP" and in fact I have many of my user's systems set up this way, must have been lucky not using the buggy softwares you refer to. Anyway, all MS needed to do was the necessary stuff to make it work better. Introduce 7-style escalation, if you really think that's a better way to accomplish this sort of thing than the (safer) sudo / "run as administrator" approach. Either way, no need to tear everything up and start again. Especially not at the user interface level. Maybe inside the kernel where only systems programmers tread.

The rest of your comments seem to be making my point as well. XP was indeed quite poor when it first shipped. It has been improved *incrementally*, as you acknowlege. Facilities have been added without tearing up the framework. If it's true that the XP kernel was an irredeemable mess by SP3, then by all means rewrite it. I explicitly pointed out that it's possible to ship a new kernel under the hood, without the user even having to notice the change. That's incremental improvement.

I'd have welcomed Windows 7 if it had preserved interfaces from XP, changing only the ones which had to be killed because the XP way was an irredeemable security weakness or suchlike. Instead they ripped everything up and started again. Loads of costs in lost productivity while you replace your old skill-set with a new skill-set (for everyone from secretaries to sysadmins). For example, many of the applets in the 7 control panel are recognisably the same ones that ran under XP, so why move them all around under a brand new control panel UI so you have to learn to find them all over again? WHY???

If they did the same thing to a car they'd have put the accelerator where the brake is at present, and replaced the wheel with a joystick. It might even be a better way to design a car, if cars didn't already exist. Car manufacturers will never make such changes, becase with cars that would kill people, not just cause lost productivity and unnecessary annoyance.

BTW I don't regard this specifically as a Microsoft problem. I'm equally scathing about whoever was responsible for Gnome 3 on Linux (which destroys Gnome 2 - you can install as many different window managers as you like on Linux, EXCEPT not Gnome 2 and Gnome 3 on the same system).

Algae are so easy compared to plants.

Algae are easy to grow. Any pail of water left in sunlight will turn green. Add some nutrients (sewage, say) and they'll turn it into thick green slime. Most people have created a pail of geen slime at some point in their lives, and it's a shame that the usual reaction has been "Yeuuuch!"

Algal photosynthesis is much more efficient than multicellular plantlife - there's no need to for them to maintain structural integrity and internal transportation networks, which holds a plant back. Algae are the plant equivalent of bacteria. Exponential growth until limited by nutrient supply.

Algae don't need fresh or clean water. They grow in the sea, in raw sewage, in toxic effluent. (They'll actualy eat sewage and produce much cleaner water) This means you can put algae farms in a desert (most sunlight) and culture tham in brackish or salt or toxic groundwater that's not a valuable and depleting resource like freshwater groundwater is.

If you can bio-engineer them to go 20x faster given 20x CO2 supply, you can get to burn coal twice. Once in a power station, the second time around in a car (as algal bio-diesel). I'd also observe there are lots of natural CO2 and carbonated water sources bubbling into the atmosphere.

Mere selective breeding will get you to algae that produce pretty good diesel fuel when you squash and filter the green soup. (What's left over is nutrient for the next generation of algae).

As you can guess I'm a fan of algae, along with solar panels. Technological civilisation doesn't have to grind to a stand-still when the oil runs out, and we don't have to continue raising atmosphereic CO2 for the next century either (with potentially dire and irreversible consequences for the climate).

Not quite

Certainly an old version of Red Hat Linux (say) will carry on working as well as it did on day one as long as there is hardware that it can run on.

However, five years after the next major version is shipped the security patch stream will dry up, and you'll discover that upgrading rather than rebuilding from RHEL5 to RHEL6 isn't supported. You can try ... most folks who have tried recommend that you don't. I've occasionally longed for a major version upgrade wizard or suchlike, but I can quite see why they don't think it's economically viable to supply and support the same. Too large a phase space to test well in advance.

XP

The real question is what did / does Vista or 7 bring us that we hadn't already got with XP?

Answer #1 - Nothing, but it gets lots more money for Microsoft.

Answer #2 - a less bugridden kernel. Maybe. But one can upgrade or completely replace a kernel without trashing the entire user interface. Why do Microsoft keep doing that? (Not just with the O/S, but Office and the other products as well).

Answer #3 - same as #1. They make lots of money out of books, training courses, (re)certification exams ...

Answer #4 - well, at least making all those old PCs obsolete makes sure new bigger faster ones get developed. Good for Intel, gamers, number-crunching scientists, and anyone who runs Linux on PCs that will be thrown away because they can't run Windows 7.

OK, I'm happier now. I'm a number-crunching scientist who runs LInux when he can.

Blue sky?

You mean, researching bolts from the blue?

http://en.wikipedia.org/wiki/Lightning#Positive_lightning