Posts by Voland's right hand 5759 publicly visible posts • joined 18 Aug 2011
For the same amount of money you:
0. Get the normal kindle.
1. Get the kindle cover - it is a must, even if an adult will be using it
2. Talk to your kind not to be stupid and remind him that the email invoice about any of his purchases lands straight into your email account.
Works fine - tested on a 13 (for 3 years now) and a 6 year old (for about a year). The only "downside" is your expense bill. It is surprising how quickly can the numbers rack up (especially if you teach your kinds speed reading). A month with a lot of travel can cost you something in the 90£+ for "book budget".
However, I will take that "downide" any day and double it :)
Surely "if you make lots of short journeys", a smaller car
Bingo. I will definitely consider a smaller car with this exact roadtrain design (provided it is not a BMW).
I was recently discussing with the SWMBO the candidates for a second small family vehicle if and when the Daihatsu Sirion MK3 she uses today will need to be converted into a spare parts bucket for the nearly identical Sirion 4x4 we have abroad. I suggested the electric Soul (she likes the conventional one). Well, her first question was - can it get you to Heathrow? At which point I juggled the 90 odd miles versus its spec-ed range in my head and parked the idea.
If, however, there was something in the Micra/Corsa/Yaris/i20 class which uses electricity as god intended (without a transmission) and has a decent petrol generator backup it would have been a candidate for "cash and carry".
And how is it different from let's say the wonderful geniuses from VW group which sell 4x4s in the UK and void your warranty if you change the tires with non-homologated ones. Makes owning one and having ideas to take it to the continent in winter quite interesting as the list in the UK is different compared to the continent and does not include a single winter tyre type. So you end up with a 4x4 which is not road legal in half of the Eu half of the year and is not fit for one of the main 4x4 purposes - to take you places where normal cars have a difficult getting to.
Not that BMW are much better either.
Bollocks
1. Did you inflate your tires to motorway pressure? The "book" pressure is for offroad / mixed driving. F.e. my Isuzu Denver 2007 manual does ~ 32 mpg at its preferred UK speed of 65 (reported by satnav), inflating the tires 20% above that gets this to 38.5. Similarly, 90 mph on the Autobahn 27 mpg with "book" pressure and 32 mpg with 20% above that. This is with Nexen tractor-like tires which have the most atrocious and fuel economy-unfriendly 4x4 thread you can think of. General Grabbers happily get you to 40mpg+ driving like a granny at 65 (the bloody Isuzu dashboard computer decides that its a fault and reports 40 from there onwards).
The new model adds 5mpg to that (6th gear is quite useful)
2. Is your load area covered? If you do not have a solid lid your aerodynamics go to hell. That is 5mpg at motorway speeds right there.
The new fuel economy rules cannot be met by something which has the aerodynamics of a shed on wheels. Let's face, cute as it may be the Defender has outlived its time. If you want an agricultural utility 4x4 these days you are better off with an Isuzu or L200.
Not sure if they are compliant to the new rules either, but they at least stand a chance to be (the Isuzu can do 45 mpg+, the L200 is not far behind).
The JVM is a fantastic piece of technology +1 to that
the Java concurrency libraries are written with a level of skill and elegance few people (including me) will ever reach. -1 one to that. In this day and age there is a simple maxima on all platforms it is (yes, I mean to shout): "MUTEXES AND REST OF POSIX THREADS WORK!!!" Java concurrency implementation is dependent on the underlying thread scheduler which java till this day does itself instead of fully relying on the OS (despite using OS threads for a while now). The consequence of that is that there is an embedded ancient piece of Solaris scheduler in it which constraints mutex rates, posix wait/notification rates and a few other key parameters in a modern app to a magic number which depends on the phase of the moon, the mood of the developer mother in law and a few other key parameters. In the meantime all OS-es and libraries as well as languages which rely on the OS native threading and notifications have moved on and are fully preemptive and NOHZ on most platforms.
You also forgot to add that this is all hobbled by dire core class implementations. I had to look at them several times recently and it is NUTS. In the day and age of nearly all CPUs having instruction sets for accelerated big integer processing (ffs, 8087 had that when it initially came out), the big numbers have no offload to JNI on supported platforms. Network functions like getnameinfo, etc are ifdefed and while present in the source remain unused and bogus obsolete analogues from the gehostby_* family are used instead - you name it. In fact, once again, my hat is off to the JVM bytecode interpreter and JIT developers, it is only because of their stellar work we do not notice how bad the core libs really are.
Exactly.
There is only one meaningful use case for IoT - preventative maintenance and there is not a single IoT enabled appliance on the market to do that at present.
You get the lot - from "kewl" glow of the "Nest thermostat", through greenie "consumption limitation" and into the outright in(s)ane like IoT blinkenlichten. None of them however computes at present despite all electronics and IT giants falling over each other to get something delusional built.
The most obvious application however - your [car | boiler | fridge | dishwasher | washing machine ] calling the engineer out for breakdown repairs is not there. Even if the maintenance is "on-fail" and not preventative it is considerably cheaper, more effective and less hassle than getting an engineer call out arranged and having it fixed or replaced. In fact, if it is less hassle I may consider it instead of fixing them myself most of the time.
Now ask the question in the context of you being a parasite specializing on feeding off Education IT (that was the niche Pi was supposed to be addressing as its main goal, right).
Before - you had to deal with Linux (horror), Kids being able to do write software which could have been used inside the school instead of the school paying you (double horror), the perspective of having to write off some of the investment into MS?E of your staff (quadruple horror), decrease in commissions from reselling Microsoft licenses (octuple horror).
AAAAAAAAAggrrhrhrhr... the Horror.
And, now, the cavalry comes led by the knight saviour on the white horse. And it carries the most wonderful news on its banner - it is not just Microsoft, it is stuffing the school with 2 licenses per student.
Chicken - egg, egg - chicken.
Managing multiple inputs, running the "school process" as a background thread which can preempt the "staring out of the window" and "reading a book under the desk" threads and managing distractions in general are 95%+ of getting high grades in school. It is also one of the reasons why girls tend to do significantly better academically up to a certain age.
They are simply better in juggling 10 things at the same time at that age. Based on non-scientific observations of my kids, the daughter can run 3-4 tasks at at the same time without botching them (f.e. doing homework while listening to music _AND_ watching pop-sci and pop-engineering shows on BBC iPlayer at the same time +/- a background thread playing Star Wars commander). Junior can handle at most one. Asking him to do two things in parallel is a recipe for disaster.
Robocallers do not use "joe average luser phone line". In the olden days it was an ISDN trunk, now it is usually a SIP trunk. If they are keeping it loaded (as robocallers do) it is trivial to pick it up as a robocall activity. The pattern of call setup + call frequency + call duration is quite easy to distinguish. Trivial Bayes on the call log should be able to pick it out with 99% certainty.
The problem is that the telcos are not willing to do anything about it. It is like assigning the cat to guard the canary. They get termination fees and they also sell their customer data to the marketeers which feed it to the robocaller outfits.
In any case, to put it bluntly, these guys were stupid. It has been the norm to set up robocaller operations outside the country jurisdiction and use VOIP for a very long time now.
Enough of the CSMA/CD
While the spec for using the frequency does not spell out CSMA/CD, it says that you are entitled to use _ONLY_ X% (under 5 if memory serves me right) duty cycle. You are _NOT_ allowed to xmit non-stop.
I tried to get that idea through to Ofcom by the way, but they did not give a damn. Basically, that clause which is present for most M2M frequencies (the 900MHz ZigBee band, the bands used by alarms, etc) is not being enforced.
We had that on our street a few years back with the added benefit of houser alarms registering a jamming attempt and activating in a 30+m radius. The culprit was traced to a duff Toyota keyfob. The owner gave the keys to his toddler to play with, the kid was teething, bit on the fob and some spit permanently shorted the ON button. Apparently there was no delay and no protection in the Toyota fob design. You short the ON and it is ON - all the time transmitting at max rate. Perfect DIY jammer.
So it is not necessarily a jammer. A mummy with a toddler driving a Toyota is equally possible.
95% of all vehicles sold are scrapped long before they reach their full mileage projection.
An autonomous self-driving vehicle should be able to clock more than 400k with ease. Compared to that a human driven vehicle will end up being scrapped < 150. It is staying in the parking lot with a layer of salt on it which kills most cars in USA and Europe, not the actual driving.
As far as maintenance - electrical drive train and autonomous driving are a marriage made in heaven (or hell if you hold GM stock). Less than 5% of the maintenance of a normal vehicle. The recharge, etc issues also become irrelevant as the vehicle proceeds to charge itself when it is unoccupied.
wooowwwwwww easy Tiger 4mb on a 286!
Mine had 8 and ran at 25MHz - a Harris surplus military chip + VLSI motherboard with working all banks memory interleave. Add to that the fact that I got p*** off at Miscrosf ineptitude at handling 1024+ cylinders and wrote my own disk drivers for the past-1024 boundary partition which were "only" 4 times faster and well... the picture was complete. All of that with a black and white 1024x768 maskless XGA for budgetary reasons.
It ran circles around all early 386es and the only reason I upgraded to 386 at all was the first installment of Tie Fighter. That bloody thing categorically refused to run on 286.
Do not get me started on how long is an average (heavily polished for PR purposes) bug squash list in an average Adobe bug-fix. And that is only stuff they decide to publish, not what they fix and do not add to the publically visible list.
700 issues (as these include features, etc) in 4 years is _NOTHING_ in a major project. In fact it is suspiciosly low.
Unless the FAA starts reporting to Lord Vetinary - a while yet.
The El Reg team is seriously handicapped here - by not being local and/or not having a few billions in the bank. So they have no strings to pull.
IMHO they should have tested somewhere where suitable lobbying by el-reg readership would have been provided. A few Eastern European countries come to mind as suitable candidates. Australia probably would have been a better choice as well. Even Canada is a better choice (hello Amazon). USA was the wrong choice to start with. In fact, my suggestion would be to repack to the crate and ship it to somewhere not under FAA jurisdiction ASAP.
One thing about the War On Terror, it is making us terrified!
It is supposed to. Read more on Stalin's rise to power and why he got as far as he did, it provides a good explanation on the necessity to keep the population terrified of "something".
You are not the only one by the way. I had a similar incident with openwrt and a (lidless) TP3020 portable access point with the console plugged in. From there on I work only harmless stuff on planes such as virtualization, java, etc. Real work - that is for the desk back home.
Why run Windows at all, given the supported software list?
Because there are limits to Nadella's "liberal orientation". The primary use case for this server is cloud deployments and Azure. Suggesting throwing out Windows as the base OS for Azure. Sure, me and you may agree, but try doing that one in Redmond.
What is interesting is preparation of this OS for customer release. I smell a BIG "Azure On Premises" and Hybrid Cloud push coming up.
Market is mature and saturated - same as PCs and laptops. It is only like for like replacement now and the reasons to replace are no longer pressing.
I have a stack of 2 semi-retired, 1 in minimal use and 2 in use in the household. The ones in use are one year old models and they still run everything fine. They are a staggering improvement on the 2 year old ones which predated them by only a year. However if you compare that to the current crop the incremental improvement is no longer staggering - it is marginal. The product has matured.
It will take 2-3 more years for the ones in use today until the batteries, charger plugs or something else gives up the ghost. Then I will consider replacement. Before that? Why should I?
Actually united ones are semi-useful. You can book something with them even during peak periods. I use them regularly for flights to Euirope in mid-winter when the prices go through the roof because of the skiers.
Try that with BA or Iberia - there is zero availability during time you actually need them and nearly zero availability to book more than 2 people on a flight at a time.
It is not a useful "reward" for a bug bounty though.
For a variety of historic reasons people use the term KVM where the correct name should be QEMU. KVM is just the x86 virtualization accel for QEMU. The QEMU codebase still handles most of the IO, memory management, etc.
By the way, some of the QEMU codebase is now reused in Xen too (if memory serves me right).
The aim of a company is to make profits that the shareholders can cash in and consume as beer 'n' burgers.
To do that, the money has to be taken back into the home tax jurisdiction of the company,
No it does not. 99% of the "profit" to the shareholder in USA from shares manipulation, not from dividends or other conventional Adam Smith Economics (TM) ROI.
Shares in this day and age are manipulated (grow or fall) on the basis of mere promise for return. In order for the share value to go up the money does not need to be repatriated. Apple is a textbook case. It has not repatriated any income for a decade or so, but its shares are growing proportional to that income on the mere promise that this income will be repatriated or otherwise converted into something real one day. There may be some repatriation at the end of the rainbow. Or maybe not. The shareholders are happy extracting value from the shares growth on the market instead of getting dividend and as long as this is the case no repatriation will be forthcoming. Ever.
Another example. Miscrosoft. It paid its first dividend ~ 20 years or so after the company was founded. Another example... and another... and another...
Granted, this is not the case for a lot of Eu companies - in most Eu countries there are various accounting and taxation regs which make not paying dividend quite painful. If the subject under discussion is a USA company... well... all you need is to trawl through the filings on NASDAQ and see exactly how much dividends did most of them pay in the last few years. Hint - not a lot.
Also, it becomes visible instantaneously the moment it prepares to fire. All the wonderful radar signature decrease magic goes straight to hell at that point.
Serbians picked it up from 60km out during the NATO bombing campaign and nailed it for 6 at that point.
In fact, when you take into account that the time to target for a lot of AA missiles is under 30 seconds the whole idea of stealth starts to look a bit ridiculous.
Yep. It is one of the only two birds I have seen which successfully outcompetes pidgeons in urban areas. The other one (equally loud and obnoxious) is the Great Tailed Grackle in Texas.
It will be interesting who the winner will be when these two pests from hell meet. My bets are on the grackle.
Question is: will it SPARC any interest?
It does not need to. It is more than sufficient to supply with non-embargo compute the parts of their industry which are subject to a western embargo: energy sector, military, etc.
Europe already seems well ARMed with other architectures, so why take the RISC? IPR can be embargoed too.
They'll either MIPS their sales targets or just not bother If this was two years ago - I would have said the same. Today, not so sure. I would not be surprised to see a mandatory school and education use mandate. That is a few million units on its own.
@ Ilmarinen
You cannot be part of the EEA without subscribing to:
1. Free movement of labour
2. Most of EU environmental regulation and a lot of the economic ones.
What Cameron and company wants is to restrict these which means _LEAVING_ the EEA period. The Swiss which already heading down that route as a result of a botched referendum and will learn the consequences the hard way next year. One of the reasons why Eu will play hardball on that one is exactly that - Britain. We will all see exactly where happens as a result of restricting labour flow when this happens because Switherland currently has the same problems as UK and will get the same painful awakening which awaits UK in case of Brexit:
1. Their health system is run on more than 50% Eu labour. It will collapse overnight if this is withdrawn
2. Various services and non-banking parts of the economy similarly heavily use Eu labour.
It is all nice to have half of the banks in the world producing GDP (thought they usually hide it instead) when you are in an operating theater and the Romanian anaesthesiologist, German surgeon and Italian head nurse are not there to attend to you. While we can live without Bulgarian and Romanians washing our cars, that particular bit is a bit difficult to live without.
By the way - I am not being flippant here, when both of my kids were born, I did not notice any British natives at consultant level, there was on trainee doctor and some of the midwives. It was exactly the case of German surgeon, Romanian anaesthesiologist and Italian head nurse. So the observation that NHS is in the same boat (or worse) than the Swiss health service is a first hand one.
Definitely voodoo math.
If the failure is once every 12.5 TB (as claimed), a nearly full 1TB RAID1 set where most sectors contain non-zero values (like the one I have on my old main house server) should fail during the weekly verify job at least once every few months. That set has been running for ~ 3 years now (it just got replaced by a new 2TB one). I need a few more coffees to recall the relevant formulae from probability theory, but it should be somewhere north of 95% for a failure during the 3 years of use.
There is definitely something wrong with the cited numbers.
Agree. I did the math recently for new house RAID array and went for RAID 1 instead. The hassle, combined with failure probabilities and rebuild time for 5, 6, 50 or 60 is just not worth it. You are better off throwing in a couple of drives as a DIY MAID on a spare machine and doing nightly backups to that.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2024
