Posts by Stephen Usher 42 publicly visible posts • joined 13 Jul 2007
It's even easier than that. They merely have to design a new processor tray and leave everything else alone.
Of course, they could help by asking AMD to just put some new firmware in a more modern version of their Radeon HD cards so that the current users can gain access to graphics card technology less than three years old as well.
Of course, the biggest problem with NTFS (and FAT) is that you can't rename or move a file which is currently in use, and this update won't fix that problem.
"Why is it a problem?", I hear you ask.
Well, if you're doing something such as a software update and you need to replace a core library with a new copy on a running system you can't. You need to reboot the system into a state where this library isn't being used and then replace the file. If, however, you were able to move the old library out of the way, still being accessible by the programs/systems using it, and then replace it you would then not need to reboot the whole system, merely restart those services/programs which use it when it's convenient. i.e. no down-time.
The problem I see with the RaspPi is not with the hardware, or even the software, such as it is. No, it's the lack of a single, bound, dead-tree manual, as they had originally envisaged, but dropped.
You see, the reason for the burgeoning of the programming teens in the early '80s only had relatively little to do with the actual hardware (other than the ability to annoy their parents with stupid noises within 5 minutes or annoying their sister by plastering their "boyfriend"'s name on the screen in multi-colours) but all to do with the excellent manuals which came with the devices. These lead the user through the basics of BASIC and on without talking down to them. They also had the teasing technical information at the end, which acted as a goal to understand.
Now, if you look at the late '80s, even the ZX Spectrum had lost its programming manual and what came with it was little more than a pamphlet telling you how to load a game. The bedroom coders stopped being inspired. The hardware hadn't changed, only the documentation had.
So, in my opinion, it would be far better for the Raspberry Pi Foundation to get a good manual writer on board, make the default boot environment a stripped down Linux kernel with a BASIC or other interpreter directly on top (which has to have simple sound creation and sister annoying ability within 5 minutes) and sell the thing as a bundle. Possibly sell the book with the hardware as a freebie which comes with it.
Systemd is seriously bad joo-joo.
It's Apple's launchd or Sun/Oracle's svcadm re-implemented stupidly with cron rolled in.
Booting quickly is less important than booting correctly, especially if you're running vital services.
From what I've seen so far, systemd may actually be ready for the big time in about five years time when it's been overhauled, but I won't hold my breath.
I've been using Lion now on a bunch of machines since release and, to be honest, it's a bit of a Vista.
It's far slower than Snow Leopard on the same hardware and the re-implementation of the Finder is clunky. Mission Control takes Spaces and removes the ability to use spacial memory to locate your applications.
It's a bit of a road crash really.
The only up-side is the increased intrinsic security against buffer overflow attacks (which is probably why it's hugely slower and more clunky than Snow Leopard).
Whilst reading the review I was getting more and more interested in the game, right until the four boss levels were mentioned. I then lost all interest in the game.
For me boss levels totally detract from a game such as this and feel like an artificial barrier added by the developers. I which they'd stop thinking that they need them.
There's lots of very early (late 70s) machines missing from there, most notably the "Grandstand" ones I remember in Dixons and Debenhams, and the early "Battle Chess" console.
These machines had a fixed set of games within them so I'm not sure if the collector would class them as consoles.
Actually, since a little known update in the law a few years ago, being naked in public is tantamount to being a criminal act.
Basically, the law was changed so that you could be done for knowingly allowing your genitals to be on view so that it may cause alarm or distress. It doesn't matter is no-one is there to be alarmed or distressed or if anyone is, just that there could be.
Seeing as the first ARPAnet node in the UK was based in the basement of the Computer Science department at UCL, this was probably the first to get a DNS entry.
From what I remember being told in the late 80s, it was in a secure location as ARPAnet at that time was still a secure network, at least for nodes outside the USA.
Hmm.. You seem to have missed the ZX Microdrive and the ZX Interface 1 when they were on sale in Boots and W.H.Smiths in 1984 then... I hope that you weren't too disappointed. :-)
As for the bike. Well, journos (and nay-sayers) always like to knock the different. As someone else said, it's basically a recumbent bike with some weather protection and a small electric motor. Just because it doesn't look like a "safety bicycle" doesn't mean that it is intrinsically bad. They said the same thing about the "safety bicycle" in the 1880s because it didn't look like a penny-farthing and that "it would never catch on."
*Would have put an Uncle Clive icon on the post but the Reg doesn't have one yet, either saintly or devilish.*
You know, I'm not sure Apple can actually get away with the condition in the contract which states that they can ban an application for any reason even if it meets all the requirements spelt out in the terms and conditions, at least under UK law, where a contract has to be "reasonable."
Maybe it requires someone with very deep pockets to challenge it in the courts?
The biggest problem for Windows patches, which force the dreaded reboot, is the fact that neither NTFS nor FAT can handle renaming currently open files allowing their replacement with newer versions.
This means that all of those DLLs and kernel driver files and all the parts of Windows necessary for the minute to minute running of the system can't be replaced on the running system. Instead they need to be scheduled to be replaced before they get used at the next reboot.
If NTFS/Windows were modified to be able to cope with renaming an open file (with the programs which are still using it finding the original version), as it is possible to do on all UNIX derived filesystems, this would mitigate a great many of the pesky reboots. (As would Microsoft creating their consumer patches as bundles, bringing systems up to the current patch level from any previous patchlevel rather than having to iteratively cycle through the patch history, each requiring a reboot.)
I'm going to ignore the fanbois and the trolls....
This bug looks to be highly intermittent, probably requiring a set sequence of events, which no-one has yet documented fully, to trigger. It probably requires a "perfect storm" event.
Just because this sort of bug is old doesn't mean that it's any easier to find and may be so illusive that even setting developers on the prowl for weeks trying to find it and fix it will do no good. So, I don't fully blame Apple, having software development and bug fixing experience myself, all be it on a smaller scale.
This is not to say that the bug isn't important, or even critical. If the cause can be found no doubt Apple will produce a fix, eventually.
Unfortunately, it seems that Apple change the library format with every major release of iTunes. Have you not noticed the "Updating Library" dialogue box?
As I see it, both Palm and Apple are being boneheaded.
Apple should see this as an opportunity. The more devices which sync with iTunes the more revenue they'll get from the iTunes store. *Ka-Ching!*
Palm should play by the rules. They should also have tried to go into discussions with Apple over trying to get an agreement to sync their device.
No analogy is completely apt, however, to answer your questions:
It you're using you neighbour's WiFi signal from inside your house you're still treading on his/her digital daisies as you're using his/her network the other side of the WiFi connection, even if you don't go onto the Internet from there.
As for the signal leaking over your physical boundary, you've not put your microwave gates up.. You should paste tin-foil on your walls. :-) Do you worry about mobile phone signals infiltrating your property even when you're not a subscriber, or even Sky TV microwaves from those satellites over the Indian Ocean? Maybe you should charge them? :-)
The problem here is one of implicit or explicit permission to use a service.
Does not configuring a wireless access point (due either to negligence of ignorance) with security implicitly give permission for someone to use that access point and the network beyond?
Let's use an analogy:
If you don't put a gate at the end of your drive is it right for anyone to enter the property and start playing in the back garden, even if they do no damage whilst playing there?
I would say that this is not right, it's trespass. Hence, using an open access point without explicit permission is also trespass (in all intents and purposes, but not legally).
Now, if the person with a WAP/Garden wants to put up a SSID/sign giving permission, then this is fine and the network/garden is anyone's to use.
(Badgers, 'cos they like cans of nice, juicy worms.)
Hmm.. maybe they should use Victorian values, Anon-Coward.
I can see bathing machines being all the rage, especially if combined with a roller-coaster.
Of course, the ladies and gentlemen would have to be segregated and the ladies and gents would also need to wear long swim-suits.
Apple didn't stuff the OS, those people who used CPAN on the system version of Perl did.
As noted by Christopher E. Stith above, no experienced admin would *EVER* mess with the system installed version of Perl as the vendor add-ins *WILL* break if you do. This is the same on all OS distributions, be it MacOS, Linux (or any flavour) or Sun Solaris/OpenSolaris.. or, indeed, Microsoft Services for Unix.
If you want a customised Perl (or Python, or <add your interpreted language here>) then you build your own version and customise it.
From the look of the errors they could happen in *ANY* procedural programming language (and probably some OOP ones as well.
Let's see, using different lengths for an array between proceedures... Hmm.. where do I see that happen most often?
C? Yes, but not that often, unless it's using static arrays.
FORTRAN? Oh yes, indeedy! Very often.
C programmers have to think about the memory allocation and hence are generally more careful (but not always). With FORTRAN(77) it's all done for you and the programmers don't worry (until their programs SEGV).
As for the second, that can happen in *ANY* language, including all those nice OOP ones which are the darlings of the safe-programming cults.
You lot are being a little harsh. If you know nothing about the subject then it's easy to fall into pitfalls.
(1) She couldn't get her ISP connection to work.
Seeing as she is obviously totally non-technical she would have to rely upon her ISPs automatic set-up program, which would only work under Windows.
Setting this up under Linux is simple IFF you know about TCP/IP networking, for anyone else it's gobbledygook, especially if it's PPP. (Would you like CHAP or PAP with your order, Sir?) That's if her ISP's modem is actually supported under Linux.
Sometimes technical people forget how complex and problematic "simple" tasks are and how much background knowledge is required.
(2) She was only following orders.
When she gets a Microsoft Word document, unsurprisingly she thinks that the only way to use it is with Microsoft Word. Remember, in her world, she's been told this and no other option exists.
Maybe double-clicking on the file would have opened it, but a great deal of the time this is not how non-computer literate people operate. Instead, they will open the program and then hope that they can find the file. After all in real life you'd not tap on a tin to get the tin opener to magically appear to open the tin, you'd go and get the tin opener first and then use the tool on the tin.
As for the anonymous coward who suggested that "Really, Linux isn't that hard... you may have to search online for some answers if you don't know how to do something". Well, firstly she had no net connection and secondly how would she know what to look for or even how to decode the technical information if she had?
I nearly coughed when I read that it used MySQL as a database. Why?!
For a music database (which is most likely to be used by one person at any one time) such a database engine is severe overkill and a potential security nightmare (as the audience for a music player won't known how to secure MySQL properly).
Surely, using a small, light, single user database engine would have been ample?
As for KDE 4, it looks pretty. I've got a Mandriva 2009.0 box running it. I wouldn't call it production ready, there are too many every-day things missing. Maybe in a year or so, when the basic functionality's been written I'll go beck to it.
I keep reading all the hype about how downloads will kill the uptake of HD physical media in the home. I can't see this happening in the short or medium term, at least within the UK.
There are small islands of high bandwidth data connection to homes currently, mostly within restricted metropolitan areas. This bandwidth is necessary for reliable HD download which doesn't look like a patchwork quilt due to compression artifacts. This isn't going to expand much more widely or that quickly.
BT's Fibre-to-the-Cabinet and 21st Century Network roll-out isn't likely to be anywhere useful for at least a decade, so large chunks of the UK will be fixed at <2Mb/s for the foreseeable future.
In other words, downloads are, and will continue to be a niche market.
As for Blu-Ray. You can blame two things for the failure to take off quickly. Firstly, onerous licensing terms and a death-grip DRM system. Secondly, most films don't look much better in HD that they do in upscaled DVD. Indeed, the films at the cinema seem to be *LOWER* resolution and grainier than DVD, at least at the multiplexes.
That's not to say that Blu-Ray won't be a success in the end. It took DVD a very long while to become mainstream, at least 8 years, and that was only when players cost less than £100 and the media was on the "3 for £20" shelves.
Of course, there is another way that Ofcom do do all this, which has already been done in the telephone and utility industries... disconnect the content providers from the channel provider. i.e. in this case split Sky's broadcasting part off into an equivalent of "BT OpenReach" or "British Gas Transco" and make it pay the same amounts to the content generation part (which would produce Sky 1, Sky Sports etc.) as the other broadcast channels, be it Virgin, Freeview or Freesat.
In this way delivery is separated from the content and a true level competitive playing field is formed for the feed into the home.
Actually, he's not an idiot, from a certain point of view....
Basically, if you're looking from the point of view of grabbing as much money as possible and milking the cattle^H^H^H^H^H^Hcustomers for as much money as possible, which is the American capitalist way after all, then he's dead on the money (in more ways than one).
If, on the other hand you're looking from the societal and customer viewpoint then he's being evil and selfish.
Truth is a three edged sword.
From what I've read on the forums links to various articles about this uproar the biggest gripe has been that the games developers have to write lots of different code so as to find the fastest way of implementing what they want to do depending upon the graphics card due to driver differences.
In fact, it seems that the biggest problem that they have is actually a driver implementation problem rather than an inherent problem with the OpenGL specification itself. i.e. the vendors have broken implementations and OpenGL is getting the flack. (The vendors probably don't actually test the OpenGL code as much as thier DX code. This seems especially true for Intel from the reactions.)
Now, there is another problem, as a previous poster alluded to. You can have fast rendering and correct rendering as long as you pick any one exclusively.
Also, the features that the games designers are clammering for today will be out of fashion in a couple of years time and will not match the hardware then, which means that the Kronos group would get it in the neck whatever they did.
(The skull and cross-bones 'cos specifying a standard is life threatening.)
Well, Sony, in the 1970s, used to be a company who produced high quality products which could be relied upon. Indeed, this was the case into the 1980s as well.
However, since the late 80s the quality plummeted. Two cases in point, a TV I bought in 1990 and a HiFi I bought at around the same time both died horribly within 5 years and the sound on the TV was terrible too. (This contrasts with the Pye (Philips) TV bought in 1984 which is still going strong.)
Sony stuff is basically tat now, it's not surprising that their profits are shrivelling.
As for the PS3, it's a good product... i.e. it's a great Bluray player which happens to play (a small selection of) games as well. ;-)
Seriously, though, they would have shifted a heck of a lot more units of the PS3 if they'd have kept hardware PS2 compatibility as the amount of PS2 software out there (and still being produced without PS3 versions) is large. Despite the extra costs per unit involved I'm sure they'd have been able to recoup the costs easily with a smaller margin but higher sales.
The problem was that the sysadmin was paranoid.. to the point where he wouldn't even write the router configuration to the router's flash memory. (Yes, if the power failed the router would lose its configuration unrecoverably. Maybe it was safe from hackers but it wasn't safe for hardware failure.. stupid sysadmin!)
Apparently he didn't give anyone the password or write it down because he didn't trust them.
Hmm.. I wonder who owns the copyright to the Sinclair code now?.....
If it's Murdoch then shouldn't we be worried about pay-per-view ZX Spectrum emulation? Or should Sky be given the QDOS for keeping it released for private use? Could this be a Quantum Leap in the way it does business?
I think that we all need to be told.
Sorry, but there has never been a guarantee that the connection logs on web servers should show what the user his/herself clicked upon. It's just that up until now that's been a (semi-)reasonable assumption.
The latest AVG spoils the fun for advertising sellers... tough! Get over it. Things change. Find another metric if you have to.
In the future browsers may walk web page links and pre-cache them for the user in case they need them... should this be banned as well just because the content providers find it inconvenient? After all, the person browsing the site is the customer and the service provider the vendor and the old mantra is "the customer is always right" after all.
Hmm.. it actually looks as though it's not really an SP£ problem, it's more of an OEM image problem which is being flagged up by the installation of SP3.
Basically, HP (and others) to save costs produce one hard disk image irrespective of the computer's processor type, and they install the Intel power management driver.
Now, it seems that until SP3 this didn't matter too much and the machine booted OK. However, something's changed in SP3 which means that the Intel power management driver crashes when it's run on an AMD chip (for which there is a separate driver which the OEM should have installed instead).
Don't blame Microsoft for OEM's penny pinching.
(Note: if M$ were to blame I'd be the first to throw the stone.)
"Maybe the BBC should start up a satellite service that regurgitates zillions of channels of endless pap consisting of game shows and inane talk shows, and age-old sit-coms? Now that would be REALLY innovative, don't you think?"
That'll be FreeSat, coming to a dish near you soon!
http://www.freesat.co.uk/home.php
The problem currently, really, is that data centre cooling systems haven't really changed in decades. Yes, the cooling systems have become bigger and some have per-rack chillers but essentially they're still just big refrigerators, wasting electricity cooling recycled air and pumping it inefficiently into the atmosphere.
What is really needed, therefore, is a more intelligent use of the waste heat and a more intelligent way of cooling the servers.
Now, let's look at the UK, for the majority of the year external temperatures are at or below optimum operating temperatures for systems. So, why not use that external air, passed through filters, directly as cooling air and then pass the clean, warmed air out of the data centre and into the offices where it can do some use on a winter's day and save some heating costs? During the summer, assuming it's warmer than this one, then you will need chillers to cool the incoming airflow and possibly some recycling of the internal air could be used, but overall the yearly energy bill would go through the floor.
the servers involved are running Oracle databases and hence have to be held down at some ancient patch level which Oracle has blessed as being able to run their database program?
(Don't laugh, there are many "enterprise" software packages which will only run on certain, ancient patch levels of OSs and which break if anything's upgraded.)
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
