Posts by Doctor Syntax 40485 publicly visible posts • joined 16 Jun 2014
@X7
Thanks for the last paragraph - until then I thought your posting was ironical.
Such companies are in the business of protecting their customers from threats. Knowledge of such threats is a company trade secret. No successful business is going to share its trade secrets with anybody, not if it wants to remain successful. And I doubt the local state security organisation would ever consider reciprocating.
"Going against his own received wisdom of not buying stakes in companies he doesn't understand"
Received?
Disregarding that, presumably the crux is that the only judgement he - or anyone else - can wield is that they think they understand a company. The evidence in this instance is that he may have thought he understood it but didn't. This seems to have been an exception for him.
An exercise for the reader: what would markets look like if everyone really did understand all the companies all the time?
"The article states it was the "decommissioned" website that was decommissioned back in September. So if it was decommissioned before the breech, why wasn't the data removed? Or was it breeched before it was decommissioned?"
It might turn on what's meant by "remove". If the database is simply dropped the data's still sitting there on disk. The next DC/Cloud customer who acquires that partition takes a look and finds it. Not a problem if the disk was encrypted but if it wasn't...
I've seen this on leased kit - the previous user's data was sitting there. I've also spent time ensuring that borrowed kit got scribbled on before returning it.
"At some point during the genesis of the universe, there would have been equal amounts of matter and antimatter, and they should have annihilated one another."
We live on a planet that clearly consists of matter. I don't think anybody has suggested that odd bits of the solar system that come crashing into us are antimatter either so it looks as if the rest of the solar system is matter and probably the rest of the galaxy as well. But if one looks at other galaxies is there any way of determining that they're all matter other than assuming uniformity? Is there any basis for knowing that half the galaxies we see aren't entirely anti-matter?
"There are wide variety of independent observations which all lead to the same conclusion that there is some matter we can't see."
But don't these all add up to much the same thing - observations failing to confirm one theoretical prediction or another?
There also seem to be a variety of independent experiments that have failed to identify what it might be. At the very least there ought to be some consideration that the observations are falsifying the hypothesis or hypotheses on which the predictions are based. Or is it going to be a case of science advancing one obituary at a time?
"If dark matter exists (i.e. if our current understanding of gravity is correct) there must be about four times as much of it as ordinary matter."
And given that dark and ordinary matter interact via gravity then surely gravity should cause them to clump together. I should weigh even more than I do given that some dark matter should have attached itself to me.
"No - the thrust is that you can't just ignore the criminal responsibility when you push insurance liability around."
Just this.
The cyclists have a good point here. This seems to have been dealt with purely as a civil matter which can be dealt with by compensation via insurance (and surely no insurance company would ever try to weasel out of their responsibilities). But where a human driver is involved there is also a criminal aspect and if there has been criminal negligence then there should be scope for prosecution.
However I think they're wrong in not putting the blame on the vehicle manufacturer alone. It's up to the vehicle manufacturer to ensure that the overall package is correct.
Consider, for example, that the vehicle manufacturer picks up some image processing library. It was actually written for an image classification system where a few false positives or negatives were acceptable and for this the library was good enough. Or it was written for a medical application where it was to be used for cytology images. Would it be right to prosecute the author of a library that has been used out of its intended context?
Consider the possibility that the S/W itself is fine but the manufacturer has seen fit to run it on H/W with less resources than the S/W was specified for, or shared the H/W with another package when that wasn't intended. Why should the author be responsible for that?
"The thing about driverless cars is that they will almost certainly carry the means to analyse an accident after it has happened, and at that point it should be *very* easy to attribute blame, whether it is the software developer, the owner of the vehicle doing something stupid, or a cyclist or pedestrian doing something reckless."
Given that we're frequently told that AI results in not even the developers being able to explain why a given decision was made this might not be at all certain, at least not without restricting the S/W to designs able to generate a continuous log of the reasons for each decision.
"Pretty easy for Microsoft, and anyone else caught in this net, to simply not offer Skype and any other services that Belgium wants to regulate."
...and explain to customers why. Then let the Belgian govt. field their public's response. If it's a big issue then it'll be a big issue for the govt. If it isn't MS can quietly reintroduce the service.
"A targetted investigation requires a warrant or court order in the U.K."
One issue here is the issuing of a warrant by some agency other than a court. For instance if someone goes to Rudd for a warrant because someone's using hashtags she'll issue it without a second thought (in fact, without a first thought for obvious reasons).
"With UBI , you'd see the end of crummy employers treating their min wage workers as disposable scum, why, because they'd be able to go and find other low skilled work where the employers treat them as human beings."
There's nothing stopping those workers doing that now except the lack of sufficient better employers. I can't see why UBI should improve that situation.
"ensure that those hoarding resources pay and those adding value through brains or labour get rewarded."
Into which category do those adding value through brains and labour whilst saving for their old age fall?
In case you feel inclined to try thinking through again remember that most "hoarding of resources", as you put it, happens through pension funds - and experience has shown it's still not enough.
"lead has been used for plumbing for centuries, certainly well into the last century, and apparently the lead content of the water it carries in pipes is not that great"
The explanation I heard was that in the past houses were not centrally heated and were occupied during the day with intermittent use of water so the water didn't sit in the pipes for long and didn't get warm there. That ensured that the concentration of lead was kept relatively low. A combination of central heating and wives going out to work reversed this so the lead levels could rise but about the same time lead piping started to be replaced by copper.
"Amazons declared profits for this year were $200 million, though they announced a $13 billion purchase of another company."
That I found surprising as it involved borrowing money. AFAIK that has been contrary to Amazon's SOP; the reason for a low profit in relation to turnover has been that they've funded growth out of sales.
"But the vested interests fighting it will be as hard to overcome as Big Oil, Big Tobacco and Big Pharma."
Given your initial proposition - a completely automated factory with nobody with income to buy its products - why would they fight it? As you describe it it would be in their interests.
But what you didn't address is where does the money come from to pay the UBI so the toasters can be sold?
"the library has removed parts of the API" - that means you need to release a new version of the library - you can't and never should do this in a bug fix/maintenance release.
Unfortunately a maintainer can do just that if they want - and have been known to do so.
I agree - an update shouldn't break any program that didn't rely on an undocumented feature. A good start would be publishing the unit tests and the unit tests should have a condition attached: new tests can be added but old ones can't be removed.
But many library programmers do exactly the same thing, which can easily in the end lead to cascading effects. Because if there's an issue with a library then all the software which compiled against it will also be affected.... what would happen if there were any bugs discovered in your average compiler? Even compilers use libraries these days, and libraries which by themselves rely on others."
It's called entropy.
"consumers" (I dislike that word, why are we never "customers" any more!)
Looking at the context it seems they're dividing customers into two classes, consumers, by which they appear to mean domestic, and business which is fair enough. But actually TFA says "users" rather than "customers" so you may have a point. However my experience is that "customer" is actually an over-used word, especially when the likes of HMRC start using it.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
