Actual braking distance isn't an issue. Your car is likely to slow down at roughly the same rate as the car in front. The important factor is the time (and distance travelled) between realising you need to stop and the brakes engaging. As long as you're further back when you actually brake than the car in front was.
Not that I want to discredit your advice. If the guy in front is acting like a moron I'll always give an extra second or so. If the guy behind is too close I'll probably go for a 4 second gap and if he has a problem he can go past and fill the gap in front of me.
Sorry, but nobody cares about ideas. Ideas are cheap. I have several every day. Why is your idea so much better than the ideas, complete with workable implementations that the designers have?
Okay - so it's a submersible aircraft rather than a flying sub. Slightly easier.
Not sure if it's acceptable for the passengers to require wetsuits. Since we're only after shallow depths pressure isn't going to be a big problem. Modern materials should be able to handle depths of a few feet. Weight is. Flooding the thing will mean less force required to keep it under.
Wings can be used as hydrofoils. They'll be a bot too good though. Perhaps they need to fold out of the way entirely or something. Extra front and back fins can be added which will provide some lift in the air and ability to keep the vehicle underwater as long as it's moving at a reasonable speed.
Perhaps a plane is the wrong way to go. Ground effect craft would have a lot of the advantages and can be made heavier without as much impact.
Having a recent warrant out for arrest, being already charged with a crime, out on bail, or already been convicted of a crime.
No longer having a warrant for arrest, having already been arrested and charged for all crimes that there was probably cause to suspect, or having been tried for all crimes with probably cause should suggest that - given a presumption of innocence - no other crimes have been committed.
I don't think the suspicions were reasonable. They seemed reasonable to the officers at the time, and I would certainly hate to see them punished for this, but allowing the evidence to stand when it's a mistake but not when it's malice doesn't seem to make any sense at all.
and hope the offeder's reputation is ruined enough that nobody will ever pay for his software again..
Sadly things don't often work out like that. Most people won't hear about this. The only thing the guy's lost at the moment is the exclusive rights to the software - something that wasn't his in the first place.
Most of the suggestions that GPL wouldn't stand up in court were from Free Software proponents. I always felt they were being a little paranoid. The GPL offers something (the right to redistribute the software) for a cost (the obligation to also redistribute source and any changes).
Traffic planners tend to use liquid flow analogies when talking about large numbers of vehicles (congestion, traffic flow, bottleneck) rather than truck-like analogies, so you end up back as a series of tubes.
Seems a pretty reasonable analogy to me. Internet bandwidth is limited. If someone sends 10Mb/s down a network then the bandwidth available to everyone else drops by 10Mb/s.
I wouldn't be so sure. Dedicated hardware is typically a lot cheaper than a general purpose CPU unless the tasks you want to do are extremely general. GPUs work very well with a simpler SIMD approach, and this can be extended to raytracing. It's an approach that works well for a lot of big number crunching tasks.
For more general purpose work, MIMD is useful. I have to wonder why Cell didn't take more cues from the Transputer. From what I've read, The Cell seems to be based on the idea of running multiple threads in parallel and having one core handle each thread. Always seemed rather inefficient. Seems that a better idea would be to package up the processes into a number of very short tasks, and assign each task to the next free core. This will, of course, require a totally different software architecture.
CPU speed is not the bottleneck. Lack of parallelism is.
I really have no idea what Window 95 does but I presume that for each piece of hardware it will initialise and wait for a response before attempting to initialise the next. Then for any startup scripts, it will load an application, run it, wait for it to terminate and then go onto the next. It's a very naive simple way of doing things that's more tailored to MS-DOS, but Microsoft programmers took a while to get out of that mindset.
I have to wonder about all the video stuff. Why can't you set the video mode in the kernel? Is X really the best solution? It always struck me as a little heavyweight for what it is (as far as I can see, a windowing system is essentially a reentrant API for drawing text and overlapping boxes). For most peoples needs I'm sure you could trim it down a lot and maintain all of the functionality that 99% of users expect.
Another thing I'm curious about is whether initrd might make sense if running using disks rather than flash.
If anything, prices will go up a dime. (Yes, for a 6 cent increase.)
Probably wouldn't happen. 99 cents is a key point. It's at the significant "less than a dollar" amount. Reduce the price to 98 cents and sales would barely go up at all. Increase to $1 and sales would drop by substantially more than the 3.3% extra they make per track. There would probably be another drop if they went up to $1.01.
Is Apple's share per song went to 24 cents, then a 10 cent increase would mean 41% more per track to Apple, but $1.09 psychologically feels substantially higher than 99 cents, and if Amazon manage to keep prices down to 99 cents then a lot of customers will go there instead.
Isn't Apple's agreement with the record labels? Surely the 9 cents comes out of the labels' share. That's what the cnn article implies. So instead of 29 cents going to Apple, 61 cents to the label and 9 cents to that guy who did nothing except all write and perform the song, it would be 29 cents to apple, 55 cents to the label and 15 cents to the worthless waste of space.
These days I'm thinking zero. Disks are slow. Chip based RAM is fairly fast. Linux is certainly less irritating than Windows when memory gets low but still not great.
If I run out of memory it's usually a lot quicker to get an "out of memory" error and kill a few other applications than to wait for the swapping to sort itself out.
Depends what you mean by "significant". May also hinge on the exact definition of "data analysis". Lots of data. Simple patterns. Excel is fine if it's the only tool you have.
Well.. If you happen to have a copy of Excel, know how to use it and have the right sort of data, it can be perfectly adequate. I use it a lot because I tend to have it installed (no idea why MS Office is seen as a requirement for development PCs), and I tend to only want to look at really simple data sets where I just want to plot a graph.
If you need to do something sufficiently complex that you feel you need to invest the time to read this book, I can't help feeling that learning to use Matlab or something might be a better use of the time.
Since I'm not an IT/admin guy, I daresay you're right. I just felt the whole "cloud computing" idea seemed like a really tricky system to set up for what sounded like quite a small benefit, and was hoping to start some discussion of worthwhile alternatives.
I'm actually rather disappointed it took so long for someone to respond.
The Cogwheel Brain by Doran Swade. Quite a hefty tome to prove my point, but a decent read anyway. There's also this article, which doesn't really do a lot to back up the argument.
Swade actually talks considerably about Ada Lovelace and the argument is pretty compelling.
People have usually decided whether they're going to hire you after the first couple of minutes. They often don't really know the reason for rejecting other than "a feeling", but still feel the need to justify their decision.
Quite so. Eric Raymond is much more well known as a writer, Larry Ellison and Bill Gates are really famous because they were successful businessmen. Even for all his work on GCC and Emacs, Richard Stallman is probably best known for the FSF.
Slashdot Mirror
Actual braking distance isn't an issue. Your car is likely to slow down at roughly the same rate as the car in front. The important factor is the time (and distance travelled) between realising you need to stop and the brakes engaging. As long as you're further back when you actually brake than the car in front was.
Not that I want to discredit your advice. If the guy in front is acting like a moron I'll always give an extra second or so. If the guy behind is too close I'll probably go for a 4 second gap and if he has a problem he can go past and fill the gap in front of me.
To whoever: There are a lot of quests in MMOs. Many of them are designed for single players.
Why? Why make a Massively multiplayer game and then try to shoehorn it into a single player game?
Sorry, but nobody cares about ideas. Ideas are cheap. I have several every day. Why is your idea so much better than the ideas, complete with workable implementations that the designers have?
Okay - so it's a submersible aircraft rather than a flying sub. Slightly easier.
Not sure if it's acceptable for the passengers to require wetsuits. Since we're only after shallow depths pressure isn't going to be a big problem. Modern materials should be able to handle depths of a few feet. Weight is. Flooding the thing will mean less force required to keep it under.
Wings can be used as hydrofoils. They'll be a bot too good though. Perhaps they need to fold out of the way entirely or something. Extra front and back fins can be added which will provide some lift in the air and ability to keep the vehicle underwater as long as it's moving at a reasonable speed.
Perhaps a plane is the wrong way to go. Ground effect craft would have a lot of the advantages and can be made heavier without as much impact.
Having a recent warrant out for arrest, being already charged with a crime, out on bail, or already been convicted of a crime.
No longer having a warrant for arrest, having already been arrested and charged for all crimes that there was probably cause to suspect, or having been tried for all crimes with probably cause should suggest that - given a presumption of innocence - no other crimes have been committed.
I don't think the suspicions were reasonable. They seemed reasonable to the officers at the time, and I would certainly hate to see them punished for this, but allowing the evidence to stand when it's a mistake but not when it's malice doesn't seem to make any sense at all.
No
Yes.
and hope the offeder's reputation is ruined enough that nobody will ever pay for his software again..
Sadly things don't often work out like that. Most people won't hear about this. The only thing the guy's lost at the moment is the exclusive rights to the software - something that wasn't his in the first place.
Most of the suggestions that GPL wouldn't stand up in court were from Free Software proponents. I always felt they were being a little paranoid. The GPL offers something (the right to redistribute the software) for a cost (the obligation to also redistribute source and any changes).
I'm surprised it got as far as court.
Traffic planners tend to use liquid flow analogies when talking about large numbers of vehicles (congestion, traffic flow, bottleneck) rather than truck-like analogies, so you end up back as a series of tubes.
Seems a pretty reasonable analogy to me. Internet bandwidth is limited. If someone sends 10Mb/s down a network then the bandwidth available to everyone else drops by 10Mb/s.
I wouldn't be so sure. Dedicated hardware is typically a lot cheaper than a general purpose CPU unless the tasks you want to do are extremely general. GPUs work very well with a simpler SIMD approach, and this can be extended to raytracing. It's an approach that works well for a lot of big number crunching tasks.
For more general purpose work, MIMD is useful. I have to wonder why Cell didn't take more cues from the Transputer. From what I've read, The Cell seems to be based on the idea of running multiple threads in parallel and having one core handle each thread. Always seemed rather inefficient. Seems that a better idea would be to package up the processes into a number of very short tasks, and assign each task to the next free core. This will, of course, require a totally different software architecture.
CPU speed is not the bottleneck. Lack of parallelism is.
I really have no idea what Window 95 does but I presume that for each piece of hardware it will initialise and wait for a response before attempting to initialise the next. Then for any startup scripts, it will load an application, run it, wait for it to terminate and then go onto the next. It's a very naive simple way of doing things that's more tailored to MS-DOS, but Microsoft programmers took a while to get out of that mindset.
I have to wonder about all the video stuff. Why can't you set the video mode in the kernel? Is X really the best solution? It always struck me as a little heavyweight for what it is (as far as I can see, a windowing system is essentially a reentrant API for drawing text and overlapping boxes). For most peoples needs I'm sure you could trim it down a lot and maintain all of the functionality that 99% of users expect.
Another thing I'm curious about is whether initrd might make sense if running using disks rather than flash.
If anything, prices will go up a dime. (Yes, for a 6 cent increase.)
Probably wouldn't happen. 99 cents is a key point. It's at the significant "less than a dollar" amount. Reduce the price to 98 cents and sales would barely go up at all. Increase to $1 and sales would drop by substantially more than the 3.3% extra they make per track. There would probably be another drop if they went up to $1.01.
Is Apple's share per song went to 24 cents, then a 10 cent increase would mean 41% more per track to Apple, but $1.09 psychologically feels substantially higher than 99 cents, and if Amazon manage to keep prices down to 99 cents then a lot of customers will go there instead.
Isn't Apple's agreement with the record labels? Surely the 9 cents comes out of the labels' share. That's what the cnn article implies. So instead of 29 cents going to Apple, 61 cents to the label and 9 cents to that guy who did nothing except all write and perform the song, it would be 29 cents to apple, 55 cents to the label and 15 cents to the worthless waste of space.
These days I'm thinking zero. Disks are slow. Chip based RAM is fairly fast. Linux is certainly less irritating than Windows when memory gets low but still not great.
If I run out of memory it's usually a lot quicker to get an "out of memory" error and kill a few other applications than to wait for the swapping to sort itself out.
Depends what you mean by "significant". May also hinge on the exact definition of "data analysis". Lots of data. Simple patterns. Excel is fine if it's the only tool you have.
Well.. If you happen to have a copy of Excel, know how to use it and have the right sort of data, it can be perfectly adequate. I use it a lot because I tend to have it installed (no idea why MS Office is seen as a requirement for development PCs), and I tend to only want to look at really simple data sets where I just want to plot a graph.
If you need to do something sufficiently complex that you feel you need to invest the time to read this book, I can't help feeling that learning to use Matlab or something might be a better use of the time.
This is exactly wrong.
Since I'm not an IT/admin guy, I daresay you're right. I just felt the whole "cloud computing" idea seemed like a really tricky system to set up for what sounded like quite a small benefit, and was hoping to start some discussion of worthwhile alternatives.
I'm actually rather disappointed it took so long for someone to respond.
Work out how to build the servers as cheaply as possible. If peak load starts to get troublesome, add some more servers.
Cloud computing is a buzzword. Big server farms are may be dull, but it's a tried and tested technology that works. Ask Google.
The Cogwheel Brain by Doran Swade. Quite a hefty tome to prove my point, but a decent read anyway. There's also this article, which doesn't really do a lot to back up the argument.
Swade actually talks considerably about Ada Lovelace and the argument is pretty compelling.
People have usually decided whether they're going to hire you after the first couple of minutes. They often don't really know the reason for rejecting other than "a feeling", but still feel the need to justify their decision.
Work on interview technique.
Quite so. Eric Raymond is much more well known as a writer, Larry Ellison and Bill Gates are really famous because they were successful businessmen. Even for all his work on GCC and Emacs, Richard Stallman is probably best known for the FSF.
Overrated. didn't do nearly as much as she was given credit for. Babbage just had a soft spot for the girl.
Try Grace Hopper. One of the UNIVAC programmers who I believe did a lot of early work on compilers.