and btw, all the drives I see in commercial storage are notebook style (2.5") sas drives. IIRC while they are 2.5 inch they are considerablly taller than modern notebook drives.
IIRC 3.5 inch platters don't like being spun at 15k rpm so you either end up with 2.5 inch platters in a case big enough for 3.5 inch ones or you use a case that actually fits the platters. So for the applications that need incrediblly high speed 2.5 inch makes sense. OTOH if capacity matters more to you than speed then 3.5 inch 7200 or 10K rpm drives make more sense.
The TIOBE Programming Community index gives an indication of the popularity of programming languages. The index is updated once a month. The ratings are based on the number of skilled engineers world-wide, courses and third party vendors. The popular search engines Google, MSN, Yahoo!, Wikipedia and YouTube are used to calculate the ratings What i'd like to know is how the hell they search for things like C, C# and java (all of which have significant other meanings) without getting a ton of false positives.
Still if you assume thier results are valid and add together the two languages the GGP called toy languages you get about 13%. Add in java which is pretty similar to C# and you are at about 32.5% (all percentages rounded to the nearest 0.5% before adding for ease of mental arithmetic).
Amateur radio enthusiasts come to mind. Something like, oh, this I had a quick look at some of the entries in their list and all the ones I looked at were piggy backing on government missions.
Do you have any evidence of "rocket enthusiasts" reaching orbit without governemnt assistance?
BTW you can get all the iplayer content through the "all a-z" menus of catch up TV on demand. I generally find this to be a much more reliable route than the iplayer specific menus.
A centrifuge should work for creating an "eath gravity" environment on the moon or mars. You would just have to angle your living pods such that it's floor was in line with the resultant of the centrifugal force and the moons gravity. Most likely the best way to do this would be to design the pods to self align with the force on them by putting them on a pivot above thier center of mass (the big advantage of this method is that you can slow down and stop without changing which direction is down inside the pods).
You may also need to make the centrifuge quite big to reduce the force gradiants.
Also bear in mind that what counts as a ligitimate reason in one country (or even in the case of the USA one state) may not be considered legitimate other places.
Also at least here in the UK how long you have worked somewhere makes a big difference to how much protection you have.
Afaict the trickiest bit with copying nes carts is actually identifying them. nes carts (unlike gameboy carts) use a huge range of mapper chips and don't have any header information to say which mapper is in use. So if you are trying to dump a cart that hasn't been dumped before you may well have to open it up to find out what mapper is in there.
If you want to actually copy the cart rather than just make an image for use on emulators you will also need to either clone the mapper chip or find a donor cart with the same mapper chip to host your copy.
There is also the security chip issue but afaict clones of that are readilly availible nowadays.
Suppose you could design a restraining device that protected you from colliding with the windshield as effectively as seat belts currently do but which posed less threat of shoulder injury. It's called a five point racing harness.
The problem is practicality rears it's ugly head, there is only so much inconvinaince and cost people will put up with in the name of protection from a relatively rare occourance.
The only reason that astronomers use observation rather than direct experiment is because they have no choice. Afaict these guys aren't either. The research ethics guys probablly aren't going to let you deliberately kill or brain damage test subjects! So a combination of observation and modeling (both computer modeling and modeling by using say crash test dummies outfitted with sensors) is probablly thier only option.
so let's put down those who have figured out how to make money! The BBC is funded by what essentially ammounts to a tax on TV usage. Even if I only want to watch ITV and SKY I still have to pay the BBC.
And the availibility of free news from the BBC has increased considerablly in recent years. A decade ago most people didn't have internet connections and only pay TV subscribers were likely to have BBC news 24 (I'm not sure if it was encrypted or not on digital terrestrial at that time but even if it wasn't digital terrestrial boxes were expensive back then) so the only news most people would have got from the BBC was a few programmes a day and a small ammount on ceefax.
Personally I think any internet news source that put itself behind a paywall would be committing suicide regadless of whether or not the BBC existed but lets not pretend that the BBC needs to make a profit.
IIRC they were pushed into drm on iplayer by some agency called "BBC Trust" which afaict is an organisation responsible for regulating the BBC.
I've always thought it rather crazy given that all the material has been previously broadcast as unencrypted DVB but I don't think it was actually the BBCs fault.
Umm, much of what Wal Mart sells is *exactly* the same brands that sell elsewhere, sometimes for less money. I've heared (I dunno how true it is) that wal-mart often bullies manufacturers into producing slightly different models that are cheaper but also lower quality for them than the stuff that other retailers get.
In many cases, store generic brands are made by the same companies that make the 'name brand' items, though I know of no way of knowing if "Safeway ketchup" is really the same as Heinz. It's not entirely about who makes an item though, it's also about what standards the customer (whether it's a big brand or a retailer) is willing to pay for.
Is saving a minuite or so per support call and/or trimming down the cost per call a little by allowing yourself (or your contractor) to hire idiots really worth pissing off your already pissed off customer even more?
Indeed, iff the pad is both perfectly random and perfectly protected from exposure then one time pads are perfectly unbreakable.
Both of these issues are problems in reality though,
Firstly you need real randomness for your pad, PRNG output DOES NOT cut it. If you use a PRNG then you effectively have a stream cipher with a key consisting of knowlage of the prng and the seeding information fed to the prng.
Distribution is also a big problem since you need as much key as data you can't use one time pad encryption to distribute your new pads, so you need physically secure communications to transfer the pad.
Define "one". A processor cannot do two operations at once. Everything is put into a line, and that one line is dealt with really really fast. I think your understanding of computer architecture is a bit behind the times.
It used to be that computers fetched an intrustruction, decoded it, executed it and moved on to the next one.
Then came pipelining, this basically meant that they could start on one instruction before finishing the previous provided there weren't any data dependency issues blocking it.
Then came things like out of order execution and virtual registers to try and keep the pipeline full more of the time.
When that still wasn't enough to keep the hugely long pipeline of the P4 full they introduced hyperthreading so work could be taken from more than one thread at a time, a P4 really can be working on two totally seperate tasks at the same time.
With core 2 they dropped hyperthreading for a while but put two complete cores on the die so again the chip could be working on two totally seperate tasks at the same time (or four if you went for the version with two dies in the package).
With core i7 they brought back hyperthreading and kept two cores on the die. So it could be working on four seperate tasks at once (or eight if you went for the version with two dies in the package).
AMD has also been going in for multicore though I don't think they have used hyperthreading.
The impression I have got is that CPU clocks aren't likely to increase increase at anything like the rate they did in the past so our gains are going to come from chips that do more and more at the same time.
Of course what happens in a crash depends on more than just the relative speed.
If say a train (more realisitic than your moving wall) and a car are both moving at 60MPH towards each other for a combined closing speed then when they crash the car will absorb a huge ammount of energy very suddenly and be almost completely crushed. The train will likely keep going with minimal damage.
If two cars are both moving at 60MPH towards each other then damage should (theoretically) be equally split so the affect should be the same (same ammount of engergy dissipated in each car) as a single car moving at 60mph and hitting an immovable object.
So the GP is kind of right if we model a wall as an immovable object (which depending on the wall in quesion may or may not be a good model) and the car hits it head on then it probablly is as bad as a head on collision.
A sanely designed wall/barrier system will be designed so it is far more likely that a care makes a glancing blow into a barrier/wall and get deflected (so most of the energy in the car stays as kenetic energy in roughly it's original direction).
Plus there are a wide variety of vehicles on the roads, when say a truck hits a car you have a situation similar (but less extreme) than the one I mentioned above with the train.
Please watch this video [bbc.co.uk] and reconsider your habit of texting while driving. "A shocking video has been made for pupils in an attempt to stop car crashes caused by the driver texting while driving.
The short film, starring young actors from south Wales, shows a teenage girl killing four people after she uses her mobile phone to send a text."
So lemme get this straight you as asking us to watch a governement propoganda film?
Note that this is all just claims based on tests under condtions that are supposed to accelerate ageing by some huge factor.
The problem with that of course is that such tests may accelerate some ageing mechanisms and not others and therefore not accurately predict what will happen decades down the line.
Iomega zipdrives, several sorts of tape drives, half a dozen different memory card standards... none of those were seen as fringe technologies at the time. Theese may not have been fringe but they were never anywhere near as widely distributed as the 3.5 inch floppy was.
Afaict many lithium ion based devices don't have a bypass path (adding one complicates the power circuitry quite a bit, especially when you are dealing with the low voltage of a single cell lithium ion pack). So running on charge has to happen via the charger.
The trouble with this is that for safety reasons a typical lithium ion charger will only start a charge if it sees a good battery pack.
Afaict this is the reason that many mobile phones won't work with the battery pack removed even if you plug in a charger.
What i've generally found is that generic motherboards (either purchased seperately or as part of a PC build from a small system builder) are far more likely to have the legacy ports than those made for use in big brand machines.
Slashdot Mirror
and btw, all the drives I see in commercial storage are notebook style (2.5") sas drives.
IIRC while they are 2.5 inch they are considerablly taller than modern notebook drives.
IIRC 3.5 inch platters don't like being spun at 15k rpm so you either end up with 2.5 inch platters in a case big enough for 3.5 inch ones or you use a case that actually fits the platters. So for the applications that need incrediblly high speed 2.5 inch makes sense. OTOH if capacity matters more to you than speed then 3.5 inch 7200 or 10K rpm drives make more sense.
The TIOBE Programming Community index gives an indication of the popularity of programming languages. The index is updated once a month. The ratings are based on the number of skilled engineers world-wide, courses and third party vendors. The popular search engines Google, MSN, Yahoo!, Wikipedia and YouTube are used to calculate the ratings
What i'd like to know is how the hell they search for things like C, C# and java (all of which have significant other meanings) without getting a ton of false positives.
Still if you assume thier results are valid and add together the two languages the GGP called toy languages you get about 13%. Add in java which is pretty similar to C# and you are at about 32.5% (all percentages rounded to the nearest 0.5% before adding for ease of mental arithmetic).
Amateur radio enthusiasts come to mind. Something like, oh, this
I had a quick look at some of the entries in their list and all the ones I looked at were piggy backing on government missions.
Do you have any evidence of "rocket enthusiasts" reaching orbit without governemnt assistance?
BTW you can get all the iplayer content through the "all a-z" menus of catch up TV on demand. I generally find this to be a much more reliable route than the iplayer specific menus.
A centrifuge should work for creating an "eath gravity" environment on the moon or mars. You would just have to angle your living pods such that it's floor was in line with the resultant of the centrifugal force and the moons gravity. Most likely the best way to do this would be to design the pods to self align with the force on them by putting them on a pivot above thier center of mass (the big advantage of this method is that you can slow down and stop without changing which direction is down inside the pods).
You may also need to make the centrifuge quite big to reduce the force gradiants.
Also bear in mind that what counts as a ligitimate reason in one country (or even in the case of the USA one state) may not be considered legitimate other places.
Also at least here in the UK how long you have worked somewhere makes a big difference to how much protection you have.
Any ideas on how to build an NES copier?
http://www.tripoint.org/kevtris/Projects/copynes/
Afaict the trickiest bit with copying nes carts is actually identifying them. nes carts (unlike gameboy carts) use a huge range of mapper chips and don't have any header information to say which mapper is in use. So if you are trying to dump a cart that hasn't been dumped before you may well have to open it up to find out what mapper is in there.
If you want to actually copy the cart rather than just make an image for use on emulators you will also need to either clone the mapper chip or find a donor cart with the same mapper chip to host your copy.
There is also the security chip issue but afaict clones of that are readilly availible nowadays.
Suppose you could design a restraining device that protected you from colliding with the windshield as effectively as seat belts currently do but which posed less threat of shoulder injury.
It's called a five point racing harness.
The problem is practicality rears it's ugly head, there is only so much inconvinaince and cost people will put up with in the name of protection from a relatively rare occourance.
The only reason that astronomers use observation rather than direct experiment is because they have no choice.
Afaict these guys aren't either. The research ethics guys probablly aren't going to let you deliberately kill or brain damage test subjects! So a combination of observation and modeling (both computer modeling and modeling by using say crash test dummies outfitted with sensors) is probablly thier only option.
so let's put down those who have figured out how to make money!
The BBC is funded by what essentially ammounts to a tax on TV usage. Even if I only want to watch ITV and SKY I still have to pay the BBC.
And the availibility of free news from the BBC has increased considerablly in recent years. A decade ago most people didn't have internet connections and only pay TV subscribers were likely to have BBC news 24 (I'm not sure if it was encrypted or not on digital terrestrial at that time but even if it wasn't digital terrestrial boxes were expensive back then) so the only news most people would have got from the BBC was a few programmes a day and a small ammount on ceefax.
Personally I think any internet news source that put itself behind a paywall would be committing suicide regadless of whether or not the BBC existed but lets not pretend that the BBC needs to make a profit.
IIRC they were pushed into drm on iplayer by some agency called "BBC Trust" which afaict is an organisation responsible for regulating the BBC.
I've always thought it rather crazy given that all the material has been previously broadcast as unencrypted DVB but I don't think it was actually the BBCs fault.
Afaict they actually have to render themselves physically unable to procreate (whether through death or otherwise) as a result of thier stupid act.
The Darwin awards are a grand invention but they can only be awarded posthumously.
Not strictly true, unable to procreate is considered sufficiant.
Umm, much of what Wal Mart sells is *exactly* the same brands that sell elsewhere, sometimes for less money.
I've heared (I dunno how true it is) that wal-mart often bullies manufacturers into producing slightly different models that are cheaper but also lower quality for them than the stuff that other retailers get.
In many cases, store generic brands are made by the same companies that make the 'name brand' items, though I know of no way of knowing if "Safeway ketchup" is really the same as Heinz.
It's not entirely about who makes an item though, it's also about what standards the customer (whether it's a big brand or a retailer) is willing to pay for.
Is saving a minuite or so per support call and/or trimming down the cost per call a little by allowing yourself (or your contractor) to hire idiots really worth pissing off your already pissed off customer even more?
Listen until they appear on the network.
Indeed, iff the pad is both perfectly random and perfectly protected from exposure then one time pads are perfectly unbreakable.
Both of these issues are problems in reality though,
Firstly you need real randomness for your pad, PRNG output DOES NOT cut it. If you use a PRNG then you effectively have a stream cipher with a key consisting of knowlage of the prng and the seeding information fed to the prng.
Distribution is also a big problem since you need as much key as data you can't use one time pad encryption to distribute your new pads, so you need physically secure communications to transfer the pad.
Define "one". A processor cannot do two operations at once. Everything is put into a line, and that one line is dealt with really really fast.
I think your understanding of computer architecture is a bit behind the times.
It used to be that computers fetched an intrustruction, decoded it, executed it and moved on to the next one.
Then came pipelining, this basically meant that they could start on one instruction before finishing the previous provided there weren't any data dependency issues blocking it.
Then came things like out of order execution and virtual registers to try and keep the pipeline full more of the time.
When that still wasn't enough to keep the hugely long pipeline of the P4 full they introduced hyperthreading so work could be taken from more than one thread at a time, a P4 really can be working on two totally seperate tasks at the same time.
With core 2 they dropped hyperthreading for a while but put two complete cores on the die so again the chip could be working on two totally seperate tasks at the same time (or four if you went for the version with two dies in the package).
With core i7 they brought back hyperthreading and kept two cores on the die. So it could be working on four seperate tasks at once (or eight if you went for the version with two dies in the package).
AMD has also been going in for multicore though I don't think they have used hyperthreading.
The impression I have got is that CPU clocks aren't likely to increase increase at anything like the rate they did in the past so our gains are going to come from chips that do more and more at the same time.
Of course what happens in a crash depends on more than just the relative speed.
If say a train (more realisitic than your moving wall) and a car are both moving at 60MPH towards each other for a combined closing speed then when they crash the car will absorb a huge ammount of energy very suddenly and be almost completely crushed. The train will likely keep going with minimal damage.
If two cars are both moving at 60MPH towards each other then damage should (theoretically) be equally split so the affect should be the same (same ammount of engergy dissipated in each car) as a single car moving at 60mph and hitting an immovable object.
So the GP is kind of right if we model a wall as an immovable object (which depending on the wall in quesion may or may not be a good model) and the car hits it head on then it probablly is as bad as a head on collision.
A sanely designed wall/barrier system will be designed so it is far more likely that a care makes a glancing blow into a barrier/wall and get deflected (so most of the energy in the car stays as kenetic energy in roughly it's original direction).
Plus there are a wide variety of vehicles on the roads, when say a truck hits a car you have a situation similar (but less extreme) than the one I mentioned above with the train.
Please watch this video [bbc.co.uk] and reconsider your habit of texting while driving.
"A shocking video has been made for pupils in an attempt to stop car crashes caused by the driver texting while driving.
The short film, starring young actors from south Wales, shows a teenage girl killing four people after she uses her mobile phone to send a text."
So lemme get this straight you as asking us to watch a governement propoganda film?
Can't you find any real videos of such accidents?
Note that this is all just claims based on tests under condtions that are supposed to accelerate ageing by some huge factor.
The problem with that of course is that such tests may accelerate some ageing mechanisms and not others and therefore not accurately predict what will happen decades down the line.
Iomega zipdrives, several sorts of tape drives, half a dozen different memory card standards... none of those were seen as fringe technologies at the time.
Theese may not have been fringe but they were never anywhere near as widely distributed as the 3.5 inch floppy was.
Afaict many lithium ion based devices don't have a bypass path (adding one complicates the power circuitry quite a bit, especially when you are dealing with the low voltage of a single cell lithium ion pack). So running on charge has to happen via the charger.
The trouble with this is that for safety reasons a typical lithium ion charger will only start a charge if it sees a good battery pack.
Afaict this is the reason that many mobile phones won't work with the battery pack removed even if you plug in a charger.
What i've generally found is that generic motherboards (either purchased seperately or as part of a PC build from a small system builder) are far more likely to have the legacy ports than those made for use in big brand machines.
Can't you just use a standard turntable and adjust the speed after digitising?