I think it's more productive/useful to have a more systematic "version control system" on the high level congress bills (the policy) rather then trying to track the nitty-gritty details of what each department/federal agency does. (the mechinisms) Then it's more possible to track intent. If one of the FAs violates the intent, it would easier to catch them as well as holding the policy makers more accountable.
While I don't think that this would stop any log-rolling, at these it would give constitutants more ideas about how good/crappy their representative has been. Combine that with a lock of 24-48 hours on any bill, and you'd make congress alot more transparent.
I was a little confused how to take screenshots and create a text file at first, but I checked the applications list and found applications called "Take Screenshot" and "Text Editor." Doesn't get much more obvious than that.
Someone has been taking their usuability studies seriously...
As an addendum, as far as I know, the technians in charge of the systems on a US navy have to know the wiring diagrams of the systems they use in case there's a malfunction and a component has to be replaced. It's insane that they can't do the same with the software running on top of electrical systems, especially incase a software component needs to be replaced. I can easily imagine a driver or core kernel service needing to be modified because a sensor malfunctions or an acuator isn't performing as needed.
But money and influnce rules in the game of military purchances. I guess the ability to adapt to changing conditions doesn't matter until people start dying.
Windows is on 90+% of the world's computers, and absolutely every younger person knows how to navigate around in it.
You must be kidding. The UIs for CCC (command, control and communication) systems on a warship are above the basic windowing system. There's NO WAY a new recruit will know how the system works without at least several months/years of training. If he has to go to the hardware manager because a driver isnt working properly during combat, there is something seriously f--ked and he's probably already dead. These have to be kiosk systems where the operator only needs to work with a couple apps at a time. Unfortunately, since they have to use windows and the kernel/windowing system breaks down on them at sea because of a unforseen situation, they will be shit out of luck until they can get a professional rep to diagonios and rebuild their kernel. With an opensource kernel/windowing system, they could update and patch in real time.
I think that the M$ marketing/sales people are absolutely brillent. They can actually convince the British and US navies that it's better to be running winCE without allowing the people onboard assess to source code then just use opensource where the code is standardized and easy to understand and could be updated in realtime.
I think the chinese are going to win the next great war because they don't give in to that kind of bullshit. Ben
Yeah, it seems like the better way to leave the asserts in the release code because users are more likely to stumble across bugs then the testers are. Your "million-dollar" code is far too complex and massize to be fully tested by any normal size QA department or for all of the corner cases to be thought of by the programmers.
A better solution however is to use exceptions and exception-handling so the application can recover from the error by alerting the user, emailing an error report, and trying to rollback to a previous stable configuration.
However, way too many people bitch and whine about how much overhead exceptions take so expect to see more crashing asserts.
The fact that the WHOLE system has to reboot if just one entertainment application goes down seems embressing/stupid to me. Sounds like overly simplistic program management. They should have used a watchdog program to ensure that each process is up and running like normal and if they crash, they get restarted individually, not the entire system. that way, if you crash your console, you won't affect the experience of the other people around you.
Besides, is oracle liable for your data loss when you lose your oracle instance? MS when you lose your IIS website?
No (because of the EULA) but they should be. Otherwise, why did you shell out the big bucks for Oracle or IIS? You might as well use apache. Likewise, if you buy RHEL and apache falls over even after you installed it properly, Redhat should be liable.
but IANAL so this could be all pie in the sky when it comes to an actual lawsuit.
"Security Researchers" are threatened with prosecution? Oh, mean hackers.
It doesn't help when normal researchers (see http://blog.wired.com/27bstroke6/2006/10/make_your _own_f.html) is threatened by the FBI for pointing out that a flawed security mechinism is flawed. It wouldn't stop the real hackers from exploiting it and causing REAL damage and the FBI did us a REAL diservice by threatening and scaring someone who could have helped make the system actually work properly.
So yess, the FBI has been lead-footed recently when dealing with "security researchers"
This meant that the robot could do a handful of simple, pre-defined motions, and the Wii-Mote was simply used to select the closest available match.
They could do better though.
They should record only very small motion paths (its accurate to 1/100 inch occuring TFA). Then have the robot play each piece when it receives it from the Wii-Remote. They'd have to record alot more motions but it would be smoother, more immediate, and more realistic in the end.
Sun's handling of Java gave Microsoft enough time to make.NET a killer platform, especially for Web apps.
I hate to break it but.NET is a bit behind the curve for net apps. its either ruby or python or PHP nowadays. I supposed you could point out iron-python but there's noing different there from normal python development wise.
When all the smoke settles, comptuters are for doing work.
So what do you want to work on? Find something that itchs you hard and try to make it better. Otherwise, I find firefox perfect for slashdotting all day and night....;)
I've been wondering about how secure the repos are. I mean, everyone seems to trust them 100%, they're not commercially/government supported, and once you get date-activated sploits in, you wouldn't be able to detect them until the activation time.
Is there an established method for making the repos more accountable? Maybe we need to buy insurance for our distros.
Sure, it's possible to bury the key inside a dongle distributed with each software player. However, it's annoying to work with and difficult to distribute
The video game industry and the mid-high end design industry tried dongles and both failed. The only place I've seen it work was at a company which made textile design software which was extremely specific and extremely expensive.
I seriously doubt any mass-media software will be tolerable with dongles.
Maybe the GP is refering to the fact that M$ is trying to compete with google in building $500 million dollar data centers to run thier live service. And what OS is M$ going to use to power this monster data center? It could be Linux, it could be Solaris. I seriously doubt it'll be server2003.
I predict within 5 years, Vista will become a legacy product, and all new computers will be shipped with Microsoft's Windows API hosted on a Linux kernel.
I think my head exploded when you said that! Do you realize how completely different the windows and posix APIs are. M$ has developed completely different standards for filetree (C:,D:,E:,... vs mounts into a single file tree), using "\" instead of "/", allowing spaces in names, different threading model, different scheduler model, a very vertical integration of components (vs. linux's relatively broad layering of functionality) and other hopelessness. Add to the fact that M$ would have to pretty much rewrite ALL of their applications to run on a posix platform and I think the likely hood of this happening is ZERO.
However, there are several companies trying to make Linux more user friendly. Just look at Gnome and KDE.
What makes you think that? AFAIK fuse is written in C (as a driver inside the kernel) and I would expect the plugin file systems to be written in C/C++ (although I know there is ruby and probably python bindings) Why should fuse be any slower then kioslaves.
The real trick is to keep the extraction software secret
Naa, you could just use a private key with a public encryption/decryption algorithm.
In fact, if you're clever enough, I sure you could build a watermark system which has a public key to verify that the watermark exists and is "correct" but only a private key would allow to you locate the watermark...
Why don't you just subtract the watermark from itself?
How do you know that the watermark is in the color domain? More likely, it'll be in the frequency domain because it holds better when the file is analog copied. Plus, the watermark will be in a random place or somewhere determined by a encyption scheme. It could even be like using a public key! I seriously doubt that you could break a well designed watermark scheme. It is NOTHING like DRM.
The article quotes one of the generals as saying the projectial is "like a ford taurus hitting the target at 380 mph" Ford taurus weigths 1383 kg (http://en.wikipedia.org/wiki/Ford_Taurus) and 380 mph -> 169 m/s 1383*169 => 234937.402 kgm/s. Projectial weighs 3.2 kg => it hits the target at 73417.9381m/s. This seems extremely high
I feel like I'm off by an order of magnitude. Guess it would be better to work off of the projection that the projectial can reach 95 miles into space to calculate initial speed.
I'm a huge kde/kioslaves fan too but IMHO fuse is really the way to go. It'll allow all unix applications to access those interesting FSs in a common manner. The only problem is that the type (fish:,http:,ipod:,etc.) gets thrown out the window and mapping isn't as flexible as having a URI. But it makes more sense in the long run...
Cheers Ben
Re:Performance, anyone?
on
Lisp and Ruby
·
· Score: 1
This might be interesting to check out. (except for using indentation as scope indicator. should be replaced with { } IMHO)
Cheers Ben
Re:Performance, anyone?
on
Lisp and Ruby
·
· Score: 1
Why would we want to rewrite a subset of Lisp in Ruby???
Maybe because not everyone enjoys trying to think in reverse polish notation. I mean compare the readability of Ruby vs Lisp: (taken from the article)
[1,2,3].map {|n| n*n }.reject {|n| n%3==1 }
to
(remove-if (lambda (n) (= (mod n 3) 1))
(mapcar (lambda (n) (* n n))
'(1 2 3)))
I barely know where to start with the lisp code, not to mention that the actual beginning data set (1 2 3) is lost in a sea of parantheses. I do believe that Lisp's AST form is extremely powerful but at an extreme cost to readability. (Unless you break up each block into a seperate function but then you start losing elegence) It's annoying to have to read the code bottom-to-top starting with the innermost list first. Why not just let the compiler/interperter maintain the AST and let us mere mortals work with a nice readable infix notation which easily expresses program flow?
I think that's why Lisp will eventually be functionality overcome by C-based/infix notation languages. The mainstream is taught infix notation (algebra) from a very early age.
Slashdot Mirror
I think it's more productive/useful to have a more systematic "version control system" on the high level congress bills (the policy) rather then trying to track the nitty-gritty details of what each department/federal agency does. (the mechinisms) Then it's more possible to track intent.
If one of the FAs violates the intent, it would easier to catch them as well as holding the policy makers more accountable.
Cheers
Ben
Thats brillent!
While I don't think that this would stop any log-rolling, at these it would give constitutants more ideas about how good/crappy their representative has been. Combine that with a lock of 24-48 hours on any bill, and you'd make congress alot more transparent.
I like it alot!
Ben
I was a little confused how to take screenshots and create a text file at first, but I checked the applications list and found applications called "Take Screenshot" and "Text Editor." Doesn't get much more obvious than that.
Someone has been taking their usuability studies seriously...
Funny how a little designing goes a long way...
Ben
As an addendum, as far as I know, the technians in charge of the systems on a US navy have to know the wiring diagrams of the systems they use in case there's a malfunction and a component has to be replaced. It's insane that they can't do the same with the software running on top of electrical systems, especially incase a software component needs to be replaced. I can easily imagine a driver or core kernel service needing to be modified because a sensor malfunctions or an acuator isn't performing as needed.
But money and influnce rules in the game of military purchances. I guess the ability to adapt to changing conditions doesn't matter until people start dying.
Ben
Windows is on 90+% of the world's computers, and absolutely every younger person knows how to navigate around in it.
You must be kidding. The UIs for CCC (command, control and communication) systems on a warship are above the basic windowing system. There's NO WAY a new recruit will know how the system works without at least several months/years of training. If he has to go to the hardware manager because a driver isnt working properly during combat, there is something seriously f--ked and he's probably already dead. These have to be kiosk systems where the operator only needs to work with a couple apps at a time. Unfortunately, since they have to use windows and the kernel/windowing system breaks down on them at sea because of a unforseen situation, they will be shit out of luck until they can get a professional rep to diagonios and rebuild their kernel. With an opensource kernel/windowing system, they could update and patch in real time.
I think that the M$ marketing/sales people are absolutely brillent. They can actually convince the British and US navies that it's better to be running winCE without allowing the people onboard assess to source code then just use opensource where the code is standardized and easy to understand and could be updated in realtime.
I think the chinese are going to win the next great war because they don't give in to that kind of bullshit.
Ben
Yeah, it seems like the better way to leave the asserts in the release code because users are more likely to stumble across bugs then the testers are.
Your "million-dollar" code is far too complex and massize to be fully tested by any normal size QA department or for all of the corner cases to be thought of by the programmers.
A better solution however is to use exceptions and exception-handling so the application can recover from the error by alerting the user, emailing an error report, and trying to rollback to a previous stable configuration.
However, way too many people bitch and whine about how much overhead exceptions take so expect to see more crashing asserts.
Cheers
Ben
The fact that the WHOLE system has to reboot if just one entertainment application goes down seems embressing/stupid to me.
Sounds like overly simplistic program management. They should have used a watchdog program to ensure that each process is up and running like normal and if they crash, they get restarted individually, not the entire system. that way, if you crash your console, you won't affect the experience of the other people around you.
Cheers
Ben
Besides, is oracle liable for your data loss when you lose your oracle instance? MS when you lose your IIS website?
No (because of the EULA) but they should be. Otherwise, why did you shell out the big bucks for Oracle or IIS? You might as well use apache.
Likewise, if you buy RHEL and apache falls over even after you installed it properly, Redhat should be liable.
but IANAL so this could be all pie in the sky when it comes to an actual lawsuit.
Cheers
Ben
"Security Researchers" are threatened with prosecution? Oh, mean hackers .
r _own_f.html) is threatened by the FBI
It doesn't help when normal researchers (see http://blog.wired.com/27bstroke6/2006/10/make_you
for pointing out that a flawed security mechinism is flawed. It wouldn't stop the real hackers from exploiting it and causing REAL damage and
the FBI did us a REAL diservice by threatening and scaring someone who could have helped make the system actually work properly.
So yess, the FBI has been lead-footed recently when dealing with "security researchers"
Ben
This meant that the robot could do a handful of simple, pre-defined motions, and the Wii-Mote was simply used to select the closest available match.
They could do better though.
They should record only very small motion paths (its accurate to 1/100 inch occuring TFA). Then have the robot play each piece when it receives it from the Wii-Remote. They'd have to record alot more motions but it would be smoother, more immediate, and more realistic in the end.
Cheers
Ben
Sun's handling of Java gave Microsoft enough time to make .NET a killer platform, especially for Web apps.
.NET is a bit behind the curve for net apps. its either ruby or python or PHP nowadays.
I hate to break it but
I supposed you could point out iron-python but there's noing different there from normal python development wise.
Cheers
Ben
When all the smoke settles, comptuters are for doing work.
;)
So what do you want to work on? Find something that itchs you hard and try to make it better.
Otherwise, I find firefox perfect for slashdotting all day and night....
Cheers
Ben
I've been wondering about how secure the repos are.
I mean, everyone seems to trust them 100%, they're not commercially/government supported, and once you get date-activated sploits in, you wouldn't be able to detect them until the activation time.
Is there an established method for making the repos more accountable? Maybe we need to buy insurance for our distros.
Cheers
Ben
Sure, it's possible to bury the key inside a dongle distributed with each software player.
However, it's annoying to work with and difficult to distribute
The video game industry and the mid-high end design industry tried dongles and both failed.
The only place I've seen it work was at a company which made textile design software
which was extremely specific and extremely expensive.
I seriously doubt any mass-media software will be tolerable with dongles.
Just my $0.02
Ben
Maybe the GP is refering to the fact that M$ is trying to compete with google in building $500 million dollar data centers to run thier live service. And what OS is M$ going to use to power this monster data center? It could be Linux, it could be Solaris. I seriously doubt it'll be server2003.
Cheers
Ben
I predict within 5 years, Vista will become a legacy product, and all new computers will be shipped with Microsoft's Windows API hosted on a Linux kernel.
I think my head exploded when you said that! Do you realize how completely different the windows and posix APIs are. M$ has developed completely different standards for filetree (C:,D:,E:,... vs mounts into a single file tree), using "\" instead of "/", allowing spaces in names, different threading model, different scheduler model, a very vertical integration of components (vs. linux's relatively broad layering of functionality) and other hopelessness. Add to the fact that M$ would have to pretty much rewrite ALL of their applications to run on a posix platform and I think the likely hood of this happening is ZERO.
However, there are several companies trying to make Linux more user friendly. Just look at Gnome and KDE.
Cheers
Ben
Ahh, thank you. I knew that trying to use momentum to calculate the speed was probably wrong.
btw 3500m/s == 7829.277mph. Pretty nice speed....
Cheers
Ben
What makes you think that? AFAIK fuse is written in C (as a driver inside the kernel) and I would expect the plugin file systems to be written in C/C++ (although I know there is ruby and probably python bindings) Why should fuse be any slower then kioslaves.
Cheers
Ben
The real trick is to keep the extraction software secret
Naa, you could just use a private key with a public encryption/decryption algorithm.
In fact, if you're clever enough, I sure you could build a watermark system which has a public key to verify that the watermark exists and is "correct" but only a private key would allow to you locate the watermark...
Not sure though,
Ben
Why don't you just subtract the watermark from itself?
How do you know that the watermark is in the color domain? More likely, it'll be in the frequency domain because it holds better when the file is analog copied. Plus, the watermark will be in a random place or somewhere determined by a encyption scheme. It could even be like using a public key! I seriously doubt that you could break a well designed watermark scheme. It is NOTHING like DRM.
Cheers
Ben
The article quotes one of the generals as saying the projectial is "like a ford taurus hitting the target at 380 mph" Ford taurus weigths 1383 kg (http://en.wikipedia.org/wiki/Ford_Taurus) and 380 mph -> 169 m/s
1383*169 => 234937.402 kgm/s. Projectial weighs 3.2 kg => it hits the target at 73417.9381m/s. This seems extremely high
I feel like I'm off by an order of magnitude. Guess it would be better to work off of the projection that the projectial can reach 95 miles into space to calculate initial speed.
I'm a huge kde/kioslaves fan too but IMHO fuse is really the way to go.
It'll allow all unix applications to access those interesting FSs in a common manner. The only problem is that the type (fish:,http:,ipod:,etc.) gets thrown out the window and mapping isn't as flexible as having a URI.
But it makes more sense in the long run...
Cheers
Ben
Umm, just found this: http://www.dwheeler.com/readable/
This might be interesting to check out. (except for using indentation as scope indicator. should be replaced with { } IMHO)
Cheers
Ben
Maybe because not everyone enjoys trying to think in reverse polish notation.
I mean compare the readability of Ruby vs Lisp: (taken from the article) to I barely know where to start with the lisp code, not to mention that the actual beginning data set (1 2 3) is lost in a sea of parantheses.
I do believe that Lisp's AST form is extremely powerful but at an extreme cost to readability. (Unless you break up each block into a seperate function but then you start losing elegence) It's annoying to have to read the code bottom-to-top starting with the innermost list first. Why not just let the compiler/interperter maintain the AST and let us mere mortals work with a nice readable infix notation which easily expresses program flow?
I think that's why Lisp will eventually be functionality overcome by C-based/infix notation languages. The mainstream is taught infix notation (algebra) from a very early age.
Cheers
Ben
By that point, it'll be too late and the M$ software will be already setup and installed.
;)
I wonder if M$ offers refunds?
Just remember kids, the first sample is always free. (or in this case, cheap)
Cheers
Ben