There are authorities that are relevant for automotive embedded software safety. Groups such as the UK's MIRA have a working group, MISRA (Motor Industry Software Reliability Association) which define a safe subset of C that is workable for embedded systems.
Specifically, undefined, unspecified and implementation-dependent behaviour of the C standard is identified, and the unsafe aspects are prohibited as part of the MISRA-C subset.
Alongside this, embedded system vendors in the automotive industry have serious validation requirements both from clients, and from internal compliance processes - the SEI's CMMI and ISO 26262 are a couple of standards that come to mind.
Keep in mind, this may be a little out of date - it's been a few years since I worked in automotive, but the people working in the industry do take it very, very seriously.
I think we can safely assume that Blackberry is about as secure as a wet paper bag in countries where the device has become "commercially successful" and the government is less than interested in maintaining privacy.
Mentioining "national security" at the end of the video is a clear sign that RIM has well and truly given in on their claims of absolute security for the sake of maintaining a moderately-successful business.
Never trust the security of communications where the keys are being handled by someone outside your organisation.
Why not just use Boot Camp and dual-boot Windows on the Macbook? A machine of that vintage should be able to run Windows 7 well, and it'll save a thousand or two.
I'd be fine with this, as long as the beancounters are forced to personally visit every single trucker in person, and attempt to extract their fees.
I'd imagine they'd soon have a 'close encounter of the truckstop kind'... perfect sort of punishment for this level of arrogance. Next they'll be demanding fees for listening to the radio while driving to work. The publishing industry will stop at nothing to fraudulently demand fees for others' works.
How is this relevant to 'news for nerds'? A startup with overenthusiastic PR says they're "aiming at mobile"... but no concrete plans or implementation?
Call me back when it's something worth giving a shit about.
The restriction on civilian GPS receivers is that the receiver should be able to show a position higher than 11mi, or a velocity greater than 515m/s, but not both simultaneously. A number of GPS manufacturers, however, have implemented this in a somewhat slack fashion and used || instead of &&.
Where's the blame targeted at Verizon? Last I checked, they were the ones pushing the profits-over-neutrality angle.
If you don't like their business practices, don't do business with them. Simple as that. Businesses think in terms of dollars; it's the only way to send a clear message.
The flaw that most articles on biometric identification, be they fingerprints, retinal scans or other, is that you only have a limited number of immutable keys to choose from. While it may not be an issue in a school setting, if anyone is able to reconstruct the fingerprint or retina picture from the stored data, or at least a fake fingerprint/picture that is functionally equivalent to the real one, it's game over. You only have two eyeballs, and ten fingerprints.
I'd rather a system that allows me to change the key once in a while.
I'd strongly suggest that you do at least a basic level of looking into theory while you're creating "practical circuits" - it's quite helpful when you're debugging to know at least roughly what's meant to happen.
One source I can recommend is the MIT Open Courseware resources - the 6.002 course on Circuits and Electronics is a good place to start; I'm an embedded software engineer who's started to push into the hardware side of things, and that set of lectures helped me turn my vague understanding of electronics (being able to read a circuit and understand what's going on) into something practical (being able to design a circuit).
Sounds great to me. I'd extend it to blood donors, with quantity donated moving you further up.
I'd support this, provided we eliminated the discrimination present in various countries' blood donation rules. It's still legal for groups such as the Red Cross to discriminate against homosexuals, reinforcing the view of risk-taking sexual behaviour being prevalent only in the homosexual community, and not the populus at large.
Development of a single ECU (from inception through to production) takes in the order of USD$500k-20m.. I doubt you'll see manufacturers opening up their software investments to the public eye any time soon. Most MCUs have flags set after programming to disable readout of the flash over programming channels.
MISRA is one standard that is used, but it's not exactly an all-of-domain standard - it simply covers common fault modes in automotive/embedded software.
The EU is pushing to have ISO26262 ready in the next 6-12 months, which will set standards for functional safety in the system domain - not simply limited to software, and with requirements set by the criticality of the given part.
For example, devices that are deemed to be safety-critical (powertrain, lighting) are held to higher standards than those that are not (immobilisers, HVAC).
We're about to see this sort of thing in the next year or two; Europe has been pushing ASIL/Functional Safety (ISO 26262) fairly heavily as of late, and it's been in the works for a few years.
Formal verification and other methods are required for those devices in the highest ASIL category - usually those devices related to the powertrain, and essential lighting etc.
It's about time the automotive industry had a wake-up call to how critical some automotive embedded software can be..
The battery life sucks because the encoding and decoding is a hell of a lot more compute-heavy than GSM. The turbo codes that are used provide a significant leap in throughput and interference immunity, but as a tradeoff, require a lot more power to decode. Also, some of the earlier amplifiers used for W-CDMA signals were quite power hungry to provide acceptable performance.
Actually, UMTS uses a lower transmit power - 250mW peak instead of 2W for GSM. That said, the interference will not be audible in the same manner due to the modulation, even if UMTS used 2W instead of 250mW.
The Government has to step up and declare Paypal a bank. Banks can't take or freeze your money simply because they don't like what you do (which Paypal often does) - only the government can do that if you are in breach of laws.
Some countries did. Australia, for example. It's made it a lot harder for Paypal to act evilly over here at least; it's a shame the rest of the company isn't forced to behave in a similar fashion.
Make sure copyright in any committed patches is assigned to you, or require public domain, and take the dual license route. If they're adverse to using the BSD license, charge them for the privilege and get a lawyer to write up a software license.
Bit of money for you, a (hopefully) reliable license for all parties, and the organisation gets the code under a non-OSS license. Everyone wins.
There are numerous handsets that'll do at least UMTS850, for example those being retailed on certain telcos in Australia. They may not be as widespread as UMTS2100/900/1800 handsets, but they're certainly out there, and being made by major manufacturers (Nokia, HTC, Samsung).
Even though the spotlight's being shone on the dodgy practices behind ACTA, what's the bet that this will still result in just plain old stonewalling until the final agreements have been ratified by treaty, and all is needed is legislation in each member state?
ACTA comes from utterly fraudulent governance, and not from the public's mandate.
Non-volatile? Like all the other "non-volatile RAM, instant-on" technologies that have gone before? MRAM, SRAM, Holographic storage... and now phase-change memory.
I've heard this marketing bullshit before. Call me when it's not vapourware.
There's also predominantly 3G networks in Australia - one of the national mobile carriers has bigger coverage on UMTS/HSPA than on GSM. For a rural example, the 300km stretch from Mildura to Broken Hill has absolutely no GSM or 3G coverage after leaving Mildura, but UMTS works for 2/3 of the way.
Cell density is required to be high in densely-populated areas with the current public appetite for data, but it doesn't mean that UMTS won't service large cells. People simply don't put the same demands on GSM cells because data throughput is awfully slow.
Slashdot Mirror
Why not skip the decade-long cycle of getting out the word, and simply introduce mandatory attendance at elections?
Many other countries do it, and it makes for significantly fairer elections with fewer batshit insane^W^Wextremist candidates.
There are authorities that are relevant for automotive embedded software safety. Groups such as the UK's MIRA have a working group, MISRA (Motor Industry Software Reliability Association) which define a safe subset of C that is workable for embedded systems.
Specifically, undefined, unspecified and implementation-dependent behaviour of the C standard is identified, and the unsafe aspects are prohibited as part of the MISRA-C subset.
Alongside this, embedded system vendors in the automotive industry have serious validation requirements both from clients, and from internal compliance processes - the SEI's CMMI and ISO 26262 are a couple of standards that come to mind.
Keep in mind, this may be a little out of date - it's been a few years since I worked in automotive, but the people working in the industry do take it very, very seriously.
Simply buy the affected customers an XBox 360. Their online services are still working..
I think we can safely assume that Blackberry is about as secure as a wet paper bag in countries where the device has become "commercially successful" and the government is less than interested in maintaining privacy.
Mentioining "national security" at the end of the video is a clear sign that RIM has well and truly given in on their claims of absolute security for the sake of maintaining a moderately-successful business.
Never trust the security of communications where the keys are being handled by someone outside your organisation.
Why not just use Boot Camp and dual-boot Windows on the Macbook? A machine of that vintage should be able to run Windows 7 well, and it'll save a thousand or two.
I'd be fine with this, as long as the beancounters are forced to personally visit every single trucker in person, and attempt to extract their fees.
I'd imagine they'd soon have a 'close encounter of the truckstop kind'... perfect sort of punishment for this level of arrogance. Next they'll be demanding fees for listening to the radio while driving to work. The publishing industry will stop at nothing to fraudulently demand fees for others' works.
How is this relevant to 'news for nerds'? A startup with overenthusiastic PR says they're "aiming at mobile"... but no concrete plans or implementation?
Call me back when it's something worth giving a shit about.
The restriction on civilian GPS receivers is that the receiver should be able to show a position higher than 11mi, or a velocity greater than 515m/s, but not both simultaneously. A number of GPS manufacturers, however, have implemented this in a somewhat slack fashion and used || instead of &&.
Where's the blame targeted at Verizon? Last I checked, they were the ones pushing the profits-over-neutrality angle.
If you don't like their business practices, don't do business with them. Simple as that. Businesses think in terms of dollars; it's the only way to send a clear message.
The flaw that most articles on biometric identification, be they fingerprints, retinal scans or other, is that you only have a limited number of immutable keys to choose from. While it may not be an issue in a school setting, if anyone is able to reconstruct the fingerprint or retina picture from the stored data, or at least a fake fingerprint/picture that is functionally equivalent to the real one, it's game over. You only have two eyeballs, and ten fingerprints.
I'd rather a system that allows me to change the key once in a while.
In a few years time, it may even be voted Psi-Scan of the Year, best budget model, 3 years running!
I'd strongly suggest that you do at least a basic level of looking into theory while you're creating "practical circuits" - it's quite helpful when you're debugging to know at least roughly what's meant to happen.
One source I can recommend is the MIT Open Courseware resources - the 6.002 course on Circuits and Electronics is a good place to start; I'm an embedded software engineer who's started to push into the hardware side of things, and that set of lectures helped me turn my vague understanding of electronics (being able to read a circuit and understand what's going on) into something practical (being able to design a circuit).
I'd support this, provided we eliminated the discrimination present in various countries' blood donation rules. It's still legal for groups such as the Red Cross to discriminate against homosexuals, reinforcing the view of risk-taking sexual behaviour being prevalent only in the homosexual community, and not the populus at large.
Development of a single ECU (from inception through to production) takes in the order of USD$500k-20m.. I doubt you'll see manufacturers opening up their software investments to the public eye any time soon. Most MCUs have flags set after programming to disable readout of the flash over programming channels.
MISRA is one standard that is used, but it's not exactly an all-of-domain standard - it simply covers common fault modes in automotive/embedded software.
The EU is pushing to have ISO26262 ready in the next 6-12 months, which will set standards for functional safety in the system domain - not simply limited to software, and with requirements set by the criticality of the given part.
For example, devices that are deemed to be safety-critical (powertrain, lighting) are held to higher standards than those that are not (immobilisers, HVAC).
We're about to see this sort of thing in the next year or two; Europe has been pushing ASIL/Functional Safety (ISO 26262) fairly heavily as of late, and it's been in the works for a few years.
Formal verification and other methods are required for those devices in the highest ASIL category - usually those devices related to the powertrain, and essential lighting etc.
It's about time the automotive industry had a wake-up call to how critical some automotive embedded software can be..
The battery life sucks because the encoding and decoding is a hell of a lot more compute-heavy than GSM. The turbo codes that are used provide a significant leap in throughput and interference immunity, but as a tradeoff, require a lot more power to decode. Also, some of the earlier amplifiers used for W-CDMA signals were quite power hungry to provide acceptable performance.
Actually, UMTS uses a lower transmit power - 250mW peak instead of 2W for GSM. That said, the interference will not be audible in the same manner due to the modulation, even if UMTS used 2W instead of 250mW.
Some countries did. Australia, for example. It's made it a lot harder for Paypal to act evilly over here at least; it's a shame the rest of the company isn't forced to behave in a similar fashion.
Very simple solution: dual-licensing.
Make sure copyright in any committed patches is assigned to you, or require public domain, and take the dual license route. If they're adverse to using the BSD license, charge them for the privilege and get a lawyer to write up a software license.
Bit of money for you, a (hopefully) reliable license for all parties, and the organisation gets the code under a non-OSS license. Everyone wins.
There are numerous handsets that'll do at least UMTS850, for example those being retailed on certain telcos in Australia. They may not be as widespread as UMTS2100/900/1800 handsets, but they're certainly out there, and being made by major manufacturers (Nokia, HTC, Samsung).
Like any GSM/UMTS network in the world?
Insert your SIM, and you're on. Only phones that won't work is those that have their IMEI reported as stolen.
Even though the spotlight's being shone on the dodgy practices behind ACTA, what's the bet that this will still result in just plain old stonewalling until the final agreements have been ratified by treaty, and all is needed is legislation in each member state?
ACTA comes from utterly fraudulent governance, and not from the public's mandate.
Non-volatile? Like all the other "non-volatile RAM, instant-on" technologies that have gone before? MRAM, SRAM, Holographic storage... and now phase-change memory.
I've heard this marketing bullshit before. Call me when it's not vapourware.
There's also predominantly 3G networks in Australia - one of the national mobile carriers has bigger coverage on UMTS/HSPA than on GSM. For a rural example, the 300km stretch from Mildura to Broken Hill has absolutely no GSM or 3G coverage after leaving Mildura, but UMTS works for 2/3 of the way.
Cell density is required to be high in densely-populated areas with the current public appetite for data, but it doesn't mean that UMTS won't service large cells. People simply don't put the same demands on GSM cells because data throughput is awfully slow.