2560 posts • joined 10 Jun 2009
Re: Dual Boot
Actually unless you are thinking hard about run-time on battery, you don't need to dual-boot. You can run Linux in a VM under Windows and it works just fine. In an ideal world I'd prefer to run Windows in a VM under Linux, but have to admint that getting Linux up and running and supporting the power-management and other integrated features of a notebook can be a a pain (or plain impossible if you didn't research what hardware to buy carefully enough).
Which again makes my point about a mature market. A plain ordinary laptop can happily run a virtual machine! (Or indeed, several VMs at once, if you buy one with an SSD and enough RAM).
I doubt that the PC is dead. For a lot of purposes, especially business or serious work, you need a keyboard and a mouse and a decent screen at arm's length.
I think what's changed is that no-one sees a compelling reason to buy a new PC every three or four years. The old one works just fine. Folks really don't want Windows 8. Lots of folks don't even want Windows 7. They'd be happier sticking to XP, except Microsoft are hitting the kill switch. Apple must be rubbing their hands in glee, since a Mac is the obvious alternative to Windows 8 for a home user who is forced off XP. Intel don't care - it's the same chip in a PC or an iMac.
Market maturation happens with every new technolgy sooner or later. The market saturates, the rate of progress slows, "New" goes from "must have" through "boring" to "bloody annoying gimmickry", and sales drop to the level necessary to replace hardware that has physically failed. I'd guess that's a new PC every six to twelve years, rather than a new PC every three to five years.
Tablets are selling like hot-cakes because the market for them is neither mature not saturated. Their time will come (and faster than it did for the PC). Most folks not living in poverty will soon have BOTH a tablet AND a PC, because each has its place.
perfect for CAMRA? If the Eurosceptics don't get it first.
Loyalty and experience
It's to do with two things - loyalty and experience.
If you pay well for local staff, they'll stay in the job long enough to acquire a large knowledge base specific to your operations, stored where it's most needed -- in their heads. A lot of this stuff can't ever be written down, because the people who know it don't necessarily know what they know until a problem arises that has to be solved asap.
Then someone arrives and cuts costs by sacking the experienced staff and relying on a bunch of outsourced (or even locally contracted) mercenaries who'll leave as soon as someone offers them more money. For that reason, staff turnover is high and the in-head knowledge base is destroyed. (Loyalty, of course, is long gone). The old system ticks over on auto-pilot for a large amount of time, and management pronounce it a success (and make further cuts?). Then the roof falls in.
If it's tangential to the primary purpose of the organisation, outsource it. If it's mission-critical, don't. Is there really anything else needs to be said?
Re: spinning rust and other stuff
Secondly: 2Tb and larger drives have a terrible failure rate. I'd gotten used to seeing drives lasting 6-8 years and now they're down at the 3-4 range again. Manufacturers dropped their warranty to 12 months for sound financial reasons.
So what do you say about my twelve 2Tb disks that haven't experienced a single failure since they were installed three years ago? Not even server-grade ones. Could be luck, but if the MTBF really is three years then half of them should have failed by now, and the odds of NONE having failed are 1 in 2^12.
I think you're generalizing from bad luck. I keep telling people that the big risk is common-mode failure caused by bad batches of disks. If you buy all your disks at once, then if one is of low reliability, the other N are likely the same. If you are running mirrored pairs, then you should buy half your disks from one manufacturer and the other half from a different manufacturer, and pair them heterogeneously. That minimizes the chance of losing the whole array to one bad batch of disks.
I suspect manufacturers cut the warranty because disks were getting so cheap, the warranty replacement process was becoming too expensive compared to the disks themselves, even if the percentage failing in warranty has not changed.
@Ledswinger good point
Very good point. For read-mostly data SSDs should be much more reliable than HDDs, as well as faster.
Wear-levelling works against you?
Couldn't wear-levelling actually work against you in a RAID environment, by causing all drives to fail at nearly the same time?
Not exactly. Without wear-levelling a small subset of blocks would get hammered and the "disk" would fail a lot faster. RAID or no RAID.
However, there's a difference in failure mode between an SSD array where all devices will predictably degrade to unacceptable at about the same time, and an HD array where the future failure of each device is pretty much unpredictable unless you are suffering from common-mode failure (ie a batch of bad disks). In any case HDD failure is linked firstly to hours in service, and then to seek activity. Volume of data written to a HDD does very little that could cause earlier failure.
RAID nearly doubles or trebles the number of write actions. With JBOD, M writes spread over N disks. With RAID-5, 2M writes spread over (N+1) disks. With RAID-6, 3M/(N+2). This will reduce the SSD life expectancy in that environment by ~2x or ~3x.
With SSDs you might want to define a new sort of array that puts differing amounts of parity on different drives, so each SSD in the array experiences a different level of write activity and wears out at a different rate. Of course, there would be a penalty w.r.t. the performance of such an array.
Any company which can contemplate spending $5bn on an office is circling the drain
And the drain is of course right there on the plans. Has to be, or rain would fall in and not get out util water pressure broke the inside windows. One BIG drain at the EXACT centre?
Re: Linux support techs
Well, they could show you a dozen ways to do imaging and deployment without paying any per-seat license charges, and with script-ability second to none. All starting with booting an appropriate stand-alone Linux environment off removeable media or off the network.
The question is whether they're sane, mad, or totally rabid.
Mad means they are trying to put a nuke onto a missile and (I hope) failing, while their country rots and starves.
Rabid means that they've hidden the nuke in a container that's been reloaded and relabelled a couple of times in third-world ports with lousy security, and which is shortly going to be delivered to the USA.
Relax - only because it's one of those things like cancer or a pIanet-busting asteroid that I can't do anything much about.
Ir was written by and for the marketing division. What did you expect?
Like it. (Don't be drinking cola as you view it! )
Re: How to make the world a better (choke, splutter) place
There is a way to make a legal system that covers all possible cases. It's actually pretty much the one that they have in North Korea! "You do what you are told to do by the Great Leader. Anything he approves of is OK. Anything he does not approve of will be punished". I was going to add "If in doubt, ask" but of course that's very dangerous advice because the Great Leader might not be amused, and that's probably illegal.
Anyway, it covers all possible cases. "The Great Leader is right. You are wrong. You will be punished. End of."
It doesn't even need interpretation of a holy book, as in theocracies, or lawyers, as in other places. If there were a way to make sure that the great leader was omniscient and omnibenevolent, it might even work.
It's even scarier than you think. A few years back I read an account of their germ warfare development program, which was worse even than what Mengele did for the Nazis. They were systematically and scientifically testing various weaponised diseases on "dissidents" and their children.
Do we think that they aren't now muking about with GM? If they started a war that they couldn't win, would they then hesitate to unleash whatever germs they have developed onto the world?
Re: definition of a fork...
Or a fork in the road. Or a fork in a tree. (A biological tree, or a binary one). "Implement used for transferring food from one's plate to one's mouth" is just one of many definitions. By the way, the original table forks had two tines, like a very small pitchfork.
The law, correctly applied, is the law, and we should follow it. That it has flaws is no excuse for criminal activity
Wrong. I cite the Nurenburg trials. "Simply following orders" (which were lawful at the time) is no defence for someone accused of crimes against humanity or genocide.
More generally, we all have the right to flout the law. It's often called civil disobedience. It is a form of protest about an unjust law or a more general inequity in society. What honorable protestors have to accept is that they do face punishment under the law that they flouted. Either the punishment is sufficiently minor that they don't care, or they calculate (rightly or wrongly) that public outrage will prevent such punishment taking place and instead result in a change to the unjust law.
There are catches though, you can't selectively apply the 5th to your testimony. If you say anything at all you lose the right to "plead the 5th"
I'm interested in how this works in practice. What if you start giving evidence, and then you are asked something unexpected that you cannot answer truthfully without self-incriminating? Especially if you are not the accused, but rather a witness who is guilty of some unrelated crime?
Memristor vs. Flash
If you could choose to store valuable data on a device that used a complicated VLSI chip to encode and store your data on a large-block-addressible medium that you know was likely to start failing after a few thousand write cycles, or one that used simple addressing logic to store it as individual bits or bytes on a medium that lasts for billions of write cycles, which would you choose?
HP will win hands-down, if they can get memristor storage down to much the same price per Gbyte.
If the theory translates into practice, memristors will be the biggest revolution in computer technology since CMOS arrived.
Re: Why are ATM fascias so complicated and intricate?
The answer? An ATM designed by Apple!
Re: Leccy? No!!
There are two opposed misconceptions out there:
That electric cars are the solution, end of.
That electric cars are a waste of time, end of.
They are likely to become an important part of a solution to a very big problem: how to keep our civilisation running when we've used most of the oil. Electric cars don't need oil, they don't even require fossil fuels to be burned. Also in future they can be part of an electricity-storage infrastructure (if they are connected whenever they are not being driven, which would require omnipresent car-charging facilities like we currently have omnipresent mains electricity).
Their current (early phase) development is being driven by rich guys having fun, which is exactly the same model as the original automobile. That's a better approach to just about everything new than soviet- or EC-style central planning!
The idea of actually scaling the penalty to the harm done (or threatened) seems right out the window
Do you realize that's about the only sane argument with which one could attempt to defend the indefensible? That particularly serious instances of computer crime may have a cost measured in billions of dollars, or even in lives lost?
Eating the laptop market ... really?
Why, when you can buy a laptop at half the price? (Soon, you'll work out that a mouse is superior to a touchpad, and a touchpad is superior to getting finger-grease all over your screen).
These things give you less hardware, and ought to be cheaper. Which is a give-away as to why everyone selling them loves them. Think of the profit margin!
Re: That could be a lie!
Have you ever tried hacksawing through mere piano wire? Or a padlock hasp?
Re: The mystery of the mysterious operatives
what *would* you need to cut through a major submarine cable?
I've seen a cross-section of such a cable. They are built somewhat like an onion. Protective plastic sheath, layer of poisonous gunk to kill marine worms that eat throuth the sheath, layer of piano wires for strength. Repeat several times. Finally a very skinny core containing the optical fibres. Can't recall anything copper. Stainless steel would be a better bet for resisting seawater.
You could probably hack through such a cable with big manually operated wrecking shears (big lever, gears, ratchet type of thing - distant relative of a bolt cutter). A saw or knife would be foiled by the first layer of piano wires. Above water an angle grinder would probably also work.
BTW those marine worms are persistent little bastards. After the first attackers have died, the water will wash away the poison and corrode through the wire. And then they'll be back for another go. Iterate to eventual destruction, if a turbidity flow doesn't kill the cable first.
Re: The mystery of the mysterious operatives
The other possibility is that they were innocently scuba diving and happened to be in the wrong place.
Possibly. OTOH if they were apprehended diving with cable shears or similar, probably not. Hopefully someone in the Egyptian security forces will check this before interrogating them. Also that they'll bear in mind that one wouldn't get far attacking an armoured submarine cable with a standard diver's emergency knife, so if that's the only blade in the picture they may well be innocent.
(*)Stop laughing at the back. It consumes less than 10w of power an hour.
You mean it runs on just ten Watts. A Watt is a unit for the rate of use of energy. What you pay for are units of energy. You could have said 0.01 kw/h per hour ... perverse but correct.
BTW I run an Atom server as well, though I'm wondering if it should morph into a Rasberry Pi soon.
"A small rod"
"oddly shaped objects"
You don't need anything radioactive! A noise diode is quite good enough.
Don't virtually all PCs have a hardware RNG these days? Although it's rarely used as such. It's the integrated audio system.
Rack up the gain on the microphone input and you'll hear noise generated by the switching of other electronic systems in the box. (Even better if a mic is connected and you pick up external noises). It shouldn't be hard to generate a fairly random number from a second's worth of samples thereof. It may not be good enough for crypto on its own, but XORing it with a "good" random number cannot make it any less good.
Re: "Garrett was able to demonstrate an application on Windows could wreck a machine."
Not designed to run Linux you say? If this is the metric people measure that criteria by, then it's clearly not designed to run Windows either.
And right now, somewhere in North Korea ....
The really important question
What do Samsung do when someone returns a bricked laptop (a) in and (b) out of warranty?
You can tell the manufaturers you want to do business with from the ones you don't, by how they handle a problem such as this one. You don't want to be told to <go away>, and to have to sue your supplier for supplying goods that were not fit for purpose. Which defition this bug amply fulfills: the machine claimed to offer UEFUI boot, but broke the specification in a way that causes catastrophic failure.
I'm asking, not suggesting anything. I really don't know.
That diagram must be faulty. FORTRAN isn't there at all.
Yes, it's a minority language these days. However, for those who do serious scientific number-crunching, it's definitely very far from being a Norwegian Blue. FORTRAN-95 and successors allow a compiler to generate efficient parallel code (which C / C++ can't, because of pointers).
Python has a big foot in that camp as well. It's great for writing top-level code, with SciPy and NumPy and lesser-known friends at lower levels to do the heavy lifting.
Re: hate python with a passion
I hated it (though with less passion) when I started learning Python. It took me about an hour to get used to it. It took me about a day to start wondering who was the idiot who thought curly brackets were a cool idea.
Re: Yes the Rothschild Bank and the US Federal Reserve....
A nuke would work, if it was placed correctly before detonation. To the side and well off-centre. You'd blast a small chunk of asteroid off at high velocity in one direction and impart an equivalent and opposite amount of momentum in the opposite direction. It would work if the delta-V was sufficient to cause the asteroid to miss earth (so the more warning we had, the better). Lots of variables, of course, and causing fragmentation of the asteroid would make things worse (if they could be any worse, that is).
Radioactivity - pah. Compared to the environmental effects of an extinction-level impact event, the radioactivity released by a single nuke would be totally insignificant. We tested hundreds of them in the atmosphere, didn't we?
Great for one-offs
I think this hits the nail on the head.
It may revolutionize prototyping. Design a complicated 3-D object on your computer, and print one. Handle it. Stress it. See what feels good or bad, what breaks or bends too easily. Revise your design. Print another prototype. No problem with going around this loop several times.
For manufacturing, you'll still bite off the large expense of having a mould manufactured, so you can knock out the item in thousands or millions for cents per unit.
It'll also mean that at the other end of the cycle, one-off spare parts for obsolete models will be similarly easy to manufacture. What is the state of the art in 3-D scanning? Take a broken part, reassemble it with superglue, 3D scan it and 3D print a replacement? Maybe not there today, but soon. Photo-copying for objects.
It may revolutionize lost-wax casting, because it'll be easier to make the wax originals. What makes sense for jewellers today (with hand-sculpted wax), may make sense for any smallish cast-metal object tomorow.
Perhaps surprisingly, no.
There's a lot of law pertaining to exhaust pipes in particular. It was established that car manufacturers had no rights over the manufacture of functionally equivalent parts which, externally, were indistinguishable from copies of a registered design, because they HAD to be exactly the same shape, diameter, etc. in order to attach to the car and not impede the operation of its engine. The internals were a different matter - there is freedom inside to make changes, and the third-party manufacturers were indeed making the innards differently to the car manufacturer.
Further, if you are making your own parts rather than manufacturing them for profit, you have even greater license. For example, you are allowed to make an instance of a mechanism described in a patent for your own curiosity, enjoyment, maintenance of a broken purchased item, etc. even though you need a license to sell such parts for profit.
As with audio and video, watch out for attempts to take away the freedoms we currently enjoy!
On error jump to human ...
I know what you mean, but I immediately thought of "Queen of Angels" by Greg Bear. The plot of that SF novel points out the difficulty of emulating any system whose software has to handle unpredictable realtime events.
Especially if it's one step away from full-human AI and 10s of light-years out.
I imagine that the problem with making the oil very hot is that losses to the surrounding air by convection increase rapidly with themperature.
Re: Solar moitor
I must have missed something. If you want to heat water from solar PV electricity, don't you just put a 12V or 24V immersion heater in your tank and connect the raw DC solar panel output to the heater?
I would have thought that the rest involves a motorized switch to connect the panels to the inverter or to the water tank as desired. If inverters are allergic to being completely disconnected from their PV panels at short notice, you could switch between all the PV output going to the inverter, and the PV output going to the inverter and the water-heater in parallel.
And if you also want to be able to heat water from mains electricity, just put a second immersion heater in the tank. OK, it may be hard to source a tank with two bastard curcular cut-outs at the top for screwing your immersion heaters into. I find that struggling with one is quite enough!
Back to the original idea, surely there's such enormous thermal inertia in a water tank that you wouldn't have to switch the heater on and off more than once every few minutes. Flicker regulations?
Re: Rural networking
Knock on the door of whoever is paying for the broadband, of course. I'd suggest that would be the "Borchester Village Internet Co-op" or suchlike. Obviously take appropriate software precautions so that only authorized users get access (such as end-to-end as an encrypted pipe, and some logging of who and when).
Not really any different to any other smallish not-for-profit's set-up, apart from the wireless pipe to the broadband connection location.
Regime change, anyone?
Same idea as rural networking, but a different application, for brave people living under repressive regimes. Build your own hidden communications network.
This more or less happened in Libya, although the rebels "stole" the existing wireless network infrastructure rather than just installing their own.
Re: Rural networking
It's not a new idea at all. Years ago I read about the installation of a working telephone booth at the Burning Man festival, using a chain of off-the-shelf routers strung across the Nevada desert. It's just that it's getting more affordable, and the RPi lets you integrate things like battery and power management to a greater degree. There are also quite a few routers that one can "jail-break" and convert into small fully programmable Linux boxes. (OpenWRT, etc.)
A related idea would be just to use a pair of solar-powered RPIs (or OpenWRT boxes) to wirelessly bridge a road. This is useful, because while you might persuade a few landowners to let you string a fibre across their land, just try getting permission to take a fibre over or even under a public highway. So, don't try. Just put a solar-powered fibre-to-wireless router on each side. You wouldn't even need directional antennae (which may not be entirely legal).
Re: My project...
Victorian terrace should be easy! It has floorboards? You can push your low-voltage low-bandwidth wiring into the cracks between the boards in ine direction, and pull wires through the space between the joists under the floorboards in the other. Modern flat with a solid concrete floor is harder, though if you have carpet you can also easily put wiring under the carpet.
Very thin copper wire or tape attached to a wall, wallpaper-glued down with tissue paper and painted over would probably also work. I once attached a multi-element FM antenna to a ceiling that way (antenna made of aluminium foil, then papered over and painted). Worked a lot better than a dangly wire, and a lot cheaper than getting a man to attach a proper antenna to the chimney. The room needed painting anyway.
Or with RPi at £30 ... go wireless?
Fairly challenging, but a case of open-source development once, fix hundreds of thousands of people's problem.
It ought to be possible to connect an Rpi, a solar panel, a battery, a weatherproof box, a USB hub, some USB wireless thingies and some directional antennae, to make a goes-anywhere wireless access point. Stick that on a pole in a field somewhere, with line of sight to another pole, and another, and another ... chain them together from a location that can get broadband, to a village or hamlet that can't.
Software that auto-assembles a network, so apart from configuring the broadband end the future problem abounts to errecting poles, screwing a box onto the top, and pointing the antennae in appropriate directions.
It's a cheap somewhat limited implementation of something that the military are playing with for battlefield communications, and disaster relief agencies for getting communications up again after a natural disaster.
Re: Thorium Breeder reactor
in particular, the Thorium cycle: breed U233 from Th232, and then fission the U233. Takes a lot more neutrons to get all the way up from U233 to Plutonium, so production of Plutonium will be minimal.
U235 production might be more of a worry, but it involves isotope separation to extract it rather than chemistry. Hard to build a Uranium isotope separation plant without being noticed.
Re: More adverts, everywhere.
So, don't patronise us and tell us to grow up for lamenting the removal of the option to exercise choice.
But your freedom to choose to block adverts has not been removed. It's just that the Google store has stopped stocking that app, so you have to get it from somewhere else and learn an extra trick or two to install it. To me, it's a bit like your default supermarket deciding not to stock one of your favourite products. Annoying.
If they'd made it impossible to block adverts without voiding your phone's warranty, THAT would be interfering with your freedom of choice. If they'd used cryptographic techniques so it was impossible to install anything they didn't want you to, and so they could retroactively take away something that they didn't want you to have , that would qualify as evil. Nobody could be that nasty, could they?... Oh.
We can't even work out how to communicate with cetaceans. What chance have we got with something with which we share no common ancestry at all?
(They sometimes save our lives by swimming us back to dry land. That shows a remarkable capacity for abstract reasoning by dolphins and orcas, and inter-species empathy in advance of our own. )
Re: No such thing as degrees Kelvin.
Which has just made me think, why haven't I ever seen use of kK or MK? (temperatures appropriate for describing stars).
(Unlike the Yg, which isn't NEARLY big enough).
Re: Anything we can do
Even if they happen to be within a few tens of light-years, even if they are currently watching Lucille Ball on Earth-TV and trying to work out what those strange techno-primitives are up to, they may still be thwarted by real-world physics. In other words, they've come to the same conclusion that our scientists have, that interstellar travel isn't ever going to happen outside of the movies. No wormholes, warp drives, ZPE drives, reactionless drives, or any of the other devices that SF authors use to get an interstellar plot going. Just rocketry and relativity and radiobiology, saying "can't be done".
My riposte to the (strong) Anthropic principle is that the universe is NOT optimised for Homo Sapiens. It's optimised for our silicon successors, who'll be able to slow their clock rates down so that a thousand light-years becomes a few years subjective. They'll prefer just about anywhere in the universe to those hot balls with moist oxidizing atmospheres, where some (mostly ridiculed) few amongst them think their ancestors must have developed, back before the Archives and the Memories.
In the worst-case scenario, they're here in our Oort clouds already, we're making far too much noise, and they're getting ready to drop a comet on us. (It worked last time, 65 Myears ago).
The day the earth caught fire ...
Premise is utterly bonkers. A bad fantasy. No science at all. Newton would have discovering relativity as a consequemnce of how fast he'd be spinning in his grave. Even "The day after Tomorrow" made more sense.
- Comment Renewable energy 'simply WON'T WORK': Top Google engineers
- Useless 'computer engineer' Barbie FIRED in three-way fsck row
- Game Theory Dragon Age Inquisition: Our chief weapons are...
- 'How a censorious and moralistic blogger ruined my evening'
- Amazon warming up 'cheapo web video' cannon to SINK Netflix