I get the impression this is more about "state" vs "federal" turf war than anything to do with providing good and competitive internet to the good citizens.
165 posts • joined 26 Feb 2008
How can a company with not much history (it's a young company), which is losing $60M a year, at an apparently increasing rate, be worth $700M. *Boogle*. If it is "just software" it's not even like they have a monopoly on "analytics in the cloud".
Tech Bubble 2.0.
High Dynamic Range
Re: Pete H
That was my point - the bus is still there, you are still moving data over it. If processor A has the data and processor B needs the data the only way to get it is to copy it over a bus (on chip or off chip, it makes no difference - there are always wires linking things together, it is just a question of scale).
All this spec seemingly changes is the need for manually keeping the different memory pools in sync - the "clever bit" is the reliance of system level hardware cache coherency - not the removal of the need to move data around the system (which is impossible).
Sorry if I didn't state the point in the most "smooth" way ...
> without having to copy chunks of data over buses, for example
Unless they have just invented a chip design which works on quantum entanglement of particles, so data can magically jump from the CPU to the GPU caches or to main memory as required, I'm pretty sure that moving data over busses is still firmly in the picture.
> I think you may be making some incorrect assumptions about what these certificates mean. They are purely and simply there to certify that a given web site belongs to a given organisation.
Yes, but they don't do this directly. The entire system is based on the fact that there is a "trusted authority" which can attest that the entire certificate chain from root to signer is valid. You implicitly have to trust everyone in that chain not to have leaked a key, or the security of the chain is DOA.
The entire scheme is trust based, and the OP is correct - most of those certificate signers I wouldn't trust at all (I have never head of them, so why would I trust them or their processes to keep their signing keys safe?). However, as most websites are only signed via one CA you don't get a choice - you either accept the certificate or go elsewhere (which is a choice, just not a very useful one).
Personally I would say that there isn't enough visibility of how any of the certificate vendors or their processes actually work, so the security processes involved are totally opaque. That's not really trust - that's just taking everything on faith - and that's no real way to build a security system.
Re: Oh dear...
It'll be fine on American roads without them ...
It's entirely possible to get very small parts of a design running VERY hot. The average chip temp will be much lower, but hot spots can cause massive problems as dissipating heat out of an area only a few 100 nm across is hard.. I seem to remember some Intel slides on Pentium4 claiming there were parts of the design around the same thermal density as a SaturnV rocket nozzle. We all saw how well that turned out for that micoarchitecture - hot spot problems were so bad it got shot.
Re: Does it matter?
> Combine that with a single flat rate tax system (possibly with a tax exempt threshold), and suddenly the cost of administering the system becomes tiny compared to what it is at the moment.
It doesn't work for a sizeable chunk of the system. There are a large number of exceptions - people who are severely ill and/or disabled - where £130 doesn't cover expenses (i.e. needing 24 hour care). Hell, around here it wouldn't even cover rent in social housing, so those out of a job would still need support.
Assuming you are giving the parents the money for the kids (it was cradle to grave, right?), it also encourages parents who shouldn't be having kids to have more of them just to get the £££. If you don't pay parents for the kids, then you get a much worse poverty trap as today with parents on benefits with 5 kids. Unless you start wanting to forcefully sterilize portions of the population, this is an "interesting" route to go down.
Society is full of corner cases - the problems with policies like this is they try and treat everyone as if they were "average" - it simply isn't true.
Re: So Stupid
... and neither is my wife.
> "I only had S3 keys on my GitHub and they were gone within five minutes!"
Assuming they got comitted to a git repo, I wonder if there were still in the version history. Seen that happen a few times on internal repos - people delete the file from HEAD, but nuking the history totally is much harder.
Re: Most people never had a choice . . .
> I think the argument against IE is relevant only in that it controls the complete stack and allows (allowed) MS to pretend (or insist) that IE is essential to the proper operation of the OS.
The real issue was that Microsoft deliberately did not follow the defined standards so only their browser worked for the majority of sites - it was an wilful abuse of a dominant market position, and made compatibility for other software difficult.
As a result of the legal ruling, and increased competition, Microsoft were basically forced to start paying more attention to the standards. Still not perfect, but a lot better than they were ...
Re: "the fact that it meshes in so well with bash and the command line"
> WHY THE FUCKING FUCKITY FUCK WOULD I WANT A COMMAND LINE!!!!?!?!?!!
TortoiseGIT for Windows?
EGit for Eclipse?
> And it has no Windows support.
I mean "Git Windows" on Google throws up about 3 or 4 different download sites at the first downloads. "No support", sure.
> with no documentation
Sometimes I think organizations roll out tools without teaching their staff the first thing about how to use them. WHAT THE FUCK HAS HAPPENED TO THE FUCKING WORLD???
> I do hope they come up with a new flavour of the month to replace git soon, as its a total stinking pile of shite.
Out of curiosity, what's wrong with it? I've used most of the "free" ones over the years (CVS, SVN, mercurial, GIT), and managed to avoid the uber systems (Perforce, etc), and TBH git seems to be the best of those that I have tried.
It's fast, flexible, and generally comes with a whole pile of features which I can't get in the others (although mercurial has some of them). Once I embraced the concept of a local repo with local commits for my working copy I suddenly became a whole load more productive. Just don't try to use it like a drop-in swap for a traditional CVS/SVN - it will drive you crazy and you'll never actually get any of the benefits.
Is it perfect? No - I miss having a global incrementing version for trunk (my only feature gripe - and one which I have "emulated" simply by tracking the hash history over time for each major branch I maintain), and it does have a learning curve, but it is one hell of a powerful tool.
Re: The fad now leaving from platform 4 ...
> Once your TV picture has a dynamic range that exceeds that of our eyes, without them dilating all the excess DR is wasted.
Not all the dynamic range in HDR needs to be "brighter" - you generally get more quantized levels available in the between the existing "LDR" min and max luma levels. In dark scenes where you eye is already dilated this allows the HDR scene to show subtle detail which is not visible in an LDR encoding.
Yes, ambient light doesn't help this (eye dilation isn't only determined by the scene), but TBH many people do watch films (the most likely HDR candidates people will actually care about) with the lights off ...
Re: I'll wait thanks
> So far their plan has been ... <snip> ... 4k ...
I'm not complaining about 4K, not because I want a better telly (hardly ever watch the one I have), but because it is finally meaning decent resolution panels being produced in volume, so it may finally kick 1080p out of PC and laptop displays. PC monitors have been static in terms of resolution for 10 years thanks to TV determining the "economics" and TV moving to something better is the only way to unblock that.
4k2k monitor - yes please. 4k2k TV - whooo ultra high definition HDR amplified compression artefacts on freeview - wonderful.
Re: Spec creep?
> It's really a minority of people who have that though - otherwise they wouldn't be able to have darkened cinemas.
You generally don't move your head in a cinema - it's a different problem. If all the VR environments provide is a 3D scene without head tracking, then you'd be right, but IIRC half the point of the head-mounted display is that you can look around.
Re: What ... you mean your notebook doesn't have a 4K screen?
I'd be happy with a full 1080p screen in a notebook ... 1376x768 is still all too common ...
Money owed > Money in bank.
Chapter 11 is designed to let companies restructure before they implode into an unrecoverable mess. If you only try it when you have $0 in the bank then you're unlikely to have enough time to recover anything.
Re: Coincidence or by Design?
> And gosh they are tiny chips, 0.1mm square is nothing for such a useful chip.
Worth noting that's just the CPU area _in_ the chip - not the whole chip. Someone will have to bolt on peripherals, bus, internal memories, pin out pads, etc to form the whole chip which an end developer would buy.
That said, in reality at the "small end" the logic gate area is significantly smaller than the area needed for pin out (I think the smallest I've seen is 2mm x 2mm, which was achieved by eschewing packaging and just dipping the silicon in a ceramic paint to insulate and protect it).
Re: What industry do these guys work in ? Clearly not a tech one.
> A process that could be easily devolved to a 3D Printer - which are hardly like rocking horse shit.
Most 3D printers I've seen can't capture the level of detail required for a finger print - not the mass-market squirty plastic kind anyway.
As above - on many sensors blutack or plasticine works just fine. Low tech and far easier ...
> If there's an incentive, things will be made simple. Look at the grunts who skim cards, using quite high tech.
True, but that's a create once and use many kind of operation. Getting one fingerprint of one guy unlocks one phone - it's not scalable to quite the same extent.
That said, if you really really want to get into one guys phone hacking is a lot of bother - just threaten to hit him with a wrench until he gives you the pin number, or cut off his finger. Attack the fleshy part - not the technology - it's normally simpler.
> attackers shouldn't be able to determine the password even if they know what method you use to create it
Yes, if every user applies a cryptographically sound key generation technique this is indeed the theory. I think Bruce's general point is that if you ask the average punter to string 3 words together the amount of entropy is far from what it looks like, and so the algorithm is not cryptographically sound - you either need longer key lengths (more words), or a new algorithm.
> Why, that's less than a week!
I assumed that he's talking about rainbow table attacks working backwards from a website which has leaked a hash, the issue is not processing time (you only have to generate the rainbow table once per salt), but the amount of space it takes to store the tables.
Assuming most users pick from a relatively small pool of words in common use, it's not an insurmountable search space. Yes, the total search space in theory is pretty enormous but, I suspect many people have a password with "Cat" in, and many more with "Dog", and more than a few with "Password". Hackers only have to skim off the 1% of the easiest passwords to get into a system- not the 1% of the geeks with the uber difficult ones which are hard to remember ;P
I read more as "there is no point investing in really expensive locks (long passwords) if everyone makes the safe out paper (i.e. the other security is implemented wrong)".
That said their approach is a race to the bottom - lets make everything as weak as the weakest part - rather than trying to improve industry best practise to reduce the number of "implemented wrong" instances in the wild.
Re: Drop in Ipad sales / saturated market?
I'm still using my iPad 1 - it gets used daily on my commute, and basically is as good a new and still has a week-long battery life.
The only reason I may consider an upgrade is that they's stopped providing iOS updates for it, so some new apps don't work if they require iOS-latest. However, given the price of a new one I really can't justify it, and I don't want to encourage companies to stop supporting perfectly functional hardware =) Given that most apps work on the current one just fine and the hardware is basically in perfect condition, an upgrade just seems like spending for the sake of it.
Re: Economics 101 - accounting using spreadsheets
> You never had an economics class did you?
Yep I did GCSE economics - a long time time ago ...
Modern macro-economics is all about macro-scale models, and data mining - would have made it far more interesting =) Micro economics (i.e. business level) is pretty much accounting, albeit not spun that way, but the basic principles are the same.
If you can't find some means to get some "computing" in to at least one aspect of an economics course you're not trying hard enough - there are lots of applicabilities to "big data" and statistical trend and correlation analysis in many real world applications of economics (insurance, crime, risk analysis, etc - it's all data mining).
I'd much rather that "coding" was to some extent introduced as part of the syllabus for other subjects.
Welcome to biology - today we're going to data mine some gene sequence information from a database. Welcome to chemistry - we're going to use some image processing tools to look at some slow-motion reaction images, or to automate some process. Economics 101 - accounting using spreadsheets, and setting up some monthly report generation scripts.
You teach coding to kids who don't like computers (who would normally hate CS on principle), and you teach some non-coding skills to the guys who love to code (who would hate subject X on principle). Done right it should be "win-win". You obviously can't do this in every lesson - but it teaches what coding is really for in industry - and how that CS is a tool which solves problems in other subjects, and not a solution in its own right (in general).
> Teaching how to break down a problem into ever smaller parts until each becomes manageable is a skill that I dare say should be valuable for *everyone*. I believe that coding is a very good vehicle for that.
The entirety of the education system _should_ be geared towards that. What's the point of all of the "knowledge" which all of the existing subjects are trying to cram in our students' heads, if they can't actually apply it to a problem and do something useful with it.
Unfortunately we have politicians who seem to equate "education" with learning as much general knowledge trivia as possible, followed by a testing methodology which is to regurgitate as much of it verbatim in 2 hours as possible. Very little of our education system encourages application of existing knowledge to new problems - indeed most of the time any "outside of the box" thinking is actively punished because the answers are not on the proscribed list of tick boxes which give marks.
You don't need to teach coding to teach "original thought" and "problem solving". My nephew "likes computers" (i.e. he plays games) - but he's not interested in coding - it's too abstract. If you want to teach something in schools, and teach it to everyone, you need to make it widely accessible to > 80% of the class. Coding won't give that, unless you wrap it up so much (logo turtles anyone) that it isn't really coding any more - so don't pretend that it is (or that it needs to be).
Re: Good Lord, we are all doomed
> who don't let you get away with submitting half-baked papers and essays.
What, you mean, actually _fail_ a student? ... but, but, ... that would ruin the statistics.
My Dad ran his own company for years - ran a very similar "aptitude" test which wasn't particularly sensitive to what course you did at university, etc. He was basically looking for the one skill you can't teach per-se - common sense - and the right attitude - a work ethic. Unless you are working in a very "top right" industry - you can teach most of the "skills "people needs on the job, but they have to want to learn it, which can be a harder thing to find.
Not Cortex-A15 ...
> The Qualcomm Snapdraqon, based on an ARM Cortex-A15,
It's not based on Cortex-A15. Krait is Qualcomm's own CPU implementation (instruction set compatible, different microarchitecture).
Right, so ...
> and the programmer whose error caused all the ruckus says there just aren't enough people scrutinizing the OpenSSL code to spot difficult-to-find bugs
... and forking it and making it OpenBSD-only is going to improve number of eyes looking at the code how?
Re: Yowser (info needed)
> At 3Ghz that's about 30 flop/cycle. How? If that was a multiply-accumulate (fmac) heavily pipelined over 2 vectors of effectively infinite length then that's 2 cycles/flop, say you have 2 such units that's 4...
Dual issue vec4 SSE = 8 per core per clock, add in a couple of non-SSE fp32 instructions = 10 per core per clock. 4 cores = 40 per clock.
Re: more lessons
> Good advice: my only improvement on that is to get a cashback credit card if you can. You can easily make 3 figures a year just from funnelling payment for things you buy anyway through the card(s).
Oh dear, another one who thinks this money grows on trees.
What really happens is that the CC company screws the merchant via transaction fees, who then increases the price of goods you were buying in the first place to cover it. Nothing banks do is ever designed to actually give you money which they haven't managed to screw out of someone else first.
Personally I'd rather the world banned these "freebies" and actually forced banks to compete on their ability to deliver a banking service, and nothing else, in particular for credit cards where the actual cost is invisible to the punter (and therefore these not-so-freebies actually look "free", unless you understand how the model actually works).
Re: Using Sinofsky As A Scapegoat
... and in principle there is nothing wrong with that goal, and from their business point of view it seems to make sense. They just made a mess of execution - the devil was in the detail for both.
I don't agree with the point that they should have started with "a new OS". In general people were happy with the old OS (it was faster than the old one, more stable than ever before, finally fixed their driver model, and generally worked). Not liking a few aspects of the UI is not a reason to bin the whole operating system ...
What it needed was continued incremental improvement on performance, memory, and stability, with _optional_ compatibility with RT/Phone, not a whole sale crammed down your face for compatibility when it wasn't actually compatible at all ...
Re: Excessive copyright term
> Big companies (DISNEY and others) get the copyright term extended whenever one of their moneymakers is nearing the end of its copyright period (the 1936 Mickey Mouse film is still in copyright!!
On think I never understood is why people thing this is this such a "bad thing"?
There is little public good (it's not a cure for cancer), it's hardly a cultural game changer which will change the world, and nothing is stopping people developing another mouse-alike cartoon, provided they don't slavishly copy it or infringe trademarks.
Why should people have a right to copy (i.e. get for free) what someone else spent time producing, just because it is digital?
We seem to be entering an age where youngsters are far more interested in getting other people's old stuff for free rather than learning how to create something themselves - cultural stagnation starts here ...
Re: Wolfpack, anyone?
> Once they go feet dry, though, some sort of self-destruct/auto-immolation might be desired, but might conflict with safe handling and storage/stowage aboard a sub.
As opposed to the warheads in a MK48, or a cruise missile? A small auto-destruct on the electronics of a drone is no problem to stow ...
It's actually quite good
I've bought one and TBH it's a really nice console. Sure, there are a few teething bugs (random UI elements like dropdown menus lose all text, and a couple of system reboots playing Ryse), but after a weekend gaming on it it seems like a nice improvement over the XBox 360.
The most noticeable change for me is noise - it's practically silent, making it a viable platform for watching films - on the 360 you had to increase the volume to drown out fans.
Is it the best thing ever - no, of course not. A big enough improvement to justify moving over - probably not at its current price point for most people. However, I just like toys, and I'm happy ...
I have a couple of these ...
... CPU's not as powerful (only an ARM Cortex-M microcontroller) but it is trivial to whack in a breadboard and hack together some fun projects interacting with the "real world" - no soldering required.
The whole toolchain is web-based and you download over USB from the browser (just drag and drop the program to the flashdisk). Not as flexible as some of its bigger brothers, but for teaching kids it hides all of the grotty bits they don't need to know about (yet) and lets them focus on the fun stuff.
Re: "Who remembers netbooks?"
> i5, 8gb ram, 128gb ssd and a 9inch screen. No fancy graphics, just a portable workhorse. Few usb 3 ports, an SD slot and hdmi out(and in?) Oh, and not stupidly expensive!
So basically you want everything a laptop has - except the optical drive and GPU - but smaller? The cost of a higher density battery to give the same battery life in a smaller form factor probably costs as much as the bits you ripped out.
... because on iOS the file management is a complete cluster fuck* which means that I have to plug my tablet into a PC if I want to move files or to or off it it. If you want to transfer from tablet to tablet then you probably can't if you don't also have a laptop. This makes it a bit of pain by itself ...
The ability to push files to a USB key natively without needing iTunes from the device (or pull from a key) would be "really useful". I mean hell, its got a normal USB port and I can plug in a printer - sweet =)
* Caveat - it may have improved since I last tried - I gave up on my iPad for anything involving more than 1 or 2 files and switched to my laptop, as it was generally a much less painful experience. Nothing has yet convinced me to try again ...
Oh no, it will be worse than that ...
The energy companies will send some toughs around in really badly fitting suits, and make comments along the lines of "gee, it would be a real shame if the lights started going out around here", and good old tax payer will somehow end up subsidizing them too.
We are already paying subsidies for "green" sources and nuclear, so why not those to ...
Yes, that last comment was in <sarc> tags ...
Re: Equal footing with Uncle Sam
> in a neutral country
"Neutral country" is simply a transient state of politics and their current economics. You get one crazy bastard in power who thinks he could use this as a lever for political/economic/personal gain and then you're screwed.
Re: I'll give him a B- grade
> It is not helped the despite the best attempts of the apple spin doctors, the 5C is already seen as the lesser cheap phone
It _is_ the lesser cheap phone. That's because it had less features, sightly less expensive materials, and well, it's cheaper. It's just not lesser-cheap enough to keep the reviewers happy. Apple can't win on this one it seems (too cheap and I predict they would have gotten kicked for producing a crap phone which too few features, and too expensive they get kicked for not cheap enough).
The interesting data seems to be that rather than buying the 5C punters are buying the 5S. Far from buying a competitor product and abandoning Apple, the 5S shipments are looping excellent. My only complaint is the psychedelic icons and the parallax gimmic - it's pointless, just like window animations on Ubuntu. At least you can turn (most) of it off.
Looking forward to the next one ...
I'm looking forward to the next post on timezones. If you can handle strings then you are, possibly, ready to handle world time. Any computer structure as fundamental as time (in most real-world systems anyway) which can be screwed by politicans deciding they want to do it differently this year, is always good for a giggle.
Re: Stuck in the past
My RAM does feel exploited ...
... oh ...
Re: Am I the only one
Re: He is effectively predicting 3D won't work out
There is a big difference between flash memory and logic gates though (density, manufacturing process, clock speeds, etc). A flash die puts out a lot less heat than a CPU die.
Thermals provide the end constraint of these systems - how do you get heat out of the middle layers? Sure you could drop the frequency to reduce the heat output, but that kind defeats the point of adding more transistors in the first place, as there are limits to how well you can parallelise so slow and wide isn't always a win.
Re: hang on...
One footnote -> if it isn't entirely in a 1:1 orbital tidal lock then the friction of the tide through any rock will itself generate vast amounts of heat. At least one of the moons in our solar system (one of Jupiter's IIRC) has a molten cores long after it should have cooled, and that's entirely down to tidal effects though rock.
Re: hang on...
Tidal locking - yes, I would have thought so to some degree, although it does't have to be a 1:1 lock. A lot depends on orbit shape. Mercury for example is "sort of tidally locked" - it has a stable 3 rotations for every 2 solar orbits; due to it's eccentric orbit this is a stable gravitational resonance point.
Baking vs Frozen - stick a nice thick frozen burger under a grill and tell me if the underside is frozen after 20 minutes. That close to the star convection through the surface from the light side should outweigh radiant losses from the dark side. The star will be dumping a _lot_ of energy into it and it has to go somewhere.
Presumably that close to the star it never cooled that much and has stayed pretty close to molten all the way along ...
> Passpoint also identifies the user, so the operator can provide value-added services such as additional content without having to resort to a cellular connection, but that's of unknown value for the moment at least.
Unknown value to the user, but "quite a lot of value" to the passport providing company. They'll know who you are, what you are buying (you were on a Starbucks WiFi, right?), and pretty much exactly where you are all day as your devices hop from hotspot to hotspot. Marketing gold mine, and ever so slightly creepy.
Tin foil phone covers anyone?