The fragility of car computers revealed.
in [UK Drivers]
|
From: Mrcheerful on 25 May 2010 06:49 Brimstone wrote: > "Doug" <jagmad(a)riseup.net> wrote in message > news:0fac0491-913a-439e-90e3-4726e85d066c(a)y12g2000vbr.googlegroups.com... > >> A wake up call eh? Not overly concerned but a little concerned maybe? >> I know I would be very concerned if my life depended on a computer >> working 100% properly throughout its useful lifetime. > > What makes you think that the computer in a car is the same as the > one on your desk Doug? > >> Thes researchers have proved conclusively that the car computer can >> cause the engine and brakes to malfunction, thus putting road users >> at risk. > > Only when interfered with by an external computer. > >> Whether the computer will actually malfunction in this way >> remains a moot point but should be a matter of serious concern to >> anyone who cares about road safety, which apparently excludes most of >> the motorists who post here. >> > It also excludes you Doug, because if you had any understanding you > would realise that there is no point in jumping up and down and > getting excited. > The only result from this research for the future is that car > designers need to make sure that new designs cannot be interfered > with by those with evil intent. I worry far more about the nut behind the wheel (or the handlebars)
From: Man at B&Q on 25 May 2010 07:47 On May 25, 11:38 am, boltar2...(a)boltar.world wrote: > On 25 May 2010 10:08:51 GMT > > Adrian <toomany2...(a)gmail.com> wrote: > >> Oh, indeed. However, the ability of any system to perform correctly and > >> react appropriately in all circumstances does depend vitally on the > >> designers' ability to foresee all possible eventualities. > > >Don't forget that there's a huge difference in complexity between (say) a > >PC app and an embedded controller - both in terms of the range of > >functionality and the scope of the environment it has to deal with. > > Depends on what controller. The embedded controllers in a train or an > aircraft will be running software a lot more complicated than your average > PC app. The control software on the Prius is probably pretty involved as > well I should imagine. All of those examples are closed systems that are designed and tested to very rigorous standards by people who know what they are doing. look up things like "functional safety" "IEC61508" and "Safety integrity Level" or "SIL", but don't expect to understand it straight away if you are still at the stage of making statement such as the above. PCs are oppen system and is simply not possible with current OS and software models to prove any real degree of reliability. MBQ
From: boltar2003 on 25 May 2010 08:07 On Tue, 25 May 2010 04:47:14 -0700 (PDT) "Man at B&Q" <manatbandq(a)hotmail.com> wrote: >> PC app. The control software on the Prius is probably pretty involved as >> well I should imagine. > >All of those examples are closed systems that are designed and tested >to very rigorous standards by people who know what they are doing. >look up things like "functional safety" "IEC61508" and "Safety >integrity Level" or "SIL", but don't expect to understand it straight >away if you are still at the stage of making statement such as the >above. Yes, thanks for the heads up, but I work in software development and I've written realtime systems so don't try and tell me my job. Unless some software has been formally proved which is VERY time consuming and expensive to do and VERY rare then there will almost always be some bugs lurking somewhere that only come to light in exceptional and unexpected combination or sequence of events. >PCs are oppen system and is simply not possible with current OS and >software models to prove any real degree of reliability. Rubbish. There are many PCs running Linux and other versions of Unix which have been running 24/7 for literally years as back end servers. If you're alluding to that pile of poorly designed poorly written shite called Windows then yes , all bets are off. B2003
From: Man at B&Q on 25 May 2010 09:20 On May 25, 1:07 pm, boltar2...(a)boltar.world wrote: > On Tue, 25 May 2010 04:47:14 -0700 (PDT) > "Man at B&Q" <manatba...(a)hotmail.com> wrote: > > >> PC app. The control software on the Prius is probably pretty involved as > >> well I should imagine. > > >All of those examples are closed systems that are designed and tested > >to very rigorous standards by people who know what they are doing. > >look up things like "functional safety" "IEC61508" and "Safety > >integrity Level" or "SIL", but don't expect to understand it straight > >away if you are still at the stage of making statement such as the > >above. > > Yes, thanks for the heads up, but I work in software development and Then you only have one viewpoint. Designing safety critical systems is as much about the hardware and *system* design than it is about software. Again. don't expect to understand if you can't grasp that point. I've > written realtime systems so don't try and tell me my job. Unless some > software has been formally proved which is VERY time consuming and expensive to > do and VERY rare then there will almost always be some bugs lurking somewhere > that only come to light in exceptional and unexpected combination or sequence > of events. The *system* is designed to fail safe in that eventuality. > >PCs are oppen system and is simply not possible with current OS and > >software models to prove any real degree of reliability. > > Rubbish. There are many PCs running Linux and other versions of Unix which > have been running 24/7 for literally years as back end servers. If you're > alluding to that pile of poorly designed poorly written shite called Windows > then yes , all bets are off. Again, closed systems, probably in a secure environment, dedicated to one specific task, much like the automotive systems we are talking about in that respect. Pretty much proves my point that a general purpopse PC (whatever OS) is open to having all kinds of 3rd party hardware and software added to it. MBQ
From: boltar2003 on 25 May 2010 09:50
On Tue, 25 May 2010 06:20:47 -0700 (PDT) "Man at B&Q" <manatbandq(a)hotmail.com> wrote: >> Yes, thanks for the heads up, but I work in software development and > >Then you only have one viewpoint. Designing safety critical systems is >as much about the hardware and *system* design than it is about >software. Again. don't expect to understand if you can't grasp that >point. The hardware will generally be stock microcontrollers probably running an ARM core which is a proven design. The peripheral circuitry will obviously be bespoke for a given task but the software will be a lot more complicated than any of the hardware. >> written realtime systems so don't try and tell me my job. Unless some >> software has been formally proved which is VERY time consuming and expens= >ive to >> do and VERY rare then there will almost always be some bugs lurking somew= >here >> that only come to light in exceptional and unexpected combination or sequ= >ence >> of events. > >The *system* is designed to fail safe in that eventuality. Tell that to the people who died in the air france crash. Whatever fail safe there was for when the pitot tubes fucked up clearly didn't work or didn't even exist. Fail safes are only as good as the people who designed them are smart. B2003 |