if you make the frame from wood, you lock up carbon. steel is a total waste of resources, also conducts lightning through the frame of the building.

plastic is the worst possible choice for the panels, almost anything else is better, as long as it is reasonably fireproof.

sometimes long life is a bad thing - plastic is damned hard to keep clean, and harbours disease - it would be far better if the panels were made of something local, cheap and recycleable, that needed to be replaced at least yearly.

with a little bit of careful thought, just about any country could produce it's own emergency shelter system out of local resources.

funny how the western world becomes less and less competent at anything, the fatter and more ikea it gets.

these people should give up saving the planet, and learn how to feed their own children properly, they are becoming as dumb as the cows that seem to be the only thing they eat.

the tests they show only serve to demonstrate how slow python string concatenation is.
they then go on to say that disk writes are bufferered more sensibly, and this is why they are faster.
the key thing to notice is that disk buffering happens in **memory**.

memory is still, and will remain, much faster.

you just have to use it properly.

clarkson has made a career out of being a dinosaur, and i think he knows it.

you're welcome in the real world jeremy and i, for one, am interested to see what you will make of your new life.

webnames.ca have been excellent, been with them going on twenty years.

build a track suit, generate power from running, use this for cooling, stick a little hamster logo on it...

when i searched i kept getting links to some stupid escape media tv serial out of the blighted states.

says a lot, really, the things our resepective cultures prefer.

(or related agency) published a detailed study on this subject in the late 90s.

as far as i remember i found this report on The Well, but i have since lost track of the file, and if anyone knows about it, i would really appreciate a chace to study it again.

the report was notable, as i remember, because it actually went so far as to suggest that weed makes better drivers - basically because they are more relaxed and therefore more likely to make the correct decisions in an emergency.

4.20, of course.

ars technica makes the point that neither radio shack nor sprint are willing to comment on this story, let alone confirm it.

http://arstechnica.com/busines...

if you value your craft. the bespoke solution is not for everyone, but there is something to be said for the perfect fit. i am able to confirm that it worked well for me, and remains the second most useful thing i ever developed.

thank you for clarifying. my opinion is that, since we have the ability, it makes a lot more sense to record both streams on both devices, as this gives a better chance of recovery, should either or both be compromised. i wonder what the truth is...

the word is adaptation, surely?

only the bbc, so far, has shown the ability to adapt anything with any degree of responsibility.

perhaps i do not understand merkin manglish - maybe adaption means "convert to a drug"?

please read the reply (currently scored 5, informative) to which i am responding. according to that, they are not different devices, they are redundant units - there are two, and BOTH record BOTH cockpit voice AND flight data, i presume on the same media. at last word the bbc is reporting that the cockpit voice is still missing, but this does not make sense, if the AC i am replying to is correct.

(currently scored 5, informative) to which i am responding. according to that, they are not different devices, they are redundant units - there are two, and BOTH record BOTH cockpit voice AND flight data, i presume on the same media. at last word the bbc is reporting that the cockpit voice is still missing, but this does not make sense, if the AC i am replying to is correct.

look at the damage that closed source has done to us, in the virtual world, and project this onto the physical world.

you must not give machines the same private autonomy you allow microsoft (for example) to take.

it is bad enough we permit closed source to handle our data, and the lessons from this are obvious.

a closed source AI operating in the physical world is a golem and should be considered evil by definition.