"while cockpit video cameras may help determine the cause of some crashes, there are plenty of recent examples where it wouldn't help."

While logging data from left engine working regime may help determine the cause of some crashes, there are plenty of recent examples where it wouldn't help.

And your point is?

"I thought we got rid of slavery a century or two ago."

Well, there are in fact professions that already have special labour conditions that would be considered border to slavery were not for the fact that it is socially accepted that those constrains should be in place. Police and military come immediately to mind.

AFAIK, at least in EU, flight controllers already come under this designation due to national security considerations and maybe pilots are also of this kind. If they are not, they could be in the future.

"Why exactly are data recorders antiquated? I mean the concept, not a specific device."

They aren't, in the slightest.

"Pilots hate this idea because it will show they are human."

Exactly that. You, of course, already know it. You want non-instrumental data recorders because not everything that happens in a flight is measured by instruments. You already have voice recordings and, as you say, they are triggered in specific moments and events. Video could be the same as long as it offers more relevant information than other techniques -and it sure can do, or else we wouldn't have developed TV, radio being enough.

"They make jokes, complain about work, talk about their weekends, etc."

Exactly. They *already* do this despite having their conversations saved so, what's the added problem about video imaginery?

If it can be used to increase flight safety it must be put in place. Full stop.

It's only that it have to cope with the fact that part of the "machinery" is human and it has, therefore, to cope with "human envelope conditions" which include there must be moments for relax and moments for intimacy, just like is already happening with the currently in place data logging facilities.

"Then we need much better auditing, because the open source development process means that any random dude can offer patches."

OMG!

Quick, stop the presses!

As if a random dude offering a patch and a team accepting a patch were the same thing.

"They mean that the tool has to be publically available... Ie, if you build a web service, you have to make it available as an "open source" of data instead of charging for access... But only for the duration of the project (usually around 3 years). You can release your code in any way you like, or not at all. The commission doesn't care.

When they say they are going to invest in open source, this is what they mean."

Citation needed.

"The paper provides /very/ little quantitative detail with which to assess efficacy. This is not science."

Oh, no, of course not. It is *citizen* science!

"I'm sure that'll be enough to get all the ISPs to delete the state from their coverage maps"

Which would represent a more valid status of their coverage than current.

Now, all that's asked for is "you said you covered this area, just do it then", which doesn't seem to be too much: other companies that feel they can get into their promises will enter the State and will eat their business.

Isn't this the very basis of capitalism?

"Always verify everything yourself and don't trust anyone.

Pay for the previous owner's internet for a month or two"

That's very easy to be said -after the fact.

What else should the prospective owner pay before buying the house just in case? Sewage? tap water? electricity? Maybe he also should burn out the house just to see if the firefighter brigade can really appear in time, right?

If they don't, the customer is free to hire a different company to lay out the infrastructure, at whatever rate it costs, which the ISP will be forced to connect to their network within a week after they get notified. If they fail to connect the line in time, the ISP will be fined by whatever they make daily.

All the bills, plus a management fee for the customer will be payed forefront by the State, which in turn will litigate -not the customer, to get their money returned.

You are right. I forgot about B cells, and I've should mentioned them since I did it about the germinative line exception.

Now: do you think talking about B cells makes any difference for the base answer to the original question, "I'm not sure how they know (or why they think they know) those sequences are representative of those in each tissue."?

Except chimeras, yes.

Are you going to claim they appear in any significant number in mammals? (and I know about prions) ...on a thread about somebody ignoring the very basic fact that multicellular organisms own just one genotype?

"But when you examine the accident after the fact, it usually turns out that it did happen in an incredibly predictable and controlled fashion."

When you examine *after the fact*?...

I don't think "predictable" means what you think it means.

For all that I know, AI stands for Artificial Intelligence, right?

I think you are not arguing on the "A" part but on the "I" part so, then, what's the difference if the "I" comes from a human or a machine?

what happens if human-driving cars end up dominating the roadways--the rules that are currently mandated to ensure safety won't necessarily be the optimal ones when most cars on the road are driven by abiding citizens. And if you assume that all other cars on the road are driven by an abiding citizen with a given set of rules, tweaking the rules on your car (say, increasing your "aggression" parameter) could lead you to dominate the roadways... at least until other drivers catch on.

Again, what's the difference?

"I never really saw anything not work in a predictable controlled fashion"

Accidents happen whenever something doesn't work in a predictable and controlled fashion and, believe me, accidents do happen. Oh! and butter does melt in your mouth.

"Non-sentient devices will always behave in a predictable, controllable fashion."

No, they won't.

And no need, either.

If you are moving and your piano falls from your window to my car parked below, this is a very nice example of harmful unexpected interaction between things. Do you think we need to embed special laws within cars and pianos to deal with it?

I, from my side, will just sue you for repairings to my damaged property and done with it and can't see why if it were a case of "my AI thingie" being damaged by "your AI thingie" would be any different.