Posts by Ken Hagan 8168 publicly visible posts • joined 14 Jun 2007
"Mind you, I'm capable of learning a new GUI, which sounds like it's a massive problem for some people.
It's an even massiver problem for the IT staff who have to support those people. Fortunately, if MS have *any* corporate direction right now, then it seems to be "baby steps every six months back to Win7". IIRC, there are two more 6-month cycles until Win7 drops out of normal support. They'll need to hurry up.
" I strongly suggest you follow the Linux Kernel Mailing List (LKML) thread a bit..."
Interesting. "what Andrew said" was that the rate limiting should be applied per-file-descriptor and this was in contrast to per-user. It was then noted that per-user would be more effective against someone who tried to get around the per-file-descriptor restriction by opening several FDs, to which Linus responded:
"I don't think we should try to protect against wilful bad behavior unless that is shown to be necessary. Yeah, if it turns out that systemd really does that just to mess with us, we'd need to extend it, but in the absence of proof to the contrary, maybe this simple attached patch works?"
And indeed it seems to work. Someone had one of the previously afflicted systems booting by Thursday. So it's all remarkably boring and grown-up and productive over there.
And elsewhere in the thread it is noted that the systemd people have fixed their side of the bug, too.
Two points in mitigation:
"You did not read the Reg article properly, you certainly have not looked at the linked material"
Well, I think quite a lot of readers don't look at the linked material. We rely on El Reg to summarise enough of it so that we have a balanced view of the situation without doing all the research ourselves. Thanks, at least from me, for the additional summary.
And in any case:
If the kernel can't protect itself against bugs in user-space programs, it isn't a very good kernel. Linus is free to have as low an opinion as he likes of the systemd people concerned, but he does need to change his kernel to address this. It's a DOS attack vector and if it was in Windows then we'd be queueing up to explain how it proves Microsoft's inherent shit-ness.
On the bright side, a similar cloud would probably be equally fatal for all the drones and robot soldiers they are building, which makes it less immoral than biological.
And historically, the evidence is actually in favour of developing such weapons. As a species, we've used all-out chemical weapons once (WW1) and thereafter only in pretty desparate conflicts where one side thought they wouldn't be noticed. Bio-weapons were certainly developed during WW2 but none of the sides were actually willing to use them for fear of retaliation. Atomic weapons were used once, when the US was certain that no-one else had them. As soon as that certainty was overturned, the willingness to use them (eg, in Korea) disappeared.
Slowly, the politicians and generals are learning. Our technical prowess makes all-out war indistinguishable from suicide. Therefore, all future wars will be fought with both sides pulling their punches and if one side looks like losing everything, it will stop pulling its punches and the "winners" will wish they hadn't.
Indeed. Let me join you in your cynicism.
If the article is to be believed, all the freed software is compilers and language tools. Given the maturity of this branch of software engineering (yacc and lex are as old as I am), I'd have thought writing a C# compiler was the least of your problems in trying to make C# or .NET useful on non-Windows platforms. Even if it weren't, Microsoft already give away a perfectly usable C# compiler.
Have they also released the extensive framework libraries that you need to do anything useful? Is this the same .NET that was pushed into the sidings with the announcement of WinRT a year or so back? Is there anyone at Microsoft who would be excited to be moved to the .NET team today?
Well, if you live such a healthy lifestyle that you fail to die of anything else (like, heart disease) then you will eventually die of cancer. This is organic veg we're talking about, not the freakin' Elixir of Life.
A more meaningful metric would be the risk of dying *early* of cancer. In fact, this would appear to be a general weakness of all "X gives you Y" type studies that end up in the popular press, but it is entirely possible that grown-up medical researchers routinely allow for this in some clever and standardised way that goes straight over the newspapers' heads and so never gets reported. Does anyone here know?
"Ah. Silly me (and one or two others) then, for spending all that time creating a responsive design for my web site."
If your website tries to look the same on all these platforms then yes, you've wasted your time. If it adapts to the target device and offers different layout, different facilities, different navigation, then you've done just what the man said and created different UIs for each case. Well done.
"Microsoft demonstrated a new Windows RT sync app that talks to some old database code using synchronous calls, but without blocking the user interface thread as synchronous calls used to do in the Victorian era (eg, 1995)."
I think you need to explain this a little more. As it stands, it sounds underwhelming.
"Most of the documents I've thrown at OO and LO have been well and truly mangled."
And if you are on the bleeding edge of the feature set then you'll have similar troubles moving documents between Office 2003 and 2007. The morals of the story are that Office formats are not a safe place to put your work, you need to stop using them, and it's only going to get harder the longer you put it off.
"I work at Nottingham Trent University, but any other UK university will be the same. As has been pointed out before in this thread, the majority of students use IT as a means to an end, mainly to write up their work and ultimately their thesis."
I'm rather surprised that everyone isn't just expected to bring their own device for such purposes. Back in the day we wrote up work with pen and paper and we were expected to buy our own. Students at secondary school are now expected to have access to a computer at home. (I don't know what the kids from deprived backgrounds do. I expect it isn't good for their education.) If you are paying several thousand in tuition fees, a cheap laptop is the least of your worries.
Now if there's some expensive software package that they need access to, that's different, but you didn't say that.
"The only problem is they need to get this released now, not in Windows 9 in 2015."
From the article...
"Myerson didn't say when this next Windows makeover would ship to customers, but he did say that Microsoft "will be making this available to all Windows 8.1 users as an update.""
So that's before Windows 9. I expect they will roll it out in six months time. They'll call it something daft. Everyone else will call it 8.3 (coz next week's offering is clearly 8.2), and *if* there is an API actually willing to admit to the true kernel version number (which is looking increasingly unlikely) then it will be something like 6.5.
"many apps would be crippled by being restricted to 32-bit (more and more *need* 64-bit to function)."
Christ on a fucking bike, mate! What are you smoking?
Outside of database servers, video editing and weather forecasting, hardly anything is *crippled* by being squeezed into 2GB of working memory.
"As HiDPI screens finally appear, Microsoft needs programmers to switch to APIs that work well with these displays."
Actually no. Programmers don't need to use new APIs at all. If you followed the guidelines laid down 30 years ago in the Book of Petzold, using GetSystemMetrics() and the like, the only thing stopping your XP application from scaling perfectly on a Hi-DPI system is the fact that later versions of Windows deliberately lie to you when you call these APIs. The "fix" is for you to recompile your application with a manifest containing GUIDs that were only published in the years after Vista was released.
"The article (please, read it) shows that it is [viable], even at the worst pessimistic scenario."
Er, no. The article's analysis is pretty flawed. People who are stuck with XP talking to old hardware have the option of isolating the XP boxes from the internet and carrying on as before, indefinitely, at zero cost, and zero risk.
The only people who need to pay for XP support are idiots in government who tethered themselves to IE6 and then went to sleep for a decade. Microsoft saw them coming and are charging three times Trevor's "viable" rate (initially, jumping to even larger multipliers in the next few years). MS will get their fee, too. The problem with the article is that there's no *larger* market for paid support if MS drop the prices to the levels suggested here (because you can isolate the machine and pay nothing), so there's no reason for MS not to gouge the small number of idiots for all they can.
"If it works initially then why should it not continue to work"
Quite, and I confidently expect isolated systems running XP to carry on working until the hardware fails. I've heard no credible claims that XP is going to stop working next week.
It's obviously different if you want to use your lathe to surf for porn on the internet. If that's what floats your boat, I suggest you get a new lathe with Windows 8 on it. (It'll serve you right.)
"Some devs have written a driver in a day."
But probably not in cases where the original hardware vendor either no longer exists, or no longer has any technical records for that particular model, or just wants to sell you a new lathe and is therefore unwilling to provide documentation.
And once your dev has reverse engineered the hardware spec, they are unlikely to be willing to guarantee the correct operation of their driver. At least, they won't be willing to sign a piece of paper that lets you recover losses from them if the driver turns round in a month's time and refuses to talk to the lathe that your business depends on. You might argue that the lathe vendor signed no such paper either, but you have a decade or more of experience to build your confidence in the original driver. The new one is a leap in the dark.
"If it does precisely the same thing in precisely the same way there is a prima facie issue of copying."
Or it is a pretty obvious idea with one particular expression that would be considered idiomatic by a large number of experienced programmers. I reckon quite a *lot* of things fall into that category, particularly if you spend time refining the spec so that it is mathematically minimal and then spend time refining your implementation to match, and then feed it to one of the fairly few compilers in widespread use, only for its optimiser to eliminate (in its own code generation style) the remaining differences between your code and the other guy's.
"If you bought a good old fashioned dead tree book written in English, would you expect to be able to translate it into a dead tree version in French for free?"
If you were French, it is hard to see how you could stop yourself doing just that. It wouldn't violate the author's (or the translator's) copyright unless you happened to render exactly the same translation, and they'd have trouble proving that in court.
By the same token, reading a book out aloud doesn't violate the copyright on the publisher's audio-book version, unless you have an audience that is wider than your immediate family. (I'm assuming the law doesn't prevent parents from reading bedtime stories. Perhaps that is naive of me.) Playing sheet music doesn't violate the copyright on someone else's CD. I could go on. Format shifting is an inevitable part of the personal use of copyrighted material.
The sort of Dan Brown parody suggested in the article wouldn't be purporting to add to the canon and wouldn't constrain what Mr Brown was able to do with his characters in the next real book. I don't see that it reduces his ability to make money from his creation or even de-values the existing books in the minds of true fans. His style has already been through the wringer of the critics and emerged in best-selling glory.
In fact, parody almost certainly means "you've arrived", in the same way that everyone using hoover, google, biro and portaloo without little trademark symbols is just something that grown up brands have to deal with.
You'd think so, but actually the economics has gone the other way.
When I write bad code, my paying customers suffer the bugs and whinge a lot and demand that I fix it at my expense or give them their money back.
When lawyers write bad code, their paying customers have to suck it up, or pay yet more money for another lot of equally bad code which might (by chance) have bugs that suit them rather than antagonise them.
Consequently, legal code appears to prefer no punctuation, long and rambling sentences running at times over several pages, and arcane vocabulary. The situation in IT, where the code has to cause the right thing to happen even when a dumb machine is reading (executing) it, strongly favours *lots* of structure (punctuation), short functions, meaningful labels for intermediates and even test cases with expected results where necessary both to ensure clarity of intent and correctness after subsequent modification.
In short, I find it *very* hard to imagine what the legal system would look like if it was implemented according to the almost-infinitely-higher standards that are commonplace in IT. And I'm one of those who don't think IT is yet up to the still-higher standards of mainstream engineering.
Words scarcely do justice to describing how totally fucked up beyond all belief the legal system is.
"But somehow I just can't see it working in Tunbridge Wells…"
I can. The sort of people who make a big fuss about Imperial units tend not to be terribly technically minded, so if you change the size of their "pint" or "pound" then they probably won't notice. You could probably swap their miles for kilometres and they'd be pleased about getting better mpg.
"But critics say that people are running illegal, highly profitable businesses by renting and then subletting many empty apartments at once - and avoiding hotel and rental-related taxes in the process."
Sooo, these people are engaging in illegal activity but registering their crimes on a well-known web-site first, to make it easy for law enforcement.
And it's worth $10bn.
I wonder if anyone has registered www.insidertradingtipoffs.com? It might be worth more than Wall Street!
"In this case how is anything a sign of intelligence?"
Once we have an answer to that question, we'll have taken our first baby steps towards AI. A related question is "What is intelligence?" and we don't have an accepted definition for that either. I suspect that once you have the answer to one then you have the answer to another, although I can't prove that since I don't have the answer to either and nor do I have sufficiently rigorous definitions of the terminology to reason about it. Nevertheless, different formulations may be equivalent and yet more or less useful than each other. I suspect that your formulation is more useful than mine, since "How do I tell?" is a call to action but "What is?" is merely an invitation to talk about it.
"I thought Kasparov used 'brute force' (ie evaluating every move to some depth) when playing his games."
Absolutely not. The number of possible moves grows so fast that a human could only evaluate 3 or 4 moves ahead by brute force. It's well-established that Grand Masters are better because they *don't* do that. At each move, they trim all the obviously crap moves and may pursue a promising strategy a dozen moves ahead.
Anyone who knew how to define "obviously" or "promising" in robust algorithmic terms could publish it in a book and live off the royalties. (There are plenty of amateurs who'd buy a book that really did teach them how to play at Master level.) The fact that no-one had done so prior to 1960 is why chess-playing was selected as a good problem for AI research. The fact that no-one has done so since Big Blue is my reason for suspecting that BB owes its success more to brute-force rather than cunning.
"What I do know is that we are listening to our customers and partners and innovating"
Interesting. Perhaps he should try listening to non-customers, like all the people who didn't upgrade to Vista, or Win8. It's rarely a good idea to confine your listening to people who like you or who are financially committed to your choices.
They had a big beta program for Win8. Thousands of people spent quite a while trying it, and both the Tech preview and Consumer preview got totally panned by just about everyone. I really don't see how he can argue that "we are listening".
"just as if Royal Mail suddenly announced they wanted to charge me for bringing mail to my door, because it's hard work"
Royal Mail already *do* charge more to bring mail to your door in a timely fashion. They also charge more for verifiable guarantees of delivery. In either case, it's up to the sender to decide whether it is worth the extra.
"We don't have many toll roads."
Well, we don't have many toll operators, but there is this big player called HMG who sell you "season tickets" to use their network.
In contrast, the phone network has lots of operators, but regulations have (so far) required them to inter-operate to the extent that each customer can just choose one as their point of access and thereafter reach anywhere on the network.
Then there's the rail network, where we have lots of operators and you can go take a running jump if you want to just use the train that's going your way.
Then there's the benefits system, which allegedly has only one operator but operates as though every last entitlement is separately administered by splinter groups from the Judean Popular People's Front.
There's lots of ways of organising a large network. The UK currently has examples of most of them. There doesn't (to me) seem to be any correlation between "good" and the ownership model. It's all in the regulatory framework.
"They have no incentive to increase service speeds and throughput because they'll get the same rates regardless."
That depends on your pricing structure.
If you only charge for an "unlimited" connection, you have no incentive to actually transfer data or do it in a timely fashion.
If you only charge "per gigabyte", you have an incentive to fatten up your pipes, but no incentive to iron out the wrinkles in your network that cause streaming services to stall.
If you charge per gigabyte and also charge for quality of service guarantees, you have every incentive to improve your network.
I have to admit I'm amazed at the quality of debate here. QoS is such a fundamental and obvious networking issue that IP and Ethernet both reserved bits in their headers decades ago even before they knew how they would be used. Since then we've had RFCs for specifying and agreeing on QoS, we've had similar "reserved bandwidth" built into USB and other standards. Whilst I could imagine that the lay audience wouldn't know this, and I can imagine that the Netflix guy chooses to ignore it because it doesn't suit his business model, I'm amazed that the El Reg readership doesn't understand that QoS (like truly unlimited connections) costs money and is only needed by a subset of customers.
Am I missing something? Are there big differences in ISP charging schemes between the different countries of the world? Perhaps El Reg needs to write a review of how each country does it, so that commentard confusion can be dispelled in future simply by linking to the article.
What is actually says in the article (here and elsewhere) is that Netflix are paying extra for a quality of service guarantee. That's something extra, that nearly everyone else doesn't get, so it costs extra, that nearly everyone else doesn't pay.
It's the difference between quantity of service and quality of service.
"Why should Netflix, who has no reason to deliver streaming video to you, be subsidising your internet connection?"
They aren't. They're paying for the extensions to the pipework to support their own business model. This is entirely optional. Had they opted to pay nothing extra, they'd get the service that everyone else gets (and is apparently satisfied with, at least to the point that they aren't switching providers).
If Netflix (or you) don't believe the ISPs financial figures, there's nothing to stop you building your own distribution network and making it available under a "fairer" pricing scheme.
"Fortune magazine's editorial judgment is now worth nothing."
I expect the editorial judgement was "Will this admittedly light-weight story pull in some readers and therefore subsidize the heavier stuff that we want to be publishing?". It's the same reason that El Reg publishes bootnotes. (At least, I think so. Maybe I've got it the wrong way round and the serious IT stories are just to pull in readers for Lester. I hope so. That sounds like a nice world.)
"Anyway. If all I want to run is Winamp, Chrome, Thunderbird, why do I need to upgrade from XP? It just works"
If that's your usage profile, you could stick something like Debian Stable on and enjoy full support, with far fewer worries about viral infections and probably a performance boost because you won't be running AV software that gobbles half your RAM and turns every disc access into an exercise in triple-checked paranoia.
Most people have at least one key application that ties them to XP, but if you don't then it's worth switching.
"Alternatively, I guess we could import all the workers and have no kids, but that isn't economically viable."
Isn't it? Last I looked, there was no shortage of people willing to be economic migrants to the UK. The main objection appears to come from those "kids" who were born here. Remove all them and you could presumably open the floodgates to mass immigration.
Of course, this may be a "cheap but not cheerful" solution. For one thing, it appears to require a sex ban on the entire adult population, which is unlikely to go down well. Still, maybe there's some sort of drug...
I understand your confusion, but would make two points.
Firstly, it is a very bad user shell. It has almost zero discoverability and has dropped various features of its predecessor that many people used. These users are therefore left with the double whammy of "it doesn't work anymore" and "there's no clue for what does work in its place".
Secondly, those sysadmins and support techs are exactly the ones who get all the grief from the users who have been dumped in a new ecosystem. I bet Steve Sinofsky never spent time on the helpdesk explaining to grown adults why they should have to relearn everything they know about the computer simply to carry on getting their job done.
"The only differences are in the CPU area: PAE (for RAM > 4GB on 32-bit systems), NX (for security), and SSE2 (for graphics, primarily). I believe some if not all of those are actually required by Windows 7, but just not listed explicitly on the page."
I'm running Win7 on a machine without NX. I'm almost certain that PAE and SSE2 aren't necessary, but won't swear to it. There are applications out there that demand SSE2, mainly because SSE2 marked the point where one could dispense with the old x87 FPU entirely, so for many folks it is just a compiler switch.
It is probably worth pointing out that all new processors announced by AMD and Intel since about 2003 have had all three. They may have continued to sell older designs for a couple of years, but I'd be surprised if *any* machine bought new since 2006 doesn't have all three.
All neutrinos are travelling at some ridiculous number of nines of the speed of light. I don't know if I ever knew why interactions are more likely with higher energy neutrinos, but I've certainly forgotten if I ever did. It's probably complicated. Perhaps someone will jump in here...
"Isn't XP64, by any other Name just Server 2003?"
Only in the same way that XP is Server 2003.
So, no, XP64 is a client OS and it goes out of support next month. Wags will argue that its driver support was so poor that it never came into support in the first place, but it is still rather sad to learn that even Microsoft have forgotten about it.
"As such, Berners-Lee was less interested in presentation and more interested by data – both storing it and discovering it. The story of the web, however, has been dominated by how it is presented – how web pages look."
There's this little startup called Google who have made quite a splash with "discovering it", you know.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
