Posts by I am the liquor 427 publicly visible posts • joined 9 Oct 2013
I think hoola has a point though. GPs were always "privatised," but they were not always "privatised in this way." There's a world of difference between an LLP owned by the GPs (or dentists) themselves, and a large corporation serving hundreds of thousands of patients with the cheapest service they can get away with.
They ought to be measuring the overall performance of the system, including the human, as long as the human is still part of the system. If the AI never crashes, but the human crashes more when they're forced to take over, then the AI hasn't achieved any safety benefit.
As far as I can see, there is not.
The Commission has reserved the right to executively alter the specifications "in the light of technical progress." But that looks rather like a chicken-and-egg situation, doesn't it. They won't allow a new connector until someone's created one, and no-one's going to create a new connector that's not allowed.
Enshrining a specific connector in law seems like a mistake. We're now stuck with USB-C for good or ill. A better idea would've been to allow any connector that meets some standard of non-proprietary openness.
The Quakers took it to the extreme, certainly, but a lot of Victorian industrialists shared those paternalistic ideals to some extent. When I first entered the world of work, most of the biggest employers nearby were companies that had started as family businesses in the 19th century, and they all had really nice sports & social clubs, with sports pitches, bowling greens, bars and function rooms... one even had a rifle range. All of those are gone now, sold off for housing development.
Well good news John, P455w0rd36 is in fact not in any password breaches known to haveibeenpwned.com.
Too bad it's been burned now!
But yes, as Len described, once a password has appeared in a breach of plaintext passwords (probably pinched from some site that stored them in the clear or used weak hashing), then it's in every password-cracker's dictionary. If they lift a database of password hashes from a site where you used one of those passwords, then they will decrypt it via a dictionary attack, even if an expensive hash function was used in that case.
Apparently haveibeenpwned has a dictionary of "hundreds of millions" of exposed passwords. So in cracking terms, any password that's in that list, no matter how long or complex, is reduced to about the level of a 5-character alphabetic password (380 million combinations).
The real answer is for the sites not to enforce composition requirements, which has been the official CESG/NCSC guidance for some years. Specify a minimum length - nothing more in terms of composition - and then check the password against a dictionary of known weak or compromised passwords, is what they should be doing.
I think that's a good call. For all the shortcomings of languages like JavaScript or of inexperienced programmers, they're secondary issues; I'd agree the number 1 barrier to maintainability of systems is choice of dependencies.
Everything has dependencies of course, but if you've stuck rigidly to "dinosaurish" dependencies like, say, Oracle or MS .Net, or something defined in an ISO standard, then the chances of your code still running in 10 or 20 years are good. If you've machine-gunned your system full of random stuff off NPM or NuGet, then chances are not so good.
It's fine that the code quality is terrible. A couple of years into the future, the framework that it's all written on is no longer flavour of the month, and most of the hundred other dependencies that were dragged in are no longer maintained at all. What's the point of building maintainable code on top of such a house of cards? Consider it disposable from the start. Throw it away and build a new one. Building code that'll be runnable and maintainable for a decade or more is such a 1990's concept.
Ponzi scheme sounds about right.
As long as we're not all smears of carbon on the pavement, it's doing what it's supposed to... or, at least, you can't say it's failed. I hang old AOL CDs round the garden to stave off the nuclear apocalypse, and to be fair, the empirical evidence for the success of that strategy is equally strong.
If there even is a standard that covers this, which there probably isn't. More likely Mazda's developers just made a wrong assumption that a JPEG file name will always end with .jpg because they usually do. Presumably other manufacturers' software was displaying them correctly.
Great idea. Wait for your data sharing agreement to be struck down by the courts, then just repackage the same old shit, with a few ineffectual changes and a different name, and you have another 2 years of business-as-usual while Max Schrems chips away at that one. Seems like that could work indefinitely, or until Max is worn out, whichever is the sooner.
Pretty much every software business error that Fred Brooks wrote about in the 1970s is still being made more often than not. Including the one in the title of the book. Most project managers I've mentioned it to over the years had never heard of The Mythical Man Month.
We're in a business that likes to think of itself as fast-moving and forward-looking. I suppose it's inevitable to have an element of being doomed to repeat the mistakes of history.
An old friend used to recount a story about accidentally driving his armoured vehicle over his rifle, rendering it suitable only for shooting round corners. Fortunately he was in enemy territory during a hot war (Iraq, the early 90s one), and got issued a new one with no questions asked. Apparently there'd have been hell to pay if he'd done the same on Salisbury Plain.
The terminating telco doesn't need to immediately block the route; they pass the info upstream and wait for the originating telco to cut off the subscriber. If the originating telco doesn't do so, then whoever's downstream of them has to cut them off. And ultimately, yes the terminating telco in the UK would have to cut off the inbound route, if the intermediate telco on that route has not dealt with nuisance calls upstream of them.
Clearly none of them are going to do this voluntarily, especially when they're all taking their cut from all these calls. International treaties and legislation will be required. Or at least the threat of legislation, if the telcos can't collectively get their house in order.
If it's ITU rules that are preventing anyone addressing this problem, then maybe the ITU is the organisation that should be tasked with solving it.
Back in the days when carpal tunnel syndrome seemed like a major health crisis, having your wrists higher than the keyboard was reckoned to be the safe way to type. You could get a squidgy wrist rest to put in front of your keyboard to raise your wrists to the correct angle.
Now everyone uses laptop keyboards, with this built-in wrist-rest design, and you hardly hear about carpal tunnel syndrome any more. Coincidence?
I think when Matthew said "it’s not something you instinctively want to use," he really meant to say "it’s not something I instinctively want to use."
Clearly the reason they still have it is that there are enough customers who will never buy anything other than a Thinkpad because of it.
You can still do normal PAYG, you don't have to buy the 1-month bundles/add-ons. Your cost per minute will be higher now if you're on 3, but you can still choose to pay just for what you use, subject to a minimum usage level of one chargeable event per 180 days to keep it alive.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2023
