I believe he exaggerated how thoroughly you need to apply the dust. Just be sure you get places that the cockroaches will walk should be enough. Under the stove, refrigerator, on the shelves, etc. And don't remove it.

Mind you, I'm sure his approach would work, I just think it's probably overkill based on what I've heard previously. You do, however, need to be sure the boric acid remains in place, because you will be continually reinfested from where-ever the original infestation came from.

P.S.: Boric acid is not non-toxic, though it's also not terrible. But keep it away from food. Wash your hands after handling it, avoid breathing the dust, etc. If you put any on food handling surfaces, be sure they are well washed before you again use them for food handling.

I don't believe they did. OTOH, the theory is still in the development stage, so perhaps that's just in a part they haven't seen yet.

Are you certain that it can't be done?

Nobody's been able to do it yet, but that's not proof. I've heard arguments as to why it's impossible, but I'm not physicist enough to understand them enough even to say whether they were convincing or not. The last time I tried to understand that area there was a general agreement that it couldn't be done, but there were several qualified physicists that felt that it hadn't been proven.

Or, from a different prespective:
Models, Logic, Desire, and Motivation.

P.S.: Single words don't map the concepts well, but the alchemists were as much philosophers as chemists (with different individuals leaning in different directions).

OK.

PyPy might be fast enough, but I've never benchmarked it, and definitely not the version for Python3.

FWIW, it's possible to write fixed length data records in Python (including Python3). It's just a real nuisance.

OTOH 3 might be a real problem, if I got far enough into a project in Python. So far the kind of project where that would be a problem has washed out earlier on other grounds.

As for 4: Nobody seems to do multi-threading well. Python's only a bit worse than others. It's gotten to the point where I write things as multi-process instead. With ANY virtual machine, that means a lot of overhead when starting things up, and a lot of continued RAM overhead. So you avoid starting processes frequently by using something analogous to threadpools, and communicating by message passing. (There's a couple of languages that handle this aspect better, but they have other problems.)

So 5) There is no perfect language. You choose the language(s) for the project. Python3 is one of the languages that I more commonly choose. D (dmd) is another. Vala is a language that I keep my eye on, because it shows a LOT of promise. But it's been promissing for years, and every time I've tried it, there's been major problems. (I feel the run-time error messages merrit the description I've heard: "encrypted Klingon".)

I've really only got two complaints with Python3:
1) It's too slow.
2) It's too hard to write fixed length blocks of data to disk.

Both of these could be addressed with optional typing, but I don't expect that to ever happen, because the trend is in the opposite direction. OTOH, any year now I expect Vala to become usable.

Perhaps they are legally required to have a warrant, but there's no punishment to them if they just ignore that restriction. At least not locally. (OTOH, the local police have been awarded a federal oversight manager with power to fire the chief of police unless he cleans up the department, so perhaps that's not common.)

We've been headed towards a centrally controlled police state ever since the Civil War. Actually, since long before that, but that was the point of inflection.

The problem is that governments want to control. In fact, that's almost the definition of a government. So they tend to be run by people who are interested in control. Those people may have other goals, but control is their common goal. And advancing technologies have made increased amounts of control realisticly possible. (Please note that I didn't say anything about "human rights". Libertarian societies can be incredibly oppressive in that area. And controlling governments can be rather generous.)

FWIW, I distrust all centralized locii of control. Each one is a single point of failure. This is why I consider the GPL to be the best license. And this is why I would favor a democratic government. (It's not because democratic governments don't make truly horrendous decisions.) But do note that democratic governments are unstable. Simple democracies tend to yield to tyrannies. (Both "tyrant" and "democracy" are from the Athenian dialect of Greek...and Athens oscillated between them.) A constitutional democracy was an attempt to stabilize it. Reasonably successful as such things go. But "plurality rules" voting was a major blunder. It needs to be "majority rules" so that the voices of those with non-central intrests are represented.

The potential benefit of monarcy is that the government will look after the long-term interrests of the country. Unfortunately, it doesn't have a very good track record in that regard. At least not when the monarch has been powerful. (Weak monarchs have a much better record in this regard.) The US government shows no more regard for the future of the country, however, than did Louis de Roi Sol. Perhaps less.

To make a sailing ship go you need both sails and a keel. (You also need a few other things that would extend the metaphor too far.) I.e., you need a propulsive force and a stabilizing force. If you lack either, then you are guaranteed disaster. OK. You also need a rudder, i.e., you've got to be able to steer a reasonable course. But governments tend to steer for increased control. Always. The only exceptions I can think of involve either incompetent hands on the rudder (which Britain was blessed with) or the collapse of the government.

If you grant the prior paragraph, then the obvious conclusion is that we need to decrease the strength of the sails. Perhaps the currents will carry us to a better destination. (Not likely, admittedly, but possible.) We don't want to destabilize things, as that yields massive fatalities.

But there are lots of problems with this simple solution. The main one is that it's not likely to lead us to any place better. But I don't think I can do anything better with this metaphor.

They worked in the factories because after the enclosure acts they would have starved to death on the farms. The cry at that point was "The sheep are eating the men!".

As you say "It's easy to scream "Foul!" when you drop context."

Actually, they weren't all the same people. The factories destroyed a lot of cottage industries, and many of those were the people put to work in the same factories. It was the same process of deskilling jobs, and then reducing the pay for those who do the jobs that is still going on. Yes, there are lots of benefits to people who are directly touched by any particular instance of it happening, but the people who *were* working in the field are treated extremely poorly with no-to-minimal recompense. And yes, it's still happening. I wouldn't say that it's rate has speeded up, because it's been so uneven over time, but we are approaching another huge increase in the rate.

But the driving engine for the factories was the enclosure acts, which basically threw people off their land, and gave it to the aristocrat that controlled the area. (Yes, he was already the lord of the land, but previously his power over the residents had had limits. Many of these were removed.)

I'm not going to praise the fuedal system, but that doesn't mean that the means used by the wealthy to increase their wealth and power were either fair or moral. Usually they were legal, if only because the laws were written by the wealthy and powerful. Which means that I don't count them heavily on the scale of "right vs. wrong".

Sorry, but I'm not convinced. That's the kind of thing that might be true, but which there is very little evidence to support. Perhaps it will become true when computer analysis of video imagry is good/fast/cheap enough.

OTOH, there's also the question of whether it should be done even were it to be shown to be effective. This is rarely debated except by people who implicitly assume that their view is the only reasonable one. There are, however, valid arguments on both sides, so such close-mindedness isn't very convincing.

You can call it a fallacy, but a lot of people starved to death for the benefit of the factory owners. They were protesting a real injustice. Perhaps their proposed solution wasn't an optimal one, but for them it was better than starving to death after being thrown out on the street.

Remember, history is written by the victors in class warfare as well as in military warfare.

There are several different people that are backing different approaches to that problem. Most of them are starved of funding, and also most of them are probably wrong. It would probably be foolish to bet that they are all wrong.

For that matter, a distributed hill-climbing algorithm would probably be able to solve that problem for any well characterized area. It would, however, likely require an amount of time approximately proportional to 2^n, where n is the number of dimensions, if you require an optimal answer rather than an almost optimal one. And n can get pretty large. And the constant of proportionality probably also depends on the problem. Note that this isn't the best current approach, it's just the easiest to understand. It's unlikely to be practical for any complex problem, though it works quite well on simple problems. There are many other plausible approaches, and many that I don't know well enough to evaluate. And in a few cases the basic approach seems reasonable, but the implementation looks rediculous.

I wouldn't bet that this problem will remain unsolved for 10 years. I'd make a bet at reasonable odds that it won't be unsolved for 20 years, or that it won't be proven to have been solved within 5 years. (It may have already have been solved by someone who doesn't have the resources to implement it on a large scale.)

Sorry, that's not a particularly plausible scenario. Not to say it wouldn't be dystopian, but I think Mack Reynolds had a much more reasonable take on it. Lots of use of tranquilizers, and bread and circuses. And "pretty much fake" rivalries between the great powers.

The one's I remember came out as Novellas in Astounding Science Fiction, or possibly Analog. I don't know if they were ever republished.

Please note, it wasn't a terrible life for anyone who didn't have one of the powers-that-be as an enemy, but it was depressing, with little hope for improvement. Every major country was a totalitarian state, though usually disguised as something else. And every country was building, or had built, a pretty rigid caste system. He didn't follow that out to it's logical conclusion, where the lower classes were fed brith control.

You're neglecting something. If the last of the current jobs are eliminated in 2045, which jobs will still be available in 2040, and how do you motivate people to prepare for them? Now try that same projection for 2035, 2030, 2025 and 2020.

Note that you will be WRONG about which jobs remain in at least SOME of your predictions. So are you going to compensate those people who guessed wrong about which jobs will be left?

It's not an easy problem, but AFAICT, the only way to even make a start at it is to have state funded university and graduate education available at a negative price. (I.e., if you can keep your grades up, you are paid room and board and get free education.) Then you adjust your estimates by adjusting the number of university positions that are available (based on academic quality, not on ability to pay). And get rid of "Publish or Perish", and schools using patents that they have filed to acquire funding. Patents resulting from state funded university work and research should be freely available for use within the country, and licensed by the government for use outside the country, with payment going to the government, not (directly, or in any tied manner) to the university.

Note that even this won't suffice to prevent massive civil unrest when over half the populace is out of work.

P.S.: Management will be the last job to be automated, because it's the managers who decide what automation will be used. Not because it's a particularly hard job to do better than most managers do. (OTOH, some managers do a pretty good job. I'm still not sure how much of what they do would be trivial if they weren't in political fights with all the other managers for status, however, and I expect that robot managers would cooperate better about that. I can't imagine that it actually leads to doing a better job.)

I'm rather sure that Watson, after suitable training, could write better C++ code than I could. I consider the STL to be an abomination, and avoid C++ at every opportunity. Given the choice of C++ or Ada, I'd choose Ada, though if efficiency isn't important I'd prefer Python3. (Actually, the compiler language that I usually prefer is D, but there are severe limitations in library support, so I often choose something else. But I'd pick even Java over C++.)