* Posts by Nigel 11

3191 publicly visible posts • joined 10 Jun 2009

Patch NOW! Microsoft slings emergency bug fix at Windows admins

Nigel 11

Re: MS, please help me understand

As a developer you should then be aware that it's pretty much impossible to release 100% bug free code, especially when your talking about something the size of Windows.

True, but ...

There is such a thing as coding with security in mind. A long time ago Microsoft hired the chief architect of the VMS operating system away from Digital, with the brief to write them a secure kernel to replace Windows 98. The result was Windows NT 3.51. It was the most secure system Microsoft ever had, possibly second only to VMS in terms of excellence.

Being secure meant that graphics performance sucked compared to Windows 98 (where there was basically no security at all). This was a completely inevitable result of securely managing the system's memory on the hardware of the day. So what did Microsoft do? It took this kernel that had been engineered for security, and blew holes in it in order to make the graphics run faster. Enter NT 4.0. Broken by design and orders from the top. Then 2000 (further security compromises), then XP(even more). Eventually what had once been one of the most secure OSes in existence (perhaps behind only VMS) became an unmaintainable kluge. Around XP SP2 they claimed to realise that security mattered and started trying to patch the holes that they had deliberately created in a once-secure design. The result was an un-maintainable kluge.

So they re-wrote it again. Enter Vista ....

You may say that was all a long time ago and you'd be right, except that you'd also be asserting that a system that was deliberately broken security-wise can then be patched back to secure by the people who broke its design.

The evidence all suggests that Microsoft simply does not understand security at all.

And if you think Linux et al are any different you're very much mistaken

Different culture. Open-source applications are of variable quality. Some are excellent, some less so.

The Linux kernel is engineered with security in mind and is overseen by Linus. He is very smart, he does not suffer fools gladly, and most importantly he has no marketing department to tell him what he has to compromise (ie, break) tomorrow, because some touchy-feely focus group of non-technical users thinks it would be a good idea to let it display pink elephants galloping faster.

More generally the Linux ecosystem learns from its mistakes. Things in active development get better. If there is a disagreement one project may fork into two, which then compete until either one branch runs out of supporters, or (occasionally) until both branches have found different niches in the open-source ecology. It's a very similar process to natural evolution. In both cases good designs prosper, poor designs die out.

PHONDLESLAB-ULOUS: Motorola Moto X Android phablet

Nigel 11

Re: Fixed battery!

how difficult the battery actually is to replace

That is key. If it's something anyone can do if they are handy with jewellers screwdrivers and aren't bothered about voiding any warranty on the device, I wouldn't be too bothered. If, on the other hand, the battery is glued in so that its replacement is completely impossible, I wouldn't buy it. Likewise if getting inside it requires special tools, or if reassembly requires one to have four thumbs.

I can understand manufacturers being worried about bad publicity and lawsuits caused by third-party substandard batteries. Requiring anyone replacing the battery to employ a screwdriver and to break a warranty seal isn't unreasonable. (If they're really smart the phone would sense this and transmit "warranty seal broken" back to base, just like printers lock "3rd-party ink cartridge used" into their firmware to void the warranty, if you use third-party cartridges).

Blackpool hotel 'fines' couple £100 for crap TripAdvisor review

Nigel 11

Hotels have license to charge a guest's card for abnormal expenses that the guest causes. Things like fouling the bed or trashing the room. It'll be in the terms and conditions, and it is probably sanctioned by common law in any case.

I hope it is proved that this is stretching that license well past breaking point.

Nigel 11

Re: No sympathy on either side.

I've stayed in £35 places that offered clean linen, a not-uncomfortable bed, an absence of bed-bugs, functional furniture and equipment, and some breakfast that was edible. Sure, you shouldn't expect more than that, but often all you want is a place to sleep.

Nigel 11

Re: No riff raff

No, Basil was funny. Kray, more like.

Nigel 11

Re: Simples...

It's in their terms and conditions, so they can tell the credit card company that it's not unauthorized. You might be able to persuade the credit card company to do a chargeback, in which case the hotel would have to sue you it recoup its "fine" (and I can't see any court supporting them in that endeavour!)

It's almost certainly unfair trading - contact trading standards

It's possibly a crime - contact the police. Not fraud though, if it's in the Terms and Conditions and the hotel can prove that you accepted them. Freedom of speech? Human rights act? Murky waters. Still, hopefully the hotelier gets a talking to from the plod.

It's an interesting legal question whether it is one transaction or two under the Consumer Credit Act. If one (room charge plus fine) it's over £100 and the credit card company is jointly liable. In that case I'd issue a small claims summons against the credit card company, on the basis that you do not believe that the hotel will stay in business long enough to get your £100 back by suing the hotel. Another reason is that the credit card company ought to take one look at it, refund your £100 plus court fee, and charge it back to the hotel, employing the fine print of its terms and conditions!

Unfortunately if it is two transactions under the law, the CCA doesn't apply.

VINYL is BACK and you can thank Sonos for that

Nigel 11

Re: a violin

Obviously composers who were brought up with the classical instruments would work to get the effect they wanted, but except for organ composers like Bach we can never be sure of how successful they were within the constraints

We can be fairly sure what they got, by playing their music on period instruments or reproductions built from the same materials to the same dimensions. An old piano is a lot less powerful than a modern one, and the balance between bass and treble notes is different. A violin or cello or oboe, on the other hand, sounds the same. (Someone with perfect absolute pitch will notice that an ancient wind-instrument "A" is a modern A-flat). An old (natural, valveless) horn sounds similar, though of course the range of notes that it can actually play is considerably lessened. A modern organ with an electric air-pump is likely to be vastly more powerful than an ancient one with a bellows pump, but every organ isand was different, and composers rarely if ever specified what organ or even what sort of organ best suited their music. I think this is because they didn't think it mattered very much. Bach sounds wonderful transposed onto anything that can convey the melody and harmony.

As to what the composer heard inside his head ... one can never know. I'm able, with a great deal of concentration, for a piece I have heard before, with no more than four parts, to "play" it inside my head by reading the manuscript. A better musician than myself doesn't need to have heard it before. A composer imagines it, then writes it down. A deaf composer cannot revise it in the light of what a rendition sounds like in reality. Terry Pratchett probably had it spot on when he observed that the gods' idea of a joke was to make a composer go deaf. Which didn't stop the music. It only stopped the distractions.

Nigel 11

Vinyl: Reproducing sound by dragging a scratched piece of low grade plastic past a tiny rock on the end of a tiny stick held between two magnets at the end of a longer stick. The vibrations in the small stick along with the mechanical noise of the turntable motor, and audio feedback from the speakers results in a tiny induced current that is fed into an amplifier along with all the electrical noise that leaks in. Only at this point is there any prospect of science of fidelity, because the previous stages are all penny farthing technologies.

-1. Would -100 if I could.

Those technologies work!

Ever thought how the dragging of horsehair anointed with gunk out of pine-trees across dried gut and wire strung tight against bits of somewhat random wood assembled by a mediaeval oik sounds so wonderful? (That's a great musician playing a Stradivarius violin, if you didn't guess)

Ever thought about how your ears work? A bit of skin stretched across a tube, with a mechanism made of three small bones conveying the vibrations into a horn-shaped liquid-filled tube full of hairs connected to transducers which convey electrical signals to your brain ... they manage to handle a dynamic range of at least six orders of magnitude, and frequencies across three orders of magnitude. The quiet threshold of hearing is just above the level at which signals would be drowned out by the thermal noise of air molecules hitting your eardrum at random. Oh, and the same contraption also enables you to walk and run on just two feet without consciously thinking about it. Hearing and balance and inertial sensing.

Nigel 11

Re: Vinyl introduces a lot of failings

44.1 khz is an appropriate choice for the technology when CDs were invented. (No, I do not know why 44.1 rather than 44.8 or some other number). Today you could certainly do better. 16 bits likewise. (Perhaps if 6-bit bytes had won out in the 1970s we'd have had 18-bit CDs, with a maximum play time about 12% less)

The relevant maths is Shannon's sampling theorem. 44.1 kHz perfect sampling allows encoding of input frequencies up to 22.05 kHz. Higher frequencies get "aliased" so input at 23kHz would get reproduced as noise at 21.10kHz (ie wrapped back along the audio spectrum from 22.05kHz). Which is fairly OK because the highest frequency you can hear is around 20KHz and the highest you can appreciate musically is half that or less. So they put a sharp analog filter in the input signal (prior to sampling) to attenuate frequencies above 20kHz.

Contrast this with the garbage generated by lossy mp3 and similar coding, with tonal artifacts inserted across the entire audio spectrum. Music encoded onto and back from a CD remains musical. Music encoded as lossy-mp3 and decoded is horribly degraded. In my case, I can't regard the resulting noise as properly musical. So I'll stick to CDs, and hopefully before they stop making CDs they'll offer losslessly encoded downloads as standard.

As to vinyl? Well, in basic form it introduces only musical distortion (mostly 3rd harmonic "warmth") and ignorable neutral noise (hiss, clicks from dust or damage). If you never, ever touch the recorded part of the vinyl surface, and never played your vinyl with anything except a deck and stylus of the highest quality, and don't ever play it more than a couple of dozen times, and change the stylus often enough, and provided the master was cut by an expert, and providing the sub-masters weren't used to press too many vinyl disks, and provided you've got sufficient decoupling between your deck and the loud bass from your speakers, ... vinyl can give a CD a decent run. But the price in money and convenience is higher than most are willing to pay.

Mastercard and Visa to ERADICATE password authentication

Nigel 11

Re: Biometrics

I've seen fingerprint authentication fooled for the cost of a camera and an inkject printer.

Or for the cost of a piece of sellotape (to lift a fingerprint), a small piece of photo-resist-coated PCB material, standard etchant, and a blob of silicone rubber. Which method has the advantage that it does not need any connivance from its victim. It's just a slight modification of the long-known method for putting a random fingerprint on an incriminating object. (Pray you have a good alibi if it's your print they lift )

Or for no cost at all. A brutal criminal will just cut off your finger(s) and leave you tied up while he empties your bank account. Mercedes used to sell cars that used the owners finger instead of a key. Until South African carjackers started cutting drivers' fingers off. Mind you, that was better than being shot dead and then having your fingers hacked off. Or vice versa. No way I'd drive any car except a rust-bucket in a country like that. Safer still to not go there at all.

No way am I ever going to carry a financial instrument that uses part of my body as a key.

US carder gets nine years in cooler, must pay back $50 MEELLION

Nigel 11

Re: I don't get it

Coz he's a criminal and the banks ... er ...

Bible THUMP: Good Book beats Darwin to most influential tome title

Nigel 11

Re: "most important book"

Those morals and value judgements existed long, long before

I'd argue that they didn't, or at least that no society had hitherto been founded upon them. The Roman and Persian empires, China, Japan, the other ancient empires I have read about, ran on completely different moral codes. Today's Islamic world and today's post-communist China likewise do not share our moral code (hence much trouble in today's world).

The enlightenment refined the moral framework and downplayed the superstition and dogma. Darwin's church was Christianity at its best. Seems to me that things have been going downhill since then.

Nigel 11

Key knowledge

Two key bits of knowledge

"Matter is made of atoms which are ordinarily indivisible. They tend to stick together, but if you try to squeeze them closer they repel each other strongly" (Paraphrase of Richard Feynmann's suggestion)

"Disease is caused by invisibly small parasites called germs and viruses" and advice on how to stop them being spread around.

Not sure if there is a single key book that encapsulates these - one on solid-state physics and chemistry, the othe on microbiology. There ought to be!

Nigel 11

Re: "most important book"

"Most important" is vague.

"Had a hugely signifficant effect on history": "Mein Kampf", "Das Kapital" both qualify.

"Has contents that everyone ought to know about": MK definitely not, DK probably not.

BTW even as an atheist, I'll agree with putting the bible (or at least the New Testament" on the list. It's the foundation stone of our accepted system of morals and value judgements. Much of what's gone wrong recently wouldn't have gone wrong, if the people in charge had adhered more strongly to that moral framework.

Microsoft .NET released from its Windows chains... but what ABOUT MONO?

Nigel 11

Re: Reap what you sow

@asdf

Are you actually a person, or a bot? Because that comment doesn't contain enough internal logic for me to work out whether and how to disagree with it.

But just for starters Wayland and Systemd are both open source. What part of open source do you not understand, and in what way is that like Microsoft 25 years ago??

Philae comet probe got down without harpoons

Nigel 11

Re: Just brilliant

how you navigate such a small craft millions of miles away so accurately onto a hurling rock,

Same way hedgehogs make love.

My HOUSE used to be a PUB: How to save the UK high street

Nigel 11

Re: "Crimbo" please don't!

Oh, but what he's talking about is not a religious festival. It's a secular commercialized phenomenon. I suspect if Jesus returned today, it would not be moneychangers in the temple that he'd take against. Bankers and big retailers more likely. (Maybe the same thing, if one reads "world" for "temple" and treats it as allegory).

Crimbo seems appropriately horrible for the secular stuff. Christmas is the religious festival that happens at the same time of the year.

Nigel 11

Re: Tesco's problems aren't just economic

I'd call Tesco a textbook example of an organisation that got taken over by accountant types who hollowed it out in a search for ever increasing profits. Things that used to work, carried on working on auto-pilot, but with nobody left who understood the how or why of them. Everything from the shop floors to corporate finance. Last one out, don't turn off the lights ....

Eventually, one change too many, or one year on auto too many, and the house of cards collapses. Same thing as RBS really, except that Tesco can be allowed to go bust should it run out of money.

(I defected from Tesco to Sainsbury's a couple of years before the horsemeat scandal. When that happened, I can't say I was surprised. The writing was on the walls, and no-one there was reading it )

Nigel 11

Re: With you on this

There is a tremendous amount of dead wood in councils, you don't need a 1-2-1 match of managers to do-ers and having worked in councils before, about 20% of the staff do 80% of the work.

Just like IT in most larger companies then.

No, I take it back. The councils are models of efficiency, compared to the programmer / manager work ratio!

Nigel 11

Re: Not all councils are hopeless

a taxi driver "in the know" told me his mate was looking at £26,000 a year in biz rates for a shoebox of a corner shop, the rent was a paltry £20,000 a year by comparison.

For a mini-cab office?

It's just occurred to me that you could run a mini-cab operation from the back of a mobile home these days. I wonder if that's actually illegal provided you pay your road tax?

Nigel 11

Not all councils are hopeless

My council is doing a reasonable job of getting boarded-up shops on the fringes of the high streets converted into houses (*back* into houses in many cases). This keeps the centre of the high street alive (and with a greater number of potential customers nearby).

Parking restrictions are not unreasonable either. There are free-for-one-hour bays. The real problem is congestion. There are many shops where if someone parked outside for even five minutes, it would bring the whole area to a stand-still. If the high street was built in the days of horses and carts, it's not really fix-able.

A suggestion that would work in some places with slightly wider high streets. Introduce even-odd day parking controls, as widely used across the channel. On even-numbered days, one side of the street is a clearway (no parking, no waiting, no stopping). On odd-numbered days it's the other side. So you have an entire side of the street where you can park, every day. Three lanes, two of moving traffic and one of parked vehicles. Retailers just have to arrange deliveries on the days when a vehicle is allowed to stop outside their stores.

Oh, and NO to allowing anyone to convert a house into a shop anywhere. If one has paid for a house in a residential neighbourhood, I think it's right that one can't suddenly find the environment degraded by a shop opening next door. Keep shops in areas zoned for retail, and then you pays your money and takes your chances if you buy a house or flat in a retail zone.

The last PC replacement cycle is about to start turning

Nigel 11
Coat

Re: There always have been people who do things in a certain way...

Obviously fun things like trying to prove you can do X even with unsuitable things

Do X? Like displaying from a remote X-windows app on an Android mobile?

Nigel 11

Re: Followed by....

I do find the idea of saving serious power by running Linux on ARM instead of Linux on Intel very appealing though.

Are you aware just how few watts you need to run an Intel system these days? If you don't want huge number-crunch power, build a system with no moving parts out of a Celeron J1800 Mobo (such as Gigabyte GA-J1800N-D2H), an Akasa Crypto case (tiny, don't connect the fan!), 8Gb RAM and an SSD. Passive-cooled, completely silent and mine doesn't even get seriously warm. If you do need huge CPU power on request, intel's Generation-4 core CPUs are good at not using watts when idle, provided your OS, drivers and applications play nice. (ie all event-driven, no polling loops).

Yes, at the really low power end ARM is better. I've used a Rasberry Pi as a "thin client" just to see what it felt like. (A bit too sluggish for my tastes, though not unusable).

Nevertheless what's eating the power is your display rather than your state-of-the-art Intel CPU.

PS Why do I have to build my own silent desktop PCs with no moving parts? It's a sign of lack of imagination, that nobody like Dell is selling such systems, to replace ancient, loudly whirring "classic" desktop systems that are still out there in vast numbers.

Nigel 11

Why does it always have to be a genuine honking workstation actually sitting on your desk?

It doesn't always. I was recently in a furniture chain where the salespeople all had Ipads and took your order sitting on one of the sofas. They could also show you what any sofa looked like in any fabric. Quite impressive.

But personally, I prefer to type on a proper full size keyboard and look at a full-sized screen (24 inch 1920x1080, lamenting the disappearance of 1920x1200). That way I don't get RSI in my fingers and I don't get eyestrain and headaches and back-ache.

I have a smartphone and a tablet and find uses for both. But if I had to choose I'd give those up rather than my desktop (assuming I could go back to my old dumb phone for calls and texts -- with a battery that lasted two weeks, some things don't improve).

At work, any employee whose job it is to sit in front of a computer for most of the working day deserves a proper desktop system. And the employer is in danger of facing a claim for causing industrial injuries, if they think otherwise.

PS the laptop probably *is* dead. It always was a poor compromise, with a shrunken keyboard with limited key-travel and a too-small screen too close to the keyboard or too far from your eyes. As for those expletive-deleted mouse-pads, the less said the better.

Horizon finds new potential in the sci-fi staple deepsleep tale

Nigel 11

A deepness in the sky

Cryosleep is essential to the plot, and even more essential to the back-story and universe-building, of what may be my all-time favorite SF story. Vernor Vinge, "A deepness in the sky". If you haven't read it, then do.

Why solid-state disks are winning the argument

Nigel 11

Re: Are they Round?

Solid-state Storage Devices. How many people above are calling them disks, as opposed to SSDs? I'd have voted for SSSDs but never mind.

Nigel 11

Re: Developers need what now?

Depends what sort of developers.

People who are coding and building, should have machines that can do it fast. Some sorts of debugging, likewise.

But people who are testing for release, should at least some of the time be testing using the crappiest PC that a customer might still be using.

My favorite peeve is websites that were never tested other than on a Gigabit internal net. You do NOT need a gigabit link to develop HTML and Javascripts. You should be exiting the building on a crappy ADSL service from a crappy ISP, and looping back in via the big bad internet. That's what some of your customers are seeing, stuck on the end of too many miles of corroded aluminium POTS cable that's somehow managing to support ADSL at a few Mbit/s (when it's not raining).

Nigel 11

Re: Too many words

because they're faster. Next question please.

Also more shockproof

Also quieter (silent)

Also less heavy

Also longer battery life, or even less weight by using a smaller battery.

edit:

Also, with dense-packed equipment in a server farm, less electricity eaten and less expensive air-con needed.

Don't know, but also suspect SSDs happier at high ambient temperatures than HDs (industrial/ embedded PCs)

Nigel 11

Re: "it is a really dumb idea to take the cheapest desktop hard drives you can find"

It's in their interests to sell you "server grade" drives at twice the price. So of course they would say that.

What they will never tell you is that it is very much in your interest to buy half your drives from one of their competitors. This is true even if it's provable that the competitor's drive is half as reliable

That's because a drive from a different manufacturer is far less likely to contain one of the same batch of defective components. Two drives with near-identical serial numbers will likely contain the same faulty components and therefore are likely to fail at or near the same time. Mirroring won't save you if this happens.

Give me a manufacturer X desktop drive and a manufacturer Y desktop drive any day, over any two identical Server grade drives with near-consecutive serial numbers.

Nigel 11

Reasons for traditional HD

1. Cost per Terabyte is still much lower for HDs.

2. I have more faith in mirroring applied to hard disks than to SSDs.

In my experience, a majority of HD problems show up in the SMART statistics well before the drive fails. Then I replace the drive pro-actively. I also try to pair drives from different manufacturers to reduce the risk of a common-mode dual-drive failure.

Will SSDs warn in advance of failure? CAN SSDs warn in advance of failure? They're a new technology (more accurately, several new technologies), and I think it'll be a few years before we know. I'm not even certain that running a mirrored pair of SSDs is useful (but given that I'm talking about multiple TB of data, it'll be a few years before I can afford to find out! )

Will multi-TB SSDs compete with multi-TB HDs? If 3D flash can expand further into the 3rd dimension, that may happen sooner than we think.

BTW why is putting Flash memory (a SSD) on the PCI bus still regarded as exotic expensive server technology? 6Gbit SATA is now the bottleneck for even consumer-grade 240Gb SSDs. Give us a small and cheap but very fast PCI card to boot and run our O/Ses from!

HOT YOUNG STAR about to GIVE BIRTH, long range images show

Nigel 11

Re: I remember when ....

For the record, it wasn't uncontroversial in Christian circles only until recently. Giordiano Bruno was burned at the stake for saying the same things and refusing to recant. (I don't know which surprises me most. that he refused to recant to save himself from an agonizing death. Or that he was so right on so little evidence! )

Antares apocalypse: Orbital points finger at turbopump FAIL

Nigel 11

Re: @ Nigel 11

Wasn't that the DC10 disaster that the manufacturer blamed on an airline using a fork-lift truck to lift engines on and off the wing? The engine didn't just fall off, it took large chunks of the wing with it, when the whole structure failed.

I believe Boeing made a comment to the effect that they also don't recommend installing engines using fork-lift trucks, but they make their wings and engine pylons strong enough to take it!

Anyway, it wasn't an engine blow-up. Google "Thompson Manchester Airport bird strike" for good video of a very scary take-off. (Wasn't a blow-up, but I don't imagine that the engine which ate the bird was generating much thrust after it did).

An engine failure while the plane is on the runway but committed to take-off is the worst case. There's quite a large scope for fatal pilot error should that happen. The Thompson jet was already in the air -- slightly less bad.

Nigel 11

The difference is that if 1/4 of the engines on a 747 blows up, the damage is contained, and the plane will be able to continue in flight -- or from point of no abort during takeoff -- and land again safely(*). Same is true of 1/2 engines on a twinjet.

(*) Design 100% sure. Reality 99.something% sure. Would prefer not to be on the flight with the exploding engine.

Nigel 11

Re: Ha! Called it!

The West would have been completely f***ed but there would have been absolutely no chance of them launching a second strike against anything that survived being nuked by the 25% of their first strike that arrived.

MAD, innit.

Intel: A tiny video drone? Disguised as a bracelet? Great! Take half a million dollars!

Nigel 11

Re: Not cool enough...

Which may be rather less difficult than other Culture-level tech. Start by trying to make a dragonfly-drone. A micro fuel-cell is probably the biggest gap in our know-how. (DAK know if that's the path nature took - large flying insects first? )

I've just realized, IMB didn't specify the normal size of mosquitoes in that part of his universe. Some of the ones that bit me in the USA, the drone might have been rather larger than I was imagining when I read the book!

Nigel 11

Re: Not small enough?

Ornithopters ... Oddly enough ,nothing man-carrying appears to be viable

And won't be. Birds max out at bustard-size, somewhat under 20kg. I'd guess that there are wing strength/stiffness issues that prevent anything heavier from getting off the ground if that involves flapping its wings. Similarly at a larger scale, fixed-wing aircraft max out when the wings can't be both long enough to generate sufficient lift and stiff enough to support the weight.

Although large birds and insects both fly by powering their wings, the aerodynamics are very different. Insect-sized wings use the vortex generated at the tip of the wing to generate extra lift. This is the explanation of how a bumblebee manages to fly with such small wings. I suspect that it's also the most efficient form of flight for small creatures and devices (simply because if it weren't, nature would have evolved the alternative! )

Nigel 11

Not small enough?

Is this the best current technology permits? We have a way to go to catch up with nature. When will we build an artificial dragonfly with the same hours-long endurance? Small powered wings are presumably much more complex than rotors, but also much more efficient.

(They aren't around today, but the fossil record shows that the dragonfly design works at much larger sizes).

FIFTEEN whole dollars on offer for cranky Pentium 4 buyers

Nigel 11

Intel, from the very beginning took the x86 architecture and made it their very own. Instruction sets that should have been agreed on a universal basis were constantly being added to by Intel without any consideration for the future.

Really? Sun, MIPS, IBM, Digital, DG, and numerous others all consulted their customers and competitors before adding to or changing their instruction sets?

I've speculated before that Intel is a smart company and recognises the danger of becoming a regulated monopoly. It therefore needs AMD to remain in business, so it has competition. Once, Intel tripped, AMD got ahead (with the Opteron and the 64-bit instruction set, and Intel had a few tough years catching up (including some unfair marketing). Now Intel is so far ahead it is in danger of putting AMD out of business. If I'm right, Intel is about to find some way to give AMD a break. Perhaps it might purchase the rights to use ATI Graphics technology? Intel's graphics architecture is most definitely not up there with its CPUs. (Edit) Hardware-wise, that is. Intel's graphics driver softwre and Linux support are excellent.

Nigel 11

Re: Pentium 4 didn't suck. @Gordan

You make a very good point, but you ignore that compiling for a particular processor, using all of the features of that processor breaks the "compile once run anywhere" ubiquity of the Intel x86 and compatible processors.

Not broken. It'll run, just less efficiently.

Most programs that seriously tax a modern CPU or even a decade-old one, have a small fraction of the code that accounts for a large fraction of the CPU usage. So it's very valuable for the 90% of the code that isn't executed so intensively to run, albeit inefficiently, on any CPU with that architecture.

As for the other 10%, distribute it as a separate library, containing something that determines which processor it is running on, and multiple compilations of the same code with different optimizations. Then dispatch to the appropriate one. As CPU and compiler technology advances, you can ship an update that replaces the library with better code (yet derived from the same source, on the unlikely assumption that no bugs needed fixing).

These days you may well also bundle versions that don't use the CPU at all, but a GPGPU if it can find one and if the gain is worth having.

So long, thanks for all the ...er, FISH BRIGHTER than boffins thought

Nigel 11

Re: I had zebra fish ...

and adaptation is unlikely

Although I did read about an octopus being kept in a marine biology lab, that handled brief spells in air rather well. The lab had a problem with fish disappearing from a tank. They rigged up a camera to catch the thief. The next night they watched their octopus lift the lid off its tank, walk across the lab to the fish tank, catch a fish, walk back to its tank, and pull the lid back over itself from the inside.

Molluscs have, of course, successfully colonised the land. (Slugs. Yeuch! ) Fortunately for us, no long-lived intelligent ones with tentacles. Not yet. Give them another fifty million years ....

Nigel 11

Re: I had zebra fish ...

Apparently even amoebae can learn to associate correlated stimuli. That's pretty good going for a monocellular organism without anything we can recognise as a nervous system.

Can any fish scale the heights that "mere" invertebrates have managed? Last night on TV, watched an octopus gather up two half coconut shells, put one on each side of itself, and pull them together to make a secure home. Found tool use, using unnatural entities dropped by human beings from above. We've not been chopping coconuts in half with machetes for very long, so it can't be instinctive ... and octopuses have only a couple of years of life in which to learn anything.

Perhaps, though, it's an unfair comparison. If an octopus is the most advanced mollusc, then shouldn't a human be seen as the most advanced fish (ie, vertebrate).

Brazil greenlights $200m internet cable to Europe in bid to outfox NSA

Nigel 11

Looking at that map...

Looking at that map, what might make a lot more sense would be a cable from Brazil to West Africa. Africa is poorly connected, Africa is rapidly developing, a longer subocean route might be more reliable than one going through or around North Africa to Europe ...

... and Brazil and African countries may have a shared political interest in appearing to be exclusing the NSA from their networks.

Perhaps once Ebola is vanquished?

Multi Jet Fusion: THAT's HP's promised 3D printer, not crazy 'leccy invention

Nigel 11

Re: Expensive refills

Recent HP printers are also reliable, if you avoid the mistake of buying the cheapest they make. The really cheap ones are designed down to a cost, then sold at a loss, to cause sales of very expensively packaged ink. So, do spend a bit more on the printer in the first place to get a quality product, and you'll find the running cost per page is *much* lower for anything except tiny usage.

We've been using HP Officejet Pro printers since the K550 and have enough of them to know that we aren't just lucky. They have 3 year warranties and regularly last for two or three times longer if lightly used, or for many 10Ks of pages if heavily used. In any case the cost of the printer expressed per page printed is negligible compared to the cost of the ink, so one treats an out-of-warranty printer failure as a consumable. And before you say that the ink is the killer cost, it isn't. HP advertise these printers as cheaper to run than a laser, and by and large it's true. For mostly black text on a white background (but with some colour), running them works out about 1.5p/page, including the occasional replacement printers.

(Yes, you do have to persuade users not to print A4 photographs, which cost up to £1 in ink ... each! The fact that these printers are not photo-quality (nor sold as such) is a distinct advantage on this front. Quality is about what you'd expect from a colour laser printer. Maybe better if you use the special HP shiny paper instead of standard copier paper).

Return of the disk drive bigness? Not for poor old, busted WD

Nigel 11

I have 3 WD 1 TB drives of different lines (Blue, Green) running in 3 different PCs for over 3 years without a hiccup.

Sigh. When one of them fails, you'll have a failure rate of 33% and suddenly WD is terrible?

Statistics 101. The accuracy of an average is related to the square root of the number of observations that went into it. So Backblaze's observations of thousands of drives probably allow them to compare failure rates to about 1.3% (ie annual failure rate 4% means ina range ~ 2.7% to 5.3% range) and just about everyone else's set of data is too small to say anything much at all.

But in any case, the real devil is common-mode-correlated failures. You can almost always protect against randomly distributed rare failures by using mirrored pairs of drives. But if you deploy two drives with near-consecutive serial numbers, failure of the first from certain causes becomes a good predictor of the imminent failure of the second. So buy one from each of two manufacturers and pair them. Even if the second manufacturer's drives are reliably known to have a higher failure rate, using it makes it far less likely that you will suffer a two-drive failure and downtime or worse.

Oh, and do make sure that someone is monitoring the drives. Recently I heard about a NAS box that had been screaming "one of my drives is dead" at the e-mail address of an ex-employee for about two years ... yes, the other drive died defore anyone noticed. TbftgoGgI.

Nigel 11

So immediately following the Asian floods that "caused" hard drive supply shortages, both firms profits spiked ?

If you have any evidence that this was a mere cover story, make it public or share it with the authorities. Because any such collusion to raise prices is illegal (and because our governments are running out of banks to hit with billion-dollar fines :-)

Simple economics tells you that such an event *should* raise profits. You make the same profit selling N drives at 6% margin as 2N drives at 3% margin. Ordinarily you are prevented from taking (say) 20% margin by a competitor who sees an opportunity to grab your share of the market and make only 15% ... and you retalliate, and margins fall back. If you have first-to-market advantage on a better product you sell it at a premium price for as long as you can. This is why the biggest drives cost more per TB than the smaller ones, even though the actual extra cost of making them is probably much less than in proportion to their capacity.

When there is a shortage of product compared to market demand, not only does price competition stop, but prices (hence margins) have to be raised to choke off demand. Price is a mechanism to make sure that the people who most need the drives get them, and the people who need them less, choose to wait a bit. The Soviets never did understand this. They thought that central planning would work better. It didn't. Politics aside [utopia to the nth power, that], no-one can solve a large travelling-salesman problem, which is what it would take to do central planning properly.

Nigel 11

Re: WD or Seagate? Hmmmm....

If you can't afford pairs of drives with mirroring, how much is the data on the drive worth?

BTW That's not a rhetorical question. Where I work we have many tens of TB of data on unmirrored (JBOD) drives. But this is a sort of a cache of simulation results that scientists think it might be useful to refer back to later. If a drive dies, so be it. They'd rather use any spare money on cacheing more results. And if they badly want to see a particular result that was lost, they can re-run the simulation.

Nigel 11

Re: Shame

I've made this comment lots of times before, but once again: all, repeat all, manufacturers have shipped batches of lemons from time to time, caused by a batch of faulty components or (occasionally) an unanticipated premature ageing problem.

If you run RAID drives in mirrored pairs, save yourself from (most) common-mode failure problems by pairing a drive from one manufacturer with a drive from another. If you have two drives with nearly consecutive serial numbers from the same manufacturer, it becomes far more likely that one will fail and then the other will fail from the common cause before the replacement drive is fully synched.

Have to say, I do like WD "Red" drives. I don't have a large enough sample to comment on reliability (FWIW no failures so far). But they certainly run cool and are uncannily quiet. Perfect for NAS boxes in a domestic or small office environment.

Ex-Soviet engines fingered after Antares ROCKET launch BLAST

Nigel 11

Safety margins

It's worth noting that safety margins are pared far finer in the launch of unmanned rockets than in any other field of engineering I can think of. The rocket equations provide ample excuse. A small gain in rocket efficiency translates into a large gain of cargo weight delivered into orbit.

I don't know what happens in practice. In principle, too many launches without a failure should result in someone deciding how to shave more weight and safety margin off future launches!

Banksy denies Banksy impostor's claim to Banksy.com – which isn't owned by Banksy

Nigel 11
Coffee/keyboard

Purgery <Definition>

Wish that could get +100 !

Nigel 11

Re: FUCKIN VANDAL

I once saw a plonker yelling at some kids who were drawing with chalk on the footpath outside their house

Could have been worse. There is a student who got a criminal conviction for drawing (well, writing) with chalk, at a demo. The college claimed, on oath, that it needed tens of thousands of pounds worth of specialist stone remediation teatment to get the chalk off. Most think that a bucket of water and a sponge would have sufficed, and that it's the college authorities that ought to have received a criminal conviction for their wanton waste of public money, to say nothing of ruining a young person's future.