For the moment, I think we can limit ourselves to the number of atoms in the solar system. One rough estimate is that there are 10e29 stars in the universe. If the atoms were divided up approximately evenly between these star's systems, then there'd be 10e82/10e29=10e53. So we have one IPv6 address for each cluster of 10e15 atoms.

Except! I've heard it estimated that about half the matter in the solar system is in the sun, and we don't want to use up the sun to build computers, because we need it to power the computers. So, 10e14 atoms per IPv6 left to work with.

So the question before the audience is:can you build a device that implements an IPv6 stack and a minimal radio transmitter that allows it to communicate with other, similar devices, using only 10e14 atoms? If so, or if it can be done in less, then we may have a problem*. Otherwise, I think we should be fine for now.

(To give you a rough estimate of what you're working with:10e14 atoms of silicon would mass about 46 nanograms.)

Submit your solutions to iwannahelpdestroytheworld@weregonnafreakingcreatethesingularity.com :)

* Although the problem may not be manifest until we convert the *entire* Earth, core and all, into these devices, along with all the other planets, and colonize the Oort cloud, and do the same there. :)

It will also vary depending on the performance of the vehicles immediately ahead of, oncoming-and-passing, or crossing ahead of the street view vehicle. Especially the first: The sensor will be running in the exhaust plumes of the vehicles ahead of the street view car, so the map will be a very non-random sampling.

On the other hand, the partculate and "volatile organic compounds" sensors will produce some very interesting data. The latter is what the federal standards call "unburned hydrocarbons" when emitted from an engine, and the output of modern engines is vanishingly small. But many species of evergreen trees emit them in enormous quantity, as part of their ongoing chemical warfare against insects that eat trees. That's what the blue haze around pine-forested mountains (such as "the Smoky Mountains") is about. You can literally destroy (by extreme and long-term contamination) an automotive conformance test cell (the room where they test the car's emissions), requiring it to be torn out and rebuilt, by placing a Christmas tree in it overnight.

I expect some towns in remote, forested, mountain areas, where people move "for their health" and "for the clean, fresh, air", to get a rude awakening. B-)

But I doubt it will affect the extremely tight standards for automobile engines - except maybe to cause a flap that tightens them further. These days many engines are so clean that running then can IMPROVE the air quality in some places (such as portions of Los Angeles, with topography that created such a thermal inversion that a single settler's campfire could leave the whole valley filled with smoke for a day or more) by inhaling and burning far more hydrocarbon and particulate pollutants than they create.

Ok, so he's the CEO of a big company that makes robots--among many other things. So I really have to wonder if he's actually as clueless as this makes him appear, or if he's cynically trying to convince stupid people that they should by his company's pseudo-friendly robots?

Or is there some third option I'm overlooking?

I mean, he might as well say, "robots must be designed to answer the ultimate question of life, the universe, and everything." That's just about as plausible, given the state-of-the-art. (And then he could try to sell us speaking robots that can say "forty-two".) :)

The lack of accidents and crime are more likely related to a general trend in crime going down from before they started turning off the lights. ... Give me at least one full year worth of data so I can compare it to the prior year, and have half of the country keep their lights on so It can be compared to the same time frame as well.

Hear, hear!

There's lots of room for methodology errors. Here's another:

Comparing murder rates between Great Britain and the US is complicated by differences in reporting. The US bumps the murder stat when there is a body and evidence of foul play. G.B. bumps it when they have a conviction.

Do they do that with other crime? If so, stable stats in the absence of street lighting might mean that any rise in crime is compensated for by a fall in identifying, apprehending, and convicting the criminals responsible. (Indeed, turning off the lights might easily result in LOWERED crime statistics at the same time it was causing a drastic increase in actual crime.)

I'm an anesthesiologist. I put people to sleep for cardiac surgery. My hospital does around 400-500 hearts a year... and we don't kill any dogs.

What hospital is that? I'll want to avoid it if I ever need heart surgery.

Seriously: How does your cardiac unit's mortality and morbidity rate stack up against those of hospitals where practice surgery on live animal, models, at least where the surgeon is new to the procedure, is more common?

For many software patents, I'd agree with you.

The problem with video compression is that many of the patents involved do represent real research, the expensive kind. They aren't one-click shopping patents. They're fundamentally pushing forward the state of the art. The people who do that work are expensive and need a lot of time, so, there has to be some way to pay for their efforts. Google's approach of subsidising all research via search ads is perhaps not as robust as one might hope for, even though it's convenient at the moment.

I don't know if DASH specifically is complex enough to deserve patent protection, but if you look at the massive efforts that go into the development of codecs like h.264, h.265 etc, the picture gets more complex. It's not pharmaceutical level research budgets but it's probably the closest the software world gets.

No, the issue is that it's open source and carriers customise the components. Android had a working online update infrastructure since day one, actually since before Apple did. But that's no use when the first thing OEMs do is repoint those mechanisms at their own servers and make huge changes to the code.

The comparisons with Linux are especially strange. Guess what? Upstreams who develop software for Linux and see it get repackaged by distributors are in exactly the same boat as Google. They see their software get packaged up, distributed, bugs possibly introduced and then upgrades may or may not make it to users. Yeah yeah, Debian say they backport security fixes. That's great when it's a popular package and a one liner. When the security fix in question is a major architectural upgrade, like adding a sandbox to an app, then users just get left behind on old versions without the upgrades because that's the "stable" version.

And of course many users are on Linux distros that stop being supported pretty quick. Then you're in the same boat as Android: old versions don't get updates.

This was a study of developers. Developers are not exactly typical users. Developers like things like vi and EMACS. And, in fact, developers can already buy keyboards with (for example) caps lock switched with control. (If they care, and are too lazy to remap their own keys.)

Do a broader study of general computer users, and then maybe we'll talk. (No real skin off my nose anyway, since if you design a keyboard layout Idon't like, I'll just remap it to be the way I do like. 'Cause I'm a developer.)

So, it's a GMO, which means the science-deniers on the left will hate it, and it reduces greenhouse gases, so the science-deniers on the right will hate it.

Basically, this is what we need, and it hasn't got a chance of success.

I'm an anesthesiologist. I put people to sleep for cardiac surgery. My hospital does around 400-500 hearts a year... and we don't kill any dogs.

So maybe I'm not up to date, or things are/were different in research hospitals.

My personal info was based on stories told by my mother, in about the '60s, when she was a special duty RN at the University of Michigan hospital, often handling cardiac recovery.

My favorite was the one where the UofMich hospital cafeteria, which had been purely open seating, established separate rooms for the staff to eat after an incident where patients' families overheard, and were traumatized by, a cardiac surgeon's response to a question. Asked how his operations the previous day had gone (referring to his experimental and/or practice surgery on a collie and another dog), he said "The blonde lived but the old bitch died."

The kids and adopted dogs story was from my wife. The surgeon in question was Dr. Albert Starr in (at least) the '60s through '80s. He was at St. Vincent's and also flew, with his team, to operate at a number of other west coast hospitals, university and otherwise.

A possible solution would be better simulations so that a student can learn by doing. I think it is a very different than working on a cadaver or simulated patient using conventional methods.

You obviously aren't familiar with surgical departments or you wouldn't have missed practice surgeries on live animals.

For instance: a typical cardiac surgeon, shortly before EACH operation on a human patient, does a practice operation of the same procedure on a live dog.

One pediatric cardiac surgeon was much beloved by his patents and their families, because (with parental permission) he would let the kid adopt the practice dog, rather than sending it to be destroyed. The kid would wake up from surgery with the new puppy beside him, with the same bandages, etc. (and a day or so farther along in recovery). The dog having been through the same procedure and having helped save the kid's life even before they met made for very strong owner/pet bonds. (There's always a live, healthy, practice dog. If the dog dies (or is severely damaged) the assumption is that the procedure failed. You DON'T do a procedure on a human if it just killed a dog. You analyze, adjust the procedure, and repeat until success.)

Getting skills up does NOT require, or usually involve, a lot of practice on JUST advanced simulations, cadavers or, live patients. The live patients are just the last step, when the skills are already finely honed, and the animal models provide immediate feedback, real situations, and automatically correct modelling of mammalian life processes.

Why don't publishers put the ads in a section of the page that can allow the rest of the page to load and render before the ad loads and renders?

Because you could stop the loading once the content you wanted was rendered, thus skipping the ad.

So the pages are set up so the ad loads and renders first.

This is nothing new. The administration has been saying this all along, and the stupidity of it has been pointed out by the people they lionize as having "done it right" in the past.

Read this: http://www.theguardian.com/com...

"How would an experienced developer get these problems in the first place?"

A lot of projects do not follow widely-accepted best practices... even if they are experienced... and that is a problem!

A remarkable number of OSS projects fail to have a public source control system (#2). That includes many established projects that everyone depends on. Actually, a number of OSS projects - and projects that people THINK are OSS but are not (because they have no license) - fail many of these points. It's not that Red Hat's internal processes are immature; Tom was trying to bring in software from someone else (Google in this case) and was fed up by the poor practices from people who should know better.

Yes, #7 refers to a best practice (let people pick their install directory) that's been around for at least 20 years and probably much longer, but it's still widely NOT followed.

Anyway, that's Tom's point; there are a lot of widely-accepted best practices that are NOT followed, and that needs to change.