50 posts • joined Wednesday 13th July 2011 08:29 GMT
@Fuzz The trick here is that the files are _not_ encrypted with different keys. Each file has a per-file symmetric key which is generated when the file is first uploaded. When the uploader wants to share the file, they share this key using PKI to protect it. Since the PKI transaction is all done client side, Mega have no way of intercepting the per-file key and decrypting the files - but do end up with two files on their system which have the same contents and the same key which can therefore be deduped.
As for the no password recovery - the whole point about this system is that Mega _never see_ the password to a user's master key because it is all generated client side. The fact that they can't do password recovery is actually a good sign here (modulo the entropy issues).
Whatever you might think of Kim Dotcom, I can't help thinking that he's got some smarter people working for him than many of the self-appointed security experts who seem incapable of understanding these basic points...
It just might work...
So what you're saying is that Nokia need an army of enthusiastic Lumia equipped techheads. The kind who spend most of their lives in the pub showing off their latest gadgets to anyone and everyone. Now all you need to do is persuade Nokia marketing to ditch the poncy faked advertising videos and spend the money on free Lumia phones for all Reg commentards instead!
Beer (as in free).
Already there - check out the WebView class in JavaFX. I know there are a lot of legacy Swing based applications around, but any new project should really be looking to use JavaFX for UI work - it's a huge step forward.
Re: £200m just for 'coaching'?
The amount of money available through the TSB for 'Technology Inspired Innovation' grants (ie, helping small businesses with early stage research) is 2 million quid and the competition is massively oversubscribed. The amount of money just announced for 'coaching' those businesses is 200 million quid for a service that most will never use. Anyone else see a problem here?
Re: "Pinterest’s vision and Rakuten’s model for e-commerce"
I doubt that the bursting of bubble 2.0 is going to make much difference when it comes to getting funding for 'genuinely worthwhile' startups in the future. The dumb money that's funding the bubble would never have invested in them in the first place.
Is this a Steve Bong! article, or did I miss something? Does this virtual tube have a stop in Shoreditch? It's now getting hard to tell the difference between big corporation marketing bollocks, government-funded Silicon Roundabout hype and attempted satire. Please make it stop!
Some things are beyond satire...
The real problem faced by Mr Bong is that no matter how hard he tries to be completely OTT with his humour, there are already too many people out there saying this kind of stuff for real. The same kind of people who mistook Nathan Barley for a lifestyle documentary and got confused when they couldn't find anywhere to sell them a Wasp T12 Speechtool.
Re: A couple of "really not corrects"
> The sad thing about this is that it all started because Sun chose to fragment Java into a multiplicity of editions, and flat-out refused to license Java SE for any handheld device.
I don't think that producing the different variants for different applications was too much of an issue. At the time, Java ME was fine for low-end devices and a 'full' SE implementation was never going to make much sense anywhere except the desktop. To me, the sad thing was the way in which Sun carried on insisting that Java ME was still good enough for the new generation of mobile devices when it clearly wasn't up to scratch. Faced with that, you can understand why Google decided to fork off and build Android instead.
Re: > sales chief falls overboard
No doubt he held on to the extensive list of high-level contacts at the various carriers and retailers which he must have built up over the years. I'm sure that will come in useful when it comes to getting a new job...
What if Oracle choose to sue for GPL violation?
I mean, Sun/Oracle did release the source code under GPL. So if they were to claim that Google's alleged copyright infringement actions were in breach of that license, would we finally see the GPL tested in court with the FSF arguing that their own license is invalid? And would this create a singularity in the American legal system, causing the whole thing to implode?
Joke icon - because someone, somewhere would otherwise mistake this for a serious legal opinion!
Re: Awesome piece...
More than that, any chance of doing some standup? Sorry, I meant to say any chance of doing some exclusive group mentoring and inspirationeering? In a place that serves beer, obviously!
I beg to disagree...
I'm sorry, but if you take an objective look at the way Google manages Android and Oracle manages Java, Oracle are still doing more things right. For starters, all the major development work on Java is done in the open via OpenJDK, which allows anyone to get their hands on the most up to date code. Compare that to Android, where Google plays favourites with the device manufacturers and then may or may not release the latest code to the rest of us once their favoured partners have got a head-start in the market. Then there's community involvement. For all it's flaws, the JCP still provides a way for people outside of Oracle to have an influence on the direction of the project. With Android there is no community input whatsoever - you eat whatever sugar laden concoction Google says you should eat.
Oops - I've just gone and defended Oracle on the Reg forums. Downvote away!
Not really from Australia...
This can't be an Australian project unless they're bending the Kickstarter rules a bit. From the Kickstarter T's and C's, you need to be based in the US and supply a US social security number to register a project.
Not really 'laboratory equipment'
You can avoid the need for FCC certification on development boards if they are are only intended for test and evaluation use in a laboratory environment. I assume that the requirements for CE exception are fairly similar. Most development boards I've used can legitimately take advantage of this - even if they are mainly targeted at hobbyists. However, once you start pitching the product for use by the general public you really need to have all the paperwork in order...
You just don't hear about the 'boring' cases
I'm sure the reason we only read about the 'elite' hacking cases is that those are the only ones that have any news interest. 'Police nab stupid bloke for trying to sell fraudulent stuff on eBay' isn't likely to make the front pages.
There is also the issue of readily available attack tools. As an analogy, a gun criminal doesn't need a PhD in ballistics - they just need an IQ large enough to know which end is the 'shooty' bit. But on the law enforcement side, you need a highly qualified forensics team to prove that a particular bullet came from a particular gun fired by a particular person if you want to prosecute a case. I don't see why cyber-crime shouldn't be subject to the same asymmetry of expertise.
Re: Not exactly
"The entire price for dev tools, especially in the FPGA market should be zero."
I'm sure the FPGA vendors would love that. Unfortunately, their upstream tool vendors probably aren't as keen on the idea. I've no doubt that this is behind Xilinx's decision to replace Modelsim with their own simulation tools.
Back to talking about ARM processors, one of the most interesting angles out there is the Microsemi (ie, Actel) take on the Cortex-M1. They roll the ARM royalty into the price of their M1 enabled devices so that there is no upfront license fee or royalty required from the end user. It means you can basically design your own fully customized Cortex-M1 based SoC for the price of a 100 quid dev board. Quite how Intel think they're going to compete with that, I'm not sure!
Re: Not exactly
I don't think that the up-front cost for embedded ARM development work is that high. For example, the eCos RTOS with GCC compiler won't cost you anything and is on a par with many commercial alternatives. If you don't mind vendor lock-in, most silicon vendors have their own free tool distributions as well.
I think there's a wider trend here. It used to cost a fortune to tool up for things like FPGA and embedded software development, but there is now so much competition from the silicon vendors that the entry level prices for these tools are tending towards zero.
Where I agree is that a good developer chooses between ARM, 8051, FPGA, etc. depending on the requirements of the job on hand. They are all just tools in the toolbox. In that context, embedded Windows on x86 is the big expensive lump hammer that's used to bludgeon any design problem into submission.
Well, it worked for HTC
It's not long since HTC were just another contract manufacturer which exactly the same sales pitch (ie, outsource all your design and manufacture to us and you do the branding). A few years later and they've now got 6% smartphone market share under their own branding and are in the process of wiping their former customers out.
But just think how many Zynga style 'social' and 'app' games you could churn out with all that money! Gaudy Skinner boxes don't cost that much to develop.
If you choose your guard interval appropriately, any 'late' signal which falls outside the guard period will be sufficiently attenuated due to the extra propagation distance that you can just treat it as background noise. Under those circumstances, there is no theoretical limit to the area you can cover with an SFN.
It's not a case of single frequency networks enabling "white space". On the contrary, the more efficiently you use the TV broadcast spectrum the less white space there is - to the point where it doesn't make economic sense to even worry about it.
DVB-T and ISDB-T already fix the main problem...
The whole issue with adjacent TV transmitters wasting spectrum by having to transmit on different frequencies is already addressed by single frequency network support in the DVB-T and ISDB-T standards. In principle, all countries which use these standards can efficiently pack all of their terrestrial TV broadcasting into a well defined chunk of spectrum - eliminating the need for location based white space detection altogether. Of course the US decided to use ATSC instead, so that's not a solution that's open to them...
Thanks for that Mr Orlowski. I decided to check out the Mail website to see the original source, and now my eyes are hurting. It's just a vast wall of randomly selected content squeezed into little square boxes with no sane way of navigating your way around it. No wonder they rate the new Metro UI so highly.
It's CFO's on the scrapheap for me...
The whole idea behind outsourcing is that you should keep your 'core competencies' in-house and that anything which does not add differentiation can be outsourced. Isn't is strange that it's always the CFO that seems to make that call when the finance function is probably the least differentiated of all business activities?
I'm saying this because I happen to use an accountancy firm which provides the full range of back office services (payroll, invoice processing, etc.) and which will maintain your management accounts for you 'in the cloud', so that you have an up-to-date view of the company financials whenever and wherever you need them. They also have a number of very experienced partners with different specialisms who can be brought in to provide board level advice when required.
My company isn't big enough to make full use of all these services at the moment - but when you can buy in all of these capabilities from outside, I'm left wondering why any SME can ever justify having a full time CFO.
If this as big a problem as claimed, I've got a suggestion for an on-topic Reg Hardware review. How about taking a look at the Simtec Entropy Key? A true hardware random number generator for 36 quid must be worth investigating.
Disclaimer: I'm in no way connected with Simtec, not astroturfing, etc...
"This project has no files"
Yes, someone has registered an SF project for Darwin on ARM - but there's been no activity on it as far as I can see. From what I've read on other sites, a significant issue seems to be getting the correct toolchain together in order to compile it. Obviously, if you're working as an intern at Apple this probably isn't a problem.
If there is an active project to get the open source version of Darwin onto ARM, I'd love to know about it!
Just to clarify...
I am actually supporting Popovkin's position here. Counterfeit components are a problem for the _whole_ aerospace industry, not just in Russia. The shuttle accidents are obviously a different matter, but there are documented cases of counterfeit parts ending up in U.S. military kit - and they probably police their supply chain more rigorously than anyone else out there.
This is actually plausible...
Counterfeit electronic components are a real problem. If someone in the supply chain decided to make a quick buck by re-marking a bunch of consumer grade components as their radiation hardened equivalents, this is the kind of failure you could expect.
LED PWM colour mixing is patented...
That's right - using PWM to do LED colour mixing is patented in the US. A company called Color Kinetics beat Rambus to the punch there. Philips bought out Color Kinetics a while back, so they're now the alpha patent troll in this area.
You mean rich Indian and Chinese shareholders...
The people avoiding the GCT here are the original owners - so the main beneficiaries will be the Mumbai based Ruias brothers who control Essar and the Hong Kong based Li Ka-Shing who controls Hutchinson Whampoa. Your other points are perfectly valid, though...
Grid != Supercomputer
You missed the bit about this thing having a _shared memory_ of up to 16TB. That's 16TB of RAM which is directly addressable from any of the processors in the system. Grids are fine for embarassingly parallelisable problems, but there are certain applications which don't map well to grids and just need one big bastard of a computer. Obviously modelling the aerodynamics of helicopters must come into this category.
I agree that non-ARM SoC's also suffer from a huge amount of vendor fragmentation as well. I was just trying to suggest that by specifying a rigid set of ARM platform requirements, Microsoft may actually help to reduce the fragmentation issue for ARM and make it more viable to produce a 'standard' ARM Linux distro. Just because the retail Windows 8 ARM devices are going to be nailed down doesn't mean that other people couldn't use the same chippery to produce Linux friendly boxes. Clouds, silver linings and all that...
The real reason for the ARM restrictions
I suspect the real reason that Microsoft had to lay down the ARM implementation restrictions is just to get a common platform subset that they could realistically work with. There is a common misconception that all existing ARM SoC's are somehow interchangeable because they use the same processor core. Anyone who has tried to run up Linux on an ARM platform will tell you that this is definitely not the case - every vendor throws in a different bunch of proprietary peripherals. Add to that a random mix of open source and binary-blob drivers - most of which never make it anywhere near Linux mainline. In short, it's a god-awful mess. If Microsoft is able to define a common ARM platform that chip vendors have to conform to, I can't help thinking they will actually be doing ARM-Linux a favour here.
Par for the course...
This kind of thing seems to be standard operating procedure for a lot of tech component suppliers. I'm in RMA hell at the moment over a brand new AMD FX-8150 which was generating L2 cache parity errors all over the place. AMD were perfectly happy to take it back, but over six weeks later I'm still trying to get them to send me a replacement...
The bottom line is that if El Reg tried to help out all their readers as per the article, they'd never have time to do anything else!
Outlook is not just for Email.
Outlook is probably the most widely used contact management and calendar application out there, even for individuals and small businesses which can't justify running Exchange. For many people, being able to keep your phone and PC contacts and calendar in sync is a more useful feature than any amount of tweetbookery crap.
IMU's can tell you position...
An IMU contains 3 accelerometers which can be used to track the X,Y and Z acceleration. Given a known starting point and rest state you can then calculate the current velocity and position by just integrating the output of the accelerometers. Obviously this accumulates sensor errors, which is why a good IMU will be a high precision instrument which will need constant recalibration to deal with temperature changes, etc. GPS is a good recalibration reference, but a step change in GPS position relative to the IMU position should be a dead giveaway for GPS jamming.
Of course, a good IMU will be an expensive piece of kit - and there's a good chance it wouldn't be cost effective for a low value asset which is expected to be regularly shot down and captured. The same goes for why it probably doesn't make sense to use the military GPS keys in there. At the end of the day, if you expect your drones to be captured and reverse engineered, you don't want anything more sophisticated than an RC plane with a bog standard GPS and a webcam in there.
Off their rockers...
I've bought phones directly from Nokia in the past, so I'm on their sales mailing list. The point at which I decided I'm not going to be buying another of their phones was when they started sending me promo Emails full of patronizing juvenile crap about 'Seeking Irregular People' for their 'Amazing Collective' and how I'd be 'sillier than a fake moustache' not to join in with their new social media hipsterfest. At least I now know that alienating boring business-bods like me is part of their marketing strategy, so I don't feel so bad about it...
I'd previously discounted using JavaFX on the basis of it's closed source nature and the 'like java but not really java' development model. I hate to admit it, but it now sounds like Oracle is doing a better job with JavaFX than Sun were - could be time to download the SDK and give it a second chance.
I't not Stalinist if you have a choice...
As long as the consumer is making an 'informed' choice to trade detailed information about their home lifestyle for a shiny piece of techno-bling, I'd say that's capitalist.
If the government makes it compulsory for every house to be fitted with a 'Non-Invasive Appliance Load Monitoring' (NIALM) system in order to obtain the same information, I'd say that's Stalinist.
I _have_ got all Salman Rushdie's records!
A little known fact about Mr Rushdie is that back when he was an advertising drone he wrote and was credited with the lyrics of a promotional record for the Burnley Building Society. You can find it on YouTube, but be quick before he pulls another hissy fit and issues a takedown request...
A fair point well made...
It's instructive that Luxembourg were able to upgrade their 2000kW transmitter as recently as 1994 - using what sounds like a fairly standard piece of Thompson-CSF kit - while at the same time the BBC insists that building long wave transmitters is a lost art and there is no way the UK can ever afford to build a new one! I noticed that there is a petition about this on the e-Gov website. Only 10 signatories so far, though...
I've heard this one before...
Surely, an alternative solution would be to give the Beeb a few million quid to design and install a set of shiny new longwave transmitters to replace the existing valve based kit.
Then the choice comes down to :
(a) Spending a relatively small amount of money on maintaining an existing, perfectly adequate system or...
(b) Ripping it all up and starting from scratch at a huge additional cost to the taxpayer.
You don't need to be Lord Truscott to know which one the government's 'industry consultants' will be recommending.
Actually they're not sensible at all - just blatantly obvious stuff that those 'skilled in the art' have been using for years. They've managed to get it through the patent system by dressing it up in long words and adding a bunch of straw-man prior art examples to fool the patent examiners. Take their following 'prior art' explanation of how integer to floating point conversion is normally done...
"Thus, before an integer unit can operate on data that is in a floating point unit, the data must first be converted into integer format, and then stored to memory. Alternatively, before a floating point unit can operate on data that is in an integer unit, the data must first be converted into floating point format, and then stored to memory. In modern microprocessors, requiring data to be converted, stored to memory, and then retrieved from memory, is very time consuming for the microprocessor, and adds significant delay in processing the data."
Maybe someone did code a floating point library like that. Once. Before being sacked for not having a f'ing clue. The bottom line is that this is no better than the crap that Apple are throwing around.
It's all about the energy density.
Just look up 'Energy Density' on Wikipedia. They have a very informative graph comparing the energy density of different energy sources. It gives you an idea of just how far battery storage has to go before it gets anywhere close to hydrocarbons.
The bottom line is that hydrocarbons are such a fantastic way of storing energy that the only way of realistically improving on them is to go for the Hydrogen option.
Beer - because there are certain hydrocarbon compounds that are wasted by putting them in your car.
What's the uplink for?
Top of my list of questions is why British Gas has suddenly decided it needs a realtime feed of everyone's electricity usage. Let's knock down a few of the more commonly used justifications:
(1) It allows them to do 'real time' pricing. Well, radio teleswitch has been doing this for years for the economy 7 tariffs - and no need for an uplink there. Broadcasting at UHF/VHF has already been demonstrated to be the best way of implementing this feature, even if the existing protocols need replacing for the new generation of meters. And guess what - there's shortly going to be a whole bunch of empty spectrum ideally suited to delivering this kind of broadcast service!
(2) It allows the electricity generators to dynamically adjust supply according to demand. Well, WTF do you think the National Grid was doing for the best part of the 20th Century? Monitoring individual consumer usage would just generate vastly more data than can be sensibly used in this context - other than by building a British Gas version of Skynet.
(3) It eliminates the need to send a meter reader round in a van. Like estimated readings already do? Like the ability to enter your meter reading on the website already does? The fact is, flesh-based meter readers will always be required because they are the only reliable way of detecting physical tampering.
The bottom line is that I can't think of any justifiable reason why I'd want British Gas to monitor my domestic energy use in real time - so I just can't see why they need that uplink.
At least this is a voluntary cost...
If you think the AlertMe setup is too expensive, just try digging into the numbers behind the national smart meter rollout. The initial estimates were giving a payback period of about 15 years. That clearly looked a bit crap, so OFGEM has managed to fix it by doubling the estimated savings that we're all going to make (YAY!). That's in raw cash terms - completely ignoring the opportunity costs involved.
Here's a request for the Reg energy desk - please take some time out from moaning about windfarms and do some digging into the real costs of the proposed smart meter rollout. After all, at 11.3 billion quid it will be on a par with that other government-backed IT boondoggle, the NHS IT system.
A nuke - because electricity was going to be too cheap to meter thanks to nuclear power. Remember that promise?
Want but don't need...
After trying all the usual places, it looks like anyone who was selling at the firesale prices is now out of stock. So my extensive collection of 'dead' technology will have to do without a TouchPad for now.
Maybe someone in HP marketing needs to learn about price elasticity of demand. At 400 quid nobody wants one. At 100 quid they can't shift them fast enough. Surely there must have been a price point between the two where they could have turned it into a viable business...
We'll all end up paying for it (again).
In the current climate, I can't see any sane IPO investors wanting to buy into the likes of Groupon and Zynga - so the ones who stand to get burned most are probably the underwriters. Do a quick search for 'groupon underwriters' and 'zynga underwriters' for a list of the banks we'll all have to be bailing out next time.
With comprehension skills like that...
...I'm amazed that Terry13 was able to find out which bus to take to get to Cambridge in the first place. Lyndsay Williams is the name on the patent in question - which suggests that they are the ideal person to give some historical background on the controversial patent. The fact that the talk happens to be in Oxford is just a statement of geographical fact.