Also whilst weapons that are designed to blind people are banned by the Geneva convention or the equivalent, many lasers used in laser targeting systems aren't explicitly designed to blind people but often can.
Even if the final system can't track a Mach 6 object, I'm sure it can be manually aimed at a stationary target making a speech 300km away.
Better if you can fire the laser in a nonvisible light wavelength, AND fire it so it passes close to a big thundercloud - ionizing the air and inducing a big visible lightning zap from the cloud to the target. Go look up laser induced lightning.
Then it'll look like an "Act of God" - the target apparently being killed by lightning.
Why shoot down missiles if you can shoot the person that orders them to be launched and make it look like it was a freak incident?
Potential apps? Sure, I can see plenty. In fact I wonder if assassination could actually be one of the main apps (secret of course - rather politically incorrect).
Interesting. Did you do a lot of update+commits or just big updates and fewer commits?
Maybe someone should have a "HDD" that's battery backed 1-2GB RAM for WALs or stuff like that. Rechargeable lead-acid gel batteries (2+ year lifespan).
Exactly. Also if people can smuggle in kilos and kilos of cocaine despite the billion dollar war on drugs, makes you wonder how successful they can be.
To stop such things the US would probably have to declare martial law and enforce curfews, border checks and all that...
And that means the enemies kinda win - the economy, freedom etc would be badly affected.
The US should have spent billions on spreading peace instead of war. Now with so much fresh blood on their hands it's going to be much harder.
"Such as not running as root (unless you are talking about Linspire)."
There you go - that's the problem. Think about why people do that.
I don't run windows as an admin by default. Nor do savvy users.
But neither do I open dubious attachments, enter the numeric password supplied in the email, and then run the executable.
There's no big difference between Linux/BSD and Windows NT/2K/XP in security, architecturally.
If you had an O/S that by default ran user programs in a sandbox - sound, graphics, write access to certain places only (or rollbackable), that would be more secure for most users.
Finding stuff worth reading - who wants to keep up with crap? There doesn't seem to be a torrent of info (please prove me wrong), just a torrent of crap. Most of the stuff is just old or crap/wrong or redundant. Why waste plenty of time reading stuff you already know/can trivially figure out or is wrong?
If you have a good idea of how things work in a particular area, a good source of info can often allow you to very quickly understand what's so innovative/cool about this "new thing". Unfortunately such sources are quite rare.
The rate of technological progress so far doesn't seem that fast to me (medical, computer, physics, bio, engineering, automotive, aerospace). It's not surprising though - since it usually takes longer to come up with something significant than to understand why it is significant.
I mean, yet another hot fusion idea that is far far from "breaking-even", with no significant advantages. Doh. Like who cares.
Whereas if you have: "Major advance in hot fusion!" or even "Cold Fusion phenomena explained near conclusively - not fusion but due to...." - now these are stuff I'd find interesting.
Or "one successful clone made from Yet Another Animal out of 500 tries". Who cares other than those directly involved. Whereas "Experimental cure for XXxx cancer does well in test stage" now that would be interesting.
Sure spaceshipone was interesting, but they are pretty far from orbit. And people have already done orbit decades ago, so there really isn't very much for a reader to catch up on is there? The aerospace field seems quite stagnant compared to the days of U2, SR71, B70, F-14 to F16, first man in orbit, man on moon, concorde, 747, Harrier (in no particular order).
Not interested in vapourware either OK?
It's hard to find stuff worth reading - so much so I even resort to reading Slashdot. Like what are you doing here anyway - checking for dupes?;).
What are you using it for? RAID5 is slow for writes. Plus AFAIK rebuilding a RAID5 is a LOT slower and hurts performance more than rebuilding stripes of mirrored drives.
What are the applications? Would you be needing more drive space or adding drives in the near future?
If it's a single application and you know you are not going to need to expand soon or add/change apps - drives are BIG compared to what app needs AND performance is _THE_ issue. Then I'd suggest optimizing around the application.
For example if you have an OLTP postgresql database app, you should put the transaction logs on a separate array optimized for writing speed, then put the tables, indexes on another array, and maybe the temporary sort directories on yet another array.
Even though striping etc helps, if you understand how the application works, it'll often perform better with each type of access on separate arrays. Of course if the next version of the application changes the way it does thing drastically then all bets are off.
You should probably put the O/S stuff on a mirror/array by itself, so the O/S can go do whatever it wants and mostly not affect your main application throughput.
Typically there are diminishing returns as you optimize more exactly to what the application needs, and you usually lose flexibility the more you tailor the configuration to a specific scenario.
So go for big wins. Forget about the last 5-10% performance gains if you are going to lose flexibility or add significant administration cost (complexity).
I still stick to using cat and | because I'm afraid that one day I may type > instead of < and on some systems that overwrites the file... After all they are next to each other on my keyboard.
In my experience (as a IT security guy) C and PHP are more dangerous than perl.
C - "runs arbitrary code of the attacker's choice" given _common_ stupid programmer mistakes.
PHP - developers fond of features that encourage bad/insecure ways of doing things - e.g. magic quotes, global track vars. Take away such popular PHPisms and PHP starts to look like perl;).
They found that the treatment/drugs supposed to prepare the mice for transplant actually _cured_ the mice. The treatment was supposed to stop the immune system from blowing away the transplanted islet cells. But after the treatment the islet cells _regrew_ back. So there was no need for a transplant.
""No one had cured them," he said. "Here was this treatment that we thought would get them ready for a transplant but - eureka! - the diabetes was cured.""
I'd stop beating the dead horse if it's dead. But it's obviously far from dead. In fact it's giving the IT Security industry a pretty good and profitable ride.
"You avoid buffer overflows in C by checking the lengths of your buffers. You stop using C strings. You use container libraries. As for C++, you avoid them by using the included string and container classes."
You avoid mistakes by getting things right. LOL.
How about stop using C and C++? True it's hard. It's harder to switch than to continue using C or C++.
But which is harder - switching to some other language or using C or C++ CORRECTLY? Using C or C++ _correctly_ appears to be beyond the competence of most programmers.
C is an unsafe design. To use your car analogy, the widespread use of C is the widespread use of cars without any protection at all - no frame around the driver, no airbags, no seat belts. Just engine, drive-train and tyres. Bare-metal...
Each driver has to build their own protection. If they get practically anything wrong, they're screwed if anything happens. By the time the drivers have added their own custom/3rd party protection most have probably lost any significant performance benefits of having the "raw" car. Or the protection is ineffective - after all most drivers aren't experts in safety - they just want to get from point A to point B.
Sure a few programmers CAN program in C. But most of them obviously can't - what are the odds of them getting EVERY statement right? The typical slashdotter here talking about programming can barely spell right or get his/her grammar correct.
You say: "My position is that _every_ statement written in a program needs to be written with consideration for its bounds of validity. I don't believe there are any magic languages out there that will save us from having to think so hard, certainly not if correctness is our ultimate goal. "
That's actually a sign of bad design to me, if you have to think so hard for each line. Either the program is badly designed, or the language is.
If you "firewall/partition off" sections of code, you can write lots of code without thinking hard and it still won't cause you grief much less "make it possible to execute arbitrary code of the attacker's choice". After all I don't secure every single valuable item in my house. I just stick to securing the entry points, a few rooms and the safe.
It seems that C makes it hard for programmers to program securely - many still screw up at the "doors and windows" of their code even though they are supposedly concentrating hard on getting those right!
Don't get me wrong - I work as an IT Security consultant, so if you programmers keep insisting on using unsafe languages and saying the problem is because programmers aren't getting "every statement" right, hey go right on ahead...
It's just kinda boring to see the same _stupid_ bugs year after year (decade after decade?), when you KNOW there are alternatives.
I've played the same game (quakeworld) through the years at 15fps, 30fps, at 60fps etc. And believe me 30fps is way far from "as good as it gets". I even resorted to 320x240 to get decent frame rates on my P166.
30fps is not terrible, but nowadays it's sluggish. For games like GTA3, 30fps is OK (it's default is locked to 30fps). But it still feels sluggish.
Compare GTA3 3rd person with Jedi Knight II 3rd person (with saber) on a decent system and you'll see the diff between 30fps and whatever it is your machine can actually go up to.
If you don't see the diff then either you don't have a decent system, or you can save a fair bit of money - you shouldn't bother upgrading for the next few years.
Of course there's another thing - FPS isn't the only thing that affects "sluggishness". Latency is another. Even if your video updates are 300FPS, it's still sluggish if you press a key and the game only acts on it 0.2 seconds later (though rendering it at 300FPS). It's not sluggish if the game renders at 100fps and acts on your inputs on the very next frame.
Still, the two are interlinked in practice. While a game could act on your inputs before drawing the next frame, you only see it in the next frame- so the feedback isn't instantaneous. At 30fps, 33ms is still a noticeable delay when you are controlling something, while that's not that bad (cause I'm a HPB used to delays of 300ms), it's not great. Plus it's 33ms on top of network delays if you're playing online.
Re:Never having played either myself....
on
Halo 2 Reviews
·
· Score: 1
Your eyes maybe. Not mine, and I doubt mine are that great.
Even if your eyes may not be able to discern individual frames, since eyes don't work by frames, you might still notice or perceive the difference in framerate.
I've watched LoTR and when the movie goes for those scenery pans, while the res is good, the "refresh rate" really sucks.
If you can tell the difference between the mouse pointer movement in Windows 95 and the mouse pointer movement in Windows NT/2000/XP then > 40 FPS makes a difference for you. I had to download a utility called ps2rate to set the mouse sampling to something decent for Win95 (100 to 200 are nice figures).
Go try it out - set your monitor refresh rate to as high as possible, then change the mouse sampling rate to 40, move the mouse pointer around and then try it at 100.
"But that's because the mouse pointer jumps", well whatever it is your eyes sure noticed the pointer jumped greater distances between samples didn't they?
BTW show me evidence that the average human eye can't discern frames beyond 30FPS.
Slashdot Mirror
Also whilst weapons that are designed to blind people are banned by the Geneva convention or the equivalent, many lasers used in laser targeting systems aren't explicitly designed to blind people but often can.
They could use this for assassination.
Even if the final system can't track a Mach 6 object, I'm sure it can be manually aimed at a stationary target making a speech 300km away.
Better if you can fire the laser in a nonvisible light wavelength, AND fire it so it passes close to a big thundercloud - ionizing the air and inducing a big visible lightning zap from the cloud to the target. Go look up laser induced lightning.
Then it'll look like an "Act of God" - the target apparently being killed by lightning.
Why shoot down missiles if you can shoot the person that orders them to be launched and make it look like it was a freak incident?
Potential apps? Sure, I can see plenty. In fact I wonder if assassination could actually be one of the main apps (secret of course - rather politically incorrect).
Shhhh! Duh.
Interesting. Did you do a lot of update+commits or just big updates and fewer commits?
Maybe someone should have a "HDD" that's battery backed 1-2GB RAM for WALs or stuff like that. Rechargeable lead-acid gel batteries (2+ year lifespan).
Would 1-2GB be good enough for such things?
Exactly. Also if people can smuggle in kilos and kilos of cocaine despite the billion dollar war on drugs, makes you wonder how successful they can be.
To stop such things the US would probably have to declare martial law and enforce curfews, border checks and all that...
And that means the enemies kinda win - the economy, freedom etc would be badly affected.
The US should have spent billions on spreading peace instead of war. Now with so much fresh blood on their hands it's going to be much harder.
"Such as not running as root (unless you are talking about Linspire)."
There you go - that's the problem. Think about why people do that.
I don't run windows as an admin by default. Nor do savvy users.
But neither do I open dubious attachments, enter the numeric password supplied in the email, and then run the executable.
There's no big difference between Linux/BSD and Windows NT/2K/XP in security, architecturally.
If you had an O/S that by default ran user programs in a sandbox - sound, graphics, write access to certain places only (or rollbackable), that would be more secure for most users.
It's not The People's fault that they install Lexmark's drivers when they buy Lexmark's printers AND they get spied on by Lexmark.
Let's not change the norms of proper social and civilized behaviour OK?
I dunno, do crack dealers make it a habit of spying on their users to find out how their users are using their crack, where, how many times etc?
I don't think crack dealers would last long in their business if they did that.
Finding stuff worth reading - who wants to keep up with crap? There doesn't seem to be a torrent of info (please prove me wrong), just a torrent of crap. Most of the stuff is just old or crap/wrong or redundant. Why waste plenty of time reading stuff you already know/can trivially figure out or is wrong?
...." - now these are stuff I'd find interesting.
;).
If you have a good idea of how things work in a particular area, a good source of info can often allow you to very quickly understand what's so innovative/cool about this "new thing". Unfortunately such sources are quite rare.
The rate of technological progress so far doesn't seem that fast to me (medical, computer, physics, bio, engineering, automotive, aerospace). It's not surprising though - since it usually takes longer to come up with something significant than to understand why it is significant.
I mean, yet another hot fusion idea that is far far from "breaking-even", with no significant advantages. Doh. Like who cares.
Whereas if you have: "Major advance in hot fusion!" or even "Cold Fusion phenomena explained near conclusively - not fusion but due to
Or "one successful clone made from Yet Another Animal out of 500 tries". Who cares other than those directly involved. Whereas "Experimental cure for XXxx cancer does well in test stage" now that would be interesting.
Sure spaceshipone was interesting, but they are pretty far from orbit. And people have already done orbit decades ago, so there really isn't very much for a reader to catch up on is there? The aerospace field seems quite stagnant compared to the days of U2, SR71, B70, F-14 to F16, first man in orbit, man on moon, concorde, 747, Harrier (in no particular order).
Not interested in vapourware either OK?
It's hard to find stuff worth reading - so much so I even resort to reading Slashdot. Like what are you doing here anyway - checking for dupes?
What value of pi did you use?
What are you using it for? RAID5 is slow for writes. Plus AFAIK rebuilding a RAID5 is a LOT slower and hurts performance more than rebuilding stripes of mirrored drives.
What are the applications? Would you be needing more drive space or adding drives in the near future?
If it's a single application and you know you are not going to need to expand soon or add/change apps - drives are BIG compared to what app needs AND performance is _THE_ issue. Then I'd suggest optimizing around the application.
For example if you have an OLTP postgresql database app, you should put the transaction logs on a separate array optimized for writing speed, then put the tables, indexes on another array, and maybe the temporary sort directories on yet another array.
Even though striping etc helps, if you understand how the application works, it'll often perform better with each type of access on separate arrays. Of course if the next version of the application changes the way it does thing drastically then all bets are off.
You should probably put the O/S stuff on a mirror/array by itself, so the O/S can go do whatever it wants and mostly not affect your main application throughput.
Typically there are diminishing returns as you optimize more exactly to what the application needs, and you usually lose flexibility the more you tailor the configuration to a specific scenario.
So go for big wins. Forget about the last 5-10% performance gains if you are going to lose flexibility or add significant administration cost (complexity).
If you get paralyzed on the segway, you can switch to one of Kamen's motorized wheelchairs ;).
Talk about creating demand...
But you're going to eat anyway right, even if you buy a roomba?
Are you going to eat significantly less if you have a roomba?
"I'm just about to finish a world wide, 3-tier, trouble ticketing system in the shape of a Maze"
"It[Xgen] is also an open source project so you could see the source code."
Right. Next thing you'd be linking to goatse.cx.
"I disagree with this comment classifying and naming software makes it sound alive. I think we should describe or label it."
Sound alive? How about "toe tagging" it then?
I still stick to using cat and | because I'm afraid that one day I may type > instead of < and on some systems that overwrites the file... After all they are next to each other on my keyboard.
How's perl dangerous?
;).
In my experience (as a IT security guy) C and PHP are more dangerous than perl.
C - "runs arbitrary code of the attacker's choice" given _common_ stupid programmer mistakes.
PHP - developers fond of features that encourage bad/insecure ways of doing things - e.g. magic quotes, global track vars. Take away such popular PHPisms and PHP starts to look like perl
Yeah if it's child porn, you might even get to share a cell...
The odds are good - I'm sure the medical establishment has far more successful ways of curing or treating mice than humans.
Read the article again till the end.
They found that the treatment/drugs supposed to prepare the mice for transplant actually _cured_ the mice. The treatment was supposed to stop the immune system from blowing away the transplanted islet cells. But after the treatment the islet cells _regrew_ back. So there was no need for a transplant.
""No one had cured them," he said. "Here was this treatment that we thought would get them ready for a transplant but - eureka! - the diabetes was cured.""
Who's trolling here?
I'd stop beating the dead horse if it's dead. But it's obviously far from dead. In fact it's giving the IT Security industry a pretty good and profitable ride.
"You avoid buffer overflows in C by checking the lengths of your buffers. You stop using C strings. You use container libraries. As for C++, you avoid them by using the included string and container classes."
You avoid mistakes by getting things right. LOL.
How about stop using C and C++? True it's hard. It's harder to switch than to continue using C or C++.
But which is harder - switching to some other language or using C or C++ CORRECTLY? Using C or C++ _correctly_ appears to be beyond the competence of most programmers.
C is an unsafe design. To use your car analogy, the widespread use of C is the widespread use of cars without any protection at all - no frame around the driver, no airbags, no seat belts. Just engine, drive-train and tyres. Bare-metal...
Each driver has to build their own protection. If they get practically anything wrong, they're screwed if anything happens. By the time the drivers have added their own custom/3rd party protection most have probably lost any significant performance benefits of having the "raw" car. Or the protection is ineffective - after all most drivers aren't experts in safety - they just want to get from point A to point B.
Sure a few programmers CAN program in C. But most of them obviously can't - what are the odds of them getting EVERY statement right? The typical slashdotter here talking about programming can barely spell right or get his/her grammar correct.
You say: "My position is that _every_ statement written in a program needs to be written with consideration for its bounds of validity. I don't believe there are any magic languages out there that will save us from having to think so hard, certainly not if correctness is our ultimate goal.
"
That's actually a sign of bad design to me, if you have to think so hard for each line. Either the program is badly designed, or the language is.
If you "firewall/partition off" sections of code, you can write lots of code without thinking hard and it still won't cause you grief much less "make it possible to execute arbitrary code of the attacker's choice". After all I don't secure every single valuable item in my house. I just stick to securing the entry points, a few rooms and the safe.
It seems that C makes it hard for programmers to program securely - many still screw up at the "doors and windows" of their code even though they are supposedly concentrating hard on getting those right!
Don't get me wrong - I work as an IT Security consultant, so if you programmers keep insisting on using unsafe languages and saying the problem is because programmers aren't getting "every statement" right, hey go right on ahead...
It's just kinda boring to see the same _stupid_ bugs year after year (decade after decade?), when you KNOW there are alternatives.
Heh, in my business the customer is often wrong.
If the customer is always right, it's far less likely they'd need IT consultants, eh?
But of course we don't put it that way...
30fps IS sluggish.
And it's nothing to do with movement speed.
I've played the same game (quakeworld) through the years at 15fps, 30fps, at 60fps etc. And believe me 30fps is way far from "as good as it gets". I even resorted to 320x240 to get decent frame rates on my P166.
30fps is not terrible, but nowadays it's sluggish. For games like GTA3, 30fps is OK (it's default is locked to 30fps). But it still feels sluggish.
Compare GTA3 3rd person with Jedi Knight II 3rd person (with saber) on a decent system and you'll see the diff between 30fps and whatever it is your machine can actually go up to.
If you don't see the diff then either you don't have a decent system, or you can save a fair bit of money - you shouldn't bother upgrading for the next few years.
Of course there's another thing - FPS isn't the only thing that affects "sluggishness". Latency is another. Even if your video updates are 300FPS, it's still sluggish if you press a key and the game only acts on it 0.2 seconds later (though rendering it at 300FPS). It's not sluggish if the game renders at 100fps and acts on your inputs on the very next frame.
Still, the two are interlinked in practice. While a game could act on your inputs before drawing the next frame, you only see it in the next frame- so the feedback isn't instantaneous. At 30fps, 33ms is still a noticeable delay when you are controlling something, while that's not that bad (cause I'm a HPB used to delays of 300ms), it's not great. Plus it's 33ms on top of network delays if you're playing online.
Your eyes maybe. Not mine, and I doubt mine are that great.
Even if your eyes may not be able to discern individual frames, since eyes don't work by frames, you might still notice or perceive the difference in framerate.
I've watched LoTR and when the movie goes for those scenery pans, while the res is good, the "refresh rate" really sucks.
If you can tell the difference between the mouse pointer movement in Windows 95 and the mouse pointer movement in Windows NT/2000/XP then > 40 FPS makes a difference for you. I had to download a utility called ps2rate to set the mouse sampling to something decent for Win95 (100 to 200 are nice figures).
Go try it out - set your monitor refresh rate to as high as possible, then change the mouse sampling rate to 40, move the mouse pointer around and then try it at 100.
"But that's because the mouse pointer jumps", well whatever it is your eyes sure noticed the pointer jumped greater distances between samples didn't they?
BTW show me evidence that the average human eye can't discern frames beyond 30FPS.