Posts by Ken Hagan 8168 publicly visible posts • joined 14 Jun 2007
"Saying that case insensitivity is good because Pascal, Ada, FORTRAN and Basic have is, well, a bit of a weak argument."
Editing "Lisp" and "Eiffel" out of the list is a bit of a weak counter-argument. It looks like you feel the argument would have been stronger with them in, so you took them out in order to mis-represent the other guy's case.
"and 100mbit fibre to the home is a reality"
In some areas.
And specifically to the home.
Meanwhile, SATA 3.2 is also a reality, works both ways, is 160 times faster, and is on an uncontended link. This consumer will be keeping his stuff in the PC as a matter of course for the time being, thanks.
Probably, and that's probably what the lawyers who reviewed the code of conduct were thinking when they signed them off. The clever bit is for the conference organisers to realise that they have this power and that being too specific in their code of conduct probably only gives the miscreants wiggle room.
A "brought-in Quality Manager" you say.
That's an interesting idea.
What do you suppose their motivation is, given that they have no financial interest in the company and the companies that they prey upon are self-selecting for "too fucking tight to hire permanent staff and too fucking dysfunctional to have any inherent quality"?
It is also quite possible that the new Kotlin apps are, for those writing them, the second time they've written an app like that (because this is the Kotlin re-write) or that those writing them are just better developers (because they've had the time and inclination to learn another language).
Similar remarks apply to (nearly?) every comparative study of the effects of language choice on code quality. It is actually extremely difficult to design such a study and remove the effects of people, experience generally, experience with this particular type of problem, and difficulty of the problems under study. I've never seen it done and I'm not sure it can be.
"After all, Hitler WAS elected..."
But not by a majority.
And he would have fared no better than his immediate predecessors in trying to maintain a government had he not done something that they didn't --> mount a military coup against the country he was supposed to be running.
"How do you make a Windows NT server 4x quicker? Stick Netware on it."
But in doing so you make it 4x more expensive. Not that NT was cheap, but Netware was pretty pricey and the difference was enough to pay for a substantial hardware upgrade.
"Either follow your country's laws and risk your status as a registrar or follow ICANN's regulations and risk getting sanctioned by your country's legal system."
Well I'm not sure I ever followed the argument here. It seemed perfectly plain to me that the ccTLD belongs to the local government, not ICANN, and that if ICANN wanted *any* influence over how it was run then they would have to have a legal presence in the country concerned, at which point they are subject to local law.
Obviously it is open to ICANN to simply punt and give up on a ccTLD. I think you'd have to say they've done that in a fair number of cases. (Can you really see ICANN dictating policy to China, Iran, Russia or North Korea?) The much-feared balkanisation of the internet has already happened and the sky remains defiantly above our heads because ICANN don't control the technical standards and those are what matters.
"since they'd all still have to point to ICANN roots for .com, .net, .org and country specific TLDs they'd be the lowest common denominator so that's all anyone would use. Which defeats the purpose of this hypothetical engineer led revolt against ICANN..."
On the contrary, I rather took this to be the point of the exercise. But on the other hand, DNS is a hierarchical system, so these hypothetical engineers can already configure their own systems to drop all the gTLDs on the floor.
"I can see that .catholic is a strange one because that religion is centered around what the Vatican (which is a separate country in and of itself) says and does. "
Nit-pick: what we in the West usually refer to as the orthodox church actually lays claim to being the catholic church. The word just means universal and consequently both sides of the Great Schism believe they are the *legitimate* claimants. This is also why the term /Roman/ Catholic exists.
Not strictly true, since a properly audited and distributed system would make it immediately clear to everyone that a particular named person *had* deleted particular named people from the database.
But I up-voted you anyway because there's no fucking chance of gov.uk doing the job properly.
"If you can't trust the government, you frankly can't trust ANYTHING and should be seriously considering renunciation..."
If you can't trust the government, you should adopt a constitution where they don't have all the power. Pretty much all of the world's long-term democracies figured that one out years ago (and in nearly every case it is the only reason they are still democracies).
Let's say I'm going away on holiday. When I come back and turn that computer on, I don't want the very first thing it does to be "notice the date and suck several gigs of shit down my wire whilst berating me for leaving the upgrade so long".
7 days is not enough. 700 days is not necessarily enough. If you feel that you really have to carry on nagging, then do so, but don't actually wrest control of *my* property from me.
"Don't see why I should have to pay for this, but ..."
You shouldn't. If Windows was completely free then you'd have no claim, but it isn't. You paid a (small) fee to have a product on your PC and apparently that product is not "fit for purpose". In the case of a PC operating system, "fit for purpose" means it is safe to leave it connected to the internet without it automatically getting fucked over by the vendor who is offering a range of licence upgrades protection schemes whereby you pay them money and they stop doing Bad Things to your computer.
You should seriously consider suing Microsoft for the costs. It is *your* computer but it is *their* decision to make upgrades compulsory and *their* choice of timing and *their* updates that turn your working system into a not-working system. Either they hand over the tools to block these updates or they pay for the damages. Sure, you could prevent the problem yourself by pulling out the network cable, but then it isn't the useful PC that you paid for.
It is fucking ridiculous negligent that MS apparently have the right to bork *your* computer on a six-monthly schedule, do not apparently feel the need to provide an off-switch and still get off scot free by saying "sorry, but computers, you know .. hard".
"Not sure if your being sarcastic or not..."
I'm not. I take your point about sandboxes being permeable, but my point is that if your sandbox is permeable then it was game over for you as soon as you started surfing the web. Hardly anyone runs trusted code in their browser. It's all "whatever the web-site feeds me". There is nothing in today's story that makes this any more scary than it was yesterday. I think it is unfair to pick on the Node.js crowd.
"That sentiment poses a particular problem for the Node.js community, where developers often rely on dozens or hundreds of code libraries (each of which may incorporate other libraries) written by someone else."
You mean ... developers often rely on links to dozens or hundreds of code libraries that can be modified after the fact by someone else, so even if they weren't a problem when the software was written they could become one if an attacker so chooses.
But this isn't actually a problem, because from the point of view of the end-user who runs the code, all JavaScript is untrusted code and therefore runs in a sandbox as a matter of course. (Well, OK, not quite all if you are the kind of person who has locally maintained apps written in JS. But I think that makes you rather unusual.)
"I mean who in their right minds, in the 21st century, comes up with a programming language that is whitespace / indentation sensitive."
Who in their right minds has ever intentionally written code that doesn't get the indentation right?
How did they get this through lint? (First seen in in the 1970s, I think. Something similar presumably exists for newer languages.)
Why are they still a programmer? (The opprobrium of colleagues should make them quit if nothing else.)
Which of their (now former) colleagues isn't using an editor that can sort out the indentation of an entire file with a single command?
"So google could just say they will stop updating android, lock down the code, encrypt it and no longer issue any new updates to non google made phones."
They could do that with the Play runtime, but everything below that (Android and Linux) is FOSS so they can no more lock it down than I can.
"support turned off the error reporting"
If true, then I have to ask "Why is it possible to turn off the error reporting?". Ignore it, by all means, but when the shit finally filters through those fan blades, the error reporting is what distinguish a problem caused by the end-user from a problem with the kit. So why would the vendor *let* the end-user remove such evidence from any future investigation?
"Probably better to invest in an external "power bank" if you find the power running low"
If I was happy to carry around an external battery, permantently plugged into the phone because the built-in one has worn out and I'm too cheap to buy a new device every other year ... then I'd prefer no battery at all inside the device.
"Social media is private space, and its owners refuse to take responsibility for policing it."
There's a story somewhere today that Farcebook "moderators" are deliberately keeping offensive material online because they believe it would hurt their bottom line to censor it. If that is the case, they are making editorial decisions, which makes them liable as publishers, and have admitted a financial incentive to publish material including hate speech and physical abuse of children.
I'm pretty sure *I'd* be in deep shit if I published such material on my own web-site and I'm pretty sure that "but this is how I make my money" wouldn't be accepted as an adequate defence.
"He didn't state it as a defence."
No, his actual defence was to bet a signed dollar that it was true.
If I were the victim, I'd seriously consider taking up the bet and then going to court to obtain the dollar (plus costs, obvs). It might be worth a fair bit if Mr Musk fails to prove his case.
"Hey, let's make this an "accessibility" issue, start a major *STINK* over it, and *FORCE* web designers to make it all "old eyes accessible" - heh"
Given that most web designers seem not to have heard of "mobile phones" as a form factor, I doubt that concern over "old eyes" will get any traction, unless you can find a lawyer to back you up. And I feel rather uncomfortable with encouraging the idea of "UI design by lawyer".
So you look it up and discover that the owner is DrEvil@gmail.com. Does that actually help you with your leeal trouble? No. You have to dig deeper to get to someone you can sue and for that you will probably need the help of your local legal system. Funnily enough, your local legal system won't find that GDPR stands in their way.
"Yes, inconvenient in some languages I am familiar with, as are the missing accents. No, not a deal breaker"
I think you'll find that for Joe Public, not being able to write your own name "properly" is a deal breaker. This really can't be handled with a "Let's all be reasonable and do it the American way" attitude.
"Given the this is a UK site, I image that the vast majority of the readers are from Europe!"
Au contraire, mon ami. Since this is an English-language web-site, most of the users are probably from just about anywhere. The vast majority of English native-speakers live outside the EU.
"If you don't connect it to the network, you don't have to worry about whether or not you get security updates."
Except that with the assumption of connectivity has come the assumption that the vendor can ship any old crap in the first manufacturing release and patch it later, so there is a fair probability that your TV won't work properly if you never give it a connection.
"I've mostly see 98 only as a deployment platform for customers, and never loved it."
Me too. I never understood the claim that NT4 needed more resources. If anything, it was the reverse: 9x was a complete dog in less than 16MB.
It was also a dog's breakfast internally, so if you had any say in your development platform you'd dual-boot. Get the application working on NT, where you had a debugger worthy of the name, and then test on 9x to find the platform dependencies. (With a little experience and care, those were pretty minimal.)
I'd adopted a similar approach a few years earlier with 3.x. I would build the 32-bit version and get it working on NT. Then I'd flick some compiler switches and test the 16-bit build on DOS (er, I mean, Windows 3.1x). Again, with a little experience, you learned the safe subset and this process was pretty well-behaved.
Basically, for most of the 90s, the trick was to do whatever it takes to avoid having to develop on DOS-based Windows. Happily, eventually our userbase realised what we had known for a decade or so -- DOS was shit and NT was both faster and leaner than they had read in the magazines, as well as being several thousand times less crash-prone. (That's no exaggeration, by the way. NT 3.x blue-screened on me about once in ten years. It's DOS cousin would usually roll over at least a day. You can do the maths yourself.)
"We forget just how painful Word was before the rise of PDF; yes Word allowed you to do object embedding, only problem was anyone else wanting to read your Word document and view all those Visio diagrams, etc. that you had so carefully embedded, had to have the relevant applications installed on their system. Obviously, you could paste-as-picture/image but that made updating a pain."
Interesting. The OLE rules actually said that embedded objects had to offer a rendering that did not require the relevant application and that containers had to save that rendering as part of the containing document, precisely to avoid the problem you've just described. I would say that it was almost impossible to actually program either the server or the container application without being aware of this, so it is interesting that you found yourself using a version of Office or Visio that had managed to screw this up.
(Update: It's probably also worth mentioning that the OLE libraries provide this capability for free. It requires conscious effort on the part of the programmer to avoid caching a graphic rendering. So that's someone going the extra mile to tick the box labelled "be annoying".)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
