Occasionally I go into Worst Buy - usually to get a hard drive they have on sale or something like that.
Occasionally I will actually go over to peruse the CD aisle to see if there is anything I might like.
Now, ignoring the fact that I consider most of the music they carry to be crap, and that which I do not consider crap I probably already own, there is one other thing that really turns me off on my Worst Buy shopping experience.
THE DAMN RADIOS.
First they have the store "radio" turned up and playing crap music and advertisements for Worst Buy crap. It's hard enough to try to think of all the songs that I don't yet own copies of without having my auditory memory overwritten by the store's garbage.
Then there are all the ThunderThump3000's over in the Automotive radio section - you know the ones the kiddies all go over to, turn up the bass, find the crap radio station, turn up the volume, and then leave.
So then the store has to turn up the store radio.
And of course the computer department has to turn up all the computers.
And the TV department turns up all the TVs.
Yes, that really is an environment I want to linger in.
Don't get me wrong - I like loud music, IF IT IS MUSIC I LIKE. But the mishmash of crap in your average Worst Buy is not music.
Actually, the raw data you get from a CCD is better than the raw data you would get from the back of the human retina. The retinal is covered with blood vessels, has a big hole in it (the blind spot), had a great deal of noise (phosphene activity).
However, the nerves just after the retina, plus the optic nerve, plus the visual cortex, do a HELL of a lot of signal processing - removing the fixed imperfections like the blind spot and the blood vessels, using the dithering created by the small jittering of the eye to increase spatial resolution, averaging out the random phosphene activity.
IF you could get the same spatial resolution coupled into the retina, you could improve vision. However, that is a BIG IF - getting the millions of electrodes into the eye and coupled to the nerve cells, giving the correct voltage levels and firing patterns, without destroying the nerves by releasing metal ions or overvoltaging them, without provoking an immune response - quite a task.
Now, the question that I have is the plasticity of the brain - consider this: imagine the above difficulties are resolved. Now, instead of using a CCD array that approximates human normal vision by using RGB, what if you made an imaging element that generated RYGCBM - instead of three response curves you use six to increase the color-space resolution. Now, normally our brains learn to parse the basically RGBY data from the eye (the rods just return luminance data). Suddenly, the brain is getting a different set of signals. Is the adult brain plastic enough to learn to process this data at all? What about a child's brain?
A point that does not seem to have been made and should be.
Liver transplants don't have to come from a cadaver (or soon-to-be a cadaver). It is possible to take a section of a healthy person's liver and transplant it. The transplant will regenerate into a full liver, and the donor's liver will also regenerate (barring infection/drug abuse/drinking/rejection).
So unlike a heart transplant, you can create media attention, locate a potential transplant donor, and get them to donate to you while they are still alive - in that way a liver is like a bone marrow transplant.
So by creating a lot of media attention he got people to consider being tested and donating that would not have done so otherwise - he did not "cut in line" ahead of somebody else. Indeed, it is possible that, due to his actions, somebody else who needed a liver might have found a match.
Various docs who frequent/. (HEY, TYRO!) can back me up on this.
Funny, but one handed keyboards have been around since the Englebart demo.
Except for CAD, they never really took off - until the modern video game.
And while I certainly would not want to type a comment like this with a one-handed keyboard, I can see where they would be damn useful in editing a document - click-drag, button press for bold, click-drag, underline (or click-drag indent, click-drag create-subroutine-skeleton, click-drag lookup-definition).
Now that the Olympic committy are going after anybody who uses a circular motif in their logo, or any derivitive of the word "Olympia", now that we are starting to allow things like "The Dream Team" into the Olympics, the spirit of human excellence, *sportsmanship*, and competition are gone.
I don't give a pair of fetid dingo's kidneys about the Olympics - I won't watch them, I don't care about them, and I wish to show my disregard - please let this become a topic.
As for the security issues - shows of "security" are going to become the norm, get used to it.
Or better still, DON'T get used to it - don't go to these places, write your government appointed representatives (I chose that wording deliberately, think about it) to complain.
Personally, I think the best security would be to enable and encourage THE PEOPLE to take part in their own defense.
Consider the window film that reduces the amount of light passing through a window, but allows for a clear image through the glass. That is "transparent", even though it is not passing all the light.
"Translucent" means that while some qunatity of light is allowed to pass, no meaningful image passes.
So it is possible to be both transparent and yet block some of the light - and for an example look no furthur than your sunglasses.
You can also implement FFTs in hardware, or use a different approach - use a more "analog-y" method like mix&filter, which allows you to run a seperate downconverter for each carrier.
As for the HW - what kind of development are you doing? What's your price range for a devel board? - are you doing this as a hobbist or professionally? If you are looking in the professional range you could get a Pentec board or an Aeroflex PXI board.
They may not be using DSPs as much as FPGAs/ASICs - a great deal of the signal processing for that sort of thing is easier done as parallel blocks of hardware than software.
The FCC is no more worried about you mucking around in the modulator/demdoulator as the driver - either will allow you to cause interference.
(A guy who does Software designed radio for a living.)
Gosh, I even went so far as to state, very clearly, that a key part of defeating this sort of thing was, and I quote my previous post, "block any client that makes more than a few requests a second."
Now, lets see. DNS packets are limited to 512 bytes per spec (and having just finished implementing an DNS client the spec is quite clear in my mind). Let us take the commonly used value of "less than ten" for the term "a few".
So, you have less than 5120 bytes/second of throughput. That's assuming that you don't flag a host that makes that many queries a second for more than a second as suspicious - and I would consider a host making more than 15 different domain lookups in as many seconds as VERY suspicious!
And for this to work you have to have a registered domain, with a registered server you control, in order to serve up your packets - remember, you cannot just fire off packets to J. Random Server - you MUST go through the proxying server, and so you MUST use a valid, registered domain.
Now, what if, instead of all that pain, I just set up a freaking HTTPS server (or even more simply set up an SSH server listening on the HTTPS port - both are SSL, and unless the proxying system can man-in-the-middle the system all the firewall can say is that an SSL connection was established.) Now I really DON'T need a domain of my own, just a server.
In short, yes, this is a threat to your network security. So is a guy with a laptop and a box of donuts ("Hey, you want these Krispy Kreames? Let me plug in for a few minutes...").
Given the set of tools needed to pull this off there are more effective ways to achive the same goal.
That is why any GOOD sysadmin will set up the system so that there is a single DNS server for the plant, and that server and that server alone is allowed to send and receive DNS packets to the greater Internet - all other machines are to use the local DNS server.
Not only does this GREATLY reduce the amount of DNS traffic a shop produces (by caching all requests locally) it helps prevent this sort of foolishness by requiring all packets to be well formed DNS packets - else the server drops them.
Then, you can block any client that makes more than a few requests a second.
Yes, it is easier to set up a firewall to be very porous to outbound traffic, but it is more secure to deny all direct access, and force clients to run through proxies for the various services.
I am all in favor of making Gnome newbie friendly - so long as it does not exclude us non-newbies.
Consider cars as an analogy:
First of all, there are many different models of car - this would be analogous to Gnome/Enlightenment/KDE/Windows/MacOS/*. Few sensible people would assert that we should all be driving Geo Metros or all be driving Grand Marquis or Peterbuilt trucks.
But even within a make of cars, there are degrees of complexity. Most people driving an automatic transmission vehicle use P, R, and D. Those other settings (N, 2, 1) are just needless complexity, right - shouldn't we just remove them? Nobody uses them, right? Now, go for a drive in the mountains. Sure, many people only use D - you can tell them by smelling for burned-up brake pads. Better drives use 2 and 1, and not their brakes - they NEED 2 and 1. And people towing a car need N.
My car has buttons for moving the pedals forward and back. The first thing I did when I took delivery was to run the pedals all the way down, being 193cm tall. Does that mean that NOBODY needs to adjust the pedals up, so we should remove that switch? Or what about the traction control off switch?
My point is that while Granny Fanny may never use those features, some of us will - SO LEAVE THEM IN YOU BASTARDS!
Put an "Expert mode" in. Default it to OFF. Let me turn it on. Let me configure whether I feel spatial navigation is right for me or not. Let ME determine what programs play MP3s if I choose to do so.
And don't treat novice users like read-only dummies - let them know there is more power available to them, should they be interested in learning about it.
There is a GREAT difference between "ignorant" (unlearned) and "stupid" (unable to learn) - and many newbies are the former, not the latter. Don't treat them (and us) as stupid.
Come on Simon! An anonymous user submits a story about how strange postings on the 'Net may be part of an ad blitz for a game, and you post it.
Hmmm. I wonder. Could the submitter of the story perhaps be employed by the media company driving the blitz? Could the submission to Slashdot be an attempt to drum up even more buzz?
Nahhh. I must be paranoid.
Come on guys - at least make the bastards PAY for space on Slashdot!
I find the statements about "we can go back to old pictures of JFK and see what he was looking at" to be questionable at best.
You need a LOT of pixels of the eye itself from which to reconstruct an image. Now, look at how much of a given normal picture the eyes of a person represent.
You *might* be able to reconstruct where the person is looking. You probably aren't going to have enough pixels to reconstruct what they saw.
To do that level of imaging you are going to need a picture of the person's eye at high resolution.
So the government spy cameras will have to zoom in on your eyes - call it about a 500 to one zoom. They will have to track your eyes as you move about.
And yes, if you wear sunglasses you can defeat this.
Now, what this WOULD be very useful for would be in combinatino with a head mounted display - since the display device has to subtend a large angle as viewed from the eye, the display device must have a good view of the eye. So combining the display device with an imaging device would allow the system to see what you at what you are looking, so you now have a pointing device. Theoretically, a wink or slow-blink could be a "select" operation.
Now, if they could get the focus point of the eye, they could REALLY make an interesting system - if you are focusing past the image, they could mute it - reduce the brightness, possibly even reduce the amount of information (iconify apps, reduce update rates, show only "critical" items, etc.) When they detect you've shifted focus to bring the display into focus, brighten up. Think of looking through a dirty windshield, then shifting focus to the dirt on the glass.
Re:Better command completion from history
on
Bash 3.0 Released
·
· Score: 1
Typing "foo" is a lot easier than CTRL-R - one keystroke rather than a combinded stroke.
Typing "foo<UP><UP>" is a lot faster when I have multiple commands in the history, as well.
Better command completion from history
on
Bash 3.0 Released
·
· Score: 2, Insightful
What I would like to see Bash do is command completion a la 4DOS:
If I have a command foo bar baz
in my history, and I type foo<TAB>
It complete that with the most recent command starting with "foo", and if I type foo<UP> it will cycle through the commands in history that start with "foo".
Perhaps it is because of how much I've neutered Javascript on my copy of Mozilla, but I cannot meaningfully take the test - what ever mechanism they are using to allow the "hover over the link to see the link" doesn't work, so I cannot check the link.
Of course, they also don't show you the full message headers, and the messages are shown as HTML messages - something I also have turned off.
So most of the first cut tests I use to check a message are disallowed - this would be like taking a test on electromagnetic theory without being allowed to use math symbols.
Does anybody else get the feeling that the writers from Star Dreck:Voyager have moved into the particle physics business?
This just reeks of the "Particle of the Week" writing that ST:V indulged in so frequently.
What next - the hypothesis that the universe will undergo a "Big Rip", but then the interaction of the accelerons and the whetions will reset the timeline and everything will be back to normal?
Slashdot Mirror
The blub linked says "Available free to all Linspire users, ..." but says nothing about non-Linspire users.
Question: is this available for non-Linspire machines?
The code link on the Linspire page seems broken - the correct one is here
Occasionally I go into Worst Buy - usually to get a hard drive they have on sale or something like that.
Occasionally I will actually go over to peruse the CD aisle to see if there is anything I might like.
Now, ignoring the fact that I consider most of the music they carry to be crap, and that which I do not consider crap I probably already own, there is one other thing that really turns me off on my Worst Buy shopping experience.
THE DAMN RADIOS.
First they have the store "radio" turned up and playing crap music and advertisements for Worst Buy crap. It's hard enough to try to think of all the songs that I don't yet own copies of without having my auditory memory overwritten by the store's garbage.
Then there are all the ThunderThump3000's over in the Automotive radio section - you know the ones the kiddies all go over to, turn up the bass, find the crap radio station, turn up the volume, and then leave.
So then the store has to turn up the store radio.
And of course the computer department has to turn up all the computers.
And the TV department turns up all the TVs.
Yes, that really is an environment I want to linger in.
Don't get me wrong - I like loud music, IF IT IS MUSIC I LIKE. But the mishmash of crap in your average Worst Buy is not music.
First - the CCD is NOT in the eye - do read the article.
Second - making a CCD with the varying resolution of the retina is not a difficult task.
Third - I don't know about YOUR camera, but MY CCD camera will do quite nicely at night without a light - you simply increase the integration time.
Actually, the raw data you get from a CCD is better than the raw data you would get from the back of the human retina. The retinal is covered with blood vessels, has a big hole in it (the blind spot), had a great deal of noise (phosphene activity).
However, the nerves just after the retina, plus the optic nerve, plus the visual cortex, do a HELL of a lot of signal processing - removing the fixed imperfections like the blind spot and the blood vessels, using the dithering created by the small jittering of the eye to increase spatial resolution, averaging out the random phosphene activity.
IF you could get the same spatial resolution coupled into the retina, you could improve vision. However, that is a BIG IF - getting the millions of electrodes into the eye and coupled to the nerve cells, giving the correct voltage levels and firing patterns, without destroying the nerves by releasing metal ions or overvoltaging them, without provoking an immune response - quite a task.
Now, the question that I have is the plasticity of the brain - consider this: imagine the above difficulties are resolved. Now, instead of using a CCD array that approximates human normal vision by using RGB, what if you made an imaging element that generated RYGCBM - instead of three response curves you use six to increase the color-space resolution. Now, normally our brains learn to parse the basically RGBY data from the eye (the rods just return luminance data). Suddenly, the brain is getting a different set of signals. Is the adult brain plastic enough to learn to process this data at all? What about a child's brain?
This can fly for 3 minutes and can return video images.
Consider flying this (covertly) into a hostage situation, then shutting down the motor - how long could it return video then?
Or corporate espionage - fly this between the drop ceiling and the real ceiling, land over the boardroom.
Oh hell yes, I can see a lot of uses right now for this.
A black hole the mass of the moon would not be stable - it would radiate its mass away as Hawking radiation.
Smaller black holes would radiate away faster.
Therefor, according to current theory, it would not be possible to create a black hole that would eat the moon.
A point that does not seem to have been made and should be.
/. (HEY, TYRO!) can back me up on this.
Liver transplants don't have to come from a cadaver (or soon-to-be a cadaver). It is possible to take a section of a healthy person's liver and transplant it. The transplant will regenerate into a full liver, and the donor's liver will also regenerate (barring infection/drug abuse/drinking/rejection).
So unlike a heart transplant, you can create media attention, locate a potential transplant donor, and get them to donate to you while they are still alive - in that way a liver is like a bone marrow transplant.
So by creating a lot of media attention he got people to consider being tested and donating that would not have done so otherwise - he did not "cut in line" ahead of somebody else. Indeed, it is possible that, due to his actions, somebody else who needed a liver might have found a match.
Various docs who frequent
Big deal. Doctor Appleby and Larry, Moe and Curly did this back in 1962.
OK, so the schools don't want Eddie Eagle to come to school and tell kids not to play with guns.
But schools are willing to let $name the ferret come to school to tell kids that they mustn't ever copy anything.
Blech.
Wait a minute - don't eagles eat ferrets?
Maybe what we need to do is get Eddie and Tux to show up at the same time as the ferret....
Funny, but one handed keyboards have been around since the Englebart demo.
Except for CAD, they never really took off - until the modern video game.
And while I certainly would not want to type a comment like this with a one-handed keyboard, I can see where they would be damn useful in editing a document - click-drag, button press for bold, click-drag, underline (or click-drag indent, click-drag create-subroutine-skeleton, click-drag lookup-definition).
Can we get an Olympic subject on /.?
That way, I can flip the ignore bit on it.
Now that the Olympic committy are going after anybody who uses a circular motif in their logo, or any derivitive of the word "Olympia", now that we are starting to allow things like "The Dream Team" into the Olympics, the spirit of human excellence, *sportsmanship*, and competition are gone.
I don't give a pair of fetid dingo's kidneys about the Olympics - I won't watch them, I don't care about them, and I wish to show my disregard - please let this become a topic.
As for the security issues - shows of "security" are going to become the norm, get used to it.
Or better still, DON'T get used to it - don't go to these places, write your government appointed representatives (I chose that wording deliberately, think about it) to complain.
Personally, I think the best security would be to enable and encourage THE PEOPLE to take part in their own defense.
Consider the window film that reduces the amount of light passing through a window, but allows for a clear image through the glass. That is "transparent", even though it is not passing all the light.
"Translucent" means that while some qunatity of light is allowed to pass, no meaningful image passes.
So it is possible to be both transparent and yet block some of the light - and for an example look no furthur than your sunglasses.
You can also implement FFTs in hardware, or use a different approach - use a more "analog-y" method like mix&filter, which allows you to run a seperate downconverter for each carrier.
As for the HW - what kind of development are you doing? What's your price range for a devel board? - are you doing this as a hobbist or professionally? If you are looking in the professional range you could get a Pentec board or an Aeroflex PXI board.
They may not be using DSPs as much as FPGAs/ASICs - a great deal of the signal processing for that sort of thing is easier done as parallel blocks of hardware than software.
The FCC is no more worried about you mucking around in the modulator/demdoulator as the driver - either will allow you to cause interference.
(A guy who does Software designed radio for a living.)
Yawn. Wake me when Intel has released real, production ready (NOT 0.2) drivers for Linux for this, or any other modern wireless network chip.
Gosh, I even went so far as to state, very clearly, that a key part of defeating this sort of thing was, and I quote my previous post, "block any client that makes more than a few requests a second."
Now, lets see. DNS packets are limited to 512 bytes per spec (and having just finished implementing an DNS client the spec is quite clear in my mind). Let us take the commonly used value of "less than ten" for the term "a few".
So, you have less than 5120 bytes/second of throughput. That's assuming that you don't flag a host that makes that many queries a second for more than a second as suspicious - and I would consider a host making more than 15 different domain lookups in as many seconds as VERY suspicious!
And for this to work you have to have a registered domain, with a registered server you control, in order to serve up your packets - remember, you cannot just fire off packets to J. Random Server - you MUST go through the proxying server, and so you MUST use a valid, registered domain.
Now, what if, instead of all that pain, I just set up a freaking HTTPS server (or even more simply set up an SSH server listening on the HTTPS port - both are SSL, and unless the proxying system can man-in-the-middle the system all the firewall can say is that an SSL connection was established.) Now I really DON'T need a domain of my own, just a server.
In short, yes, this is a threat to your network security. So is a guy with a laptop and a box of donuts ("Hey, you want these Krispy Kreames? Let me plug in for a few minutes...").
Given the set of tools needed to pull this off there are more effective ways to achive the same goal.
That is why any GOOD sysadmin will set up the system so that there is a single DNS server for the plant, and that server and that server alone is allowed to send and receive DNS packets to the greater Internet - all other machines are to use the local DNS server.
Not only does this GREATLY reduce the amount of DNS traffic a shop produces (by caching all requests locally) it helps prevent this sort of foolishness by requiring all packets to be well formed DNS packets - else the server drops them.
Then, you can block any client that makes more than a few requests a second.
Yes, it is easier to set up a firewall to be very porous to outbound traffic, but it is more secure to deny all direct access, and force clients to run through proxies for the various services.
I am all in favor of making Gnome newbie friendly - so long as it does not exclude us non-newbies.
Consider cars as an analogy:
First of all, there are many different models of car - this would be analogous to Gnome/Enlightenment/KDE/Windows/MacOS/*. Few sensible people would assert that we should all be driving Geo Metros or all be driving Grand Marquis or Peterbuilt trucks.
But even within a make of cars, there are degrees of complexity. Most people driving an automatic transmission vehicle use P, R, and D. Those other settings (N, 2, 1) are just needless complexity, right - shouldn't we just remove them? Nobody uses them, right? Now, go for a drive in the mountains. Sure, many people only use D - you can tell them by smelling for burned-up brake pads. Better drives use 2 and 1, and not their brakes - they NEED 2 and 1. And people towing a car need N.
My car has buttons for moving the pedals forward and back. The first thing I did when I took delivery was to run the pedals all the way down, being 193cm tall. Does that mean that NOBODY needs to adjust the pedals up, so we should remove that switch? Or what about the traction control off switch?
My point is that while Granny Fanny may never use those features, some of us will - SO LEAVE THEM IN YOU BASTARDS!
Put an "Expert mode" in. Default it to OFF. Let me turn it on. Let me configure whether I feel spatial navigation is right for me or not. Let ME determine what programs play MP3s if I choose to do so.
And don't treat novice users like read-only dummies - let them know there is more power available to them, should they be interested in learning about it.
There is a GREAT difference between "ignorant" (unlearned) and "stupid" (unable to learn) - and many newbies are the former, not the latter. Don't treat them (and us) as stupid.
Yet Another Web Based Advertising Blitz.
Come on Simon! An anonymous user submits a story about how strange postings on the 'Net may be part of an ad blitz for a game, and you post it.
Hmmm. I wonder. Could the submitter of the story perhaps be employed by the media company driving the blitz? Could the submission to Slashdot be an attempt to drum up even more buzz?
Nahhh. I must be paranoid.
Come on guys - at least make the bastards PAY for space on Slashdot!
I find the statements about "we can go back to old pictures of JFK and see what he was looking at" to be questionable at best.
You need a LOT of pixels of the eye itself from which to reconstruct an image. Now, look at how much of a given normal picture the eyes of a person represent.
You *might* be able to reconstruct where the person is looking. You probably aren't going to have enough pixels to reconstruct what they saw.
To do that level of imaging you are going to need a picture of the person's eye at high resolution.
So the government spy cameras will have to zoom in on your eyes - call it about a 500 to one zoom. They will have to track your eyes as you move about.
And yes, if you wear sunglasses you can defeat this.
Now, what this WOULD be very useful for would be in combinatino with a head mounted display - since the display device has to subtend a large angle as viewed from the eye, the display device must have a good view of the eye. So combining the display device with an imaging device would allow the system to see what you at what you are looking, so you now have a pointing device. Theoretically, a wink or slow-blink could be a "select" operation.
Now, if they could get the focus point of the eye, they could REALLY make an interesting system - if you are focusing past the image, they could mute it - reduce the brightness, possibly even reduce the amount of information (iconify apps, reduce update rates, show only "critical" items, etc.) When they detect you've shifted focus to bring the display into focus, brighten up. Think of looking through a dirty windshield, then shifting focus to the dirt on the glass.
Typing "foo" is a lot easier than CTRL-R - one keystroke rather than a combinded stroke.
Typing "foo<UP><UP>" is a lot faster when I have multiple commands in the history, as well.
What I would like to see Bash do is command completion a la 4DOS:
If I have a command
foo bar baz
in my history, and I type
foo<TAB>
It complete that with the most recent command starting with "foo", and if I type
foo<UP>
it will cycle through the commands in history that start with "foo".
Perhaps it is because of how much I've neutered Javascript on my copy of Mozilla, but I cannot meaningfully take the test - what ever mechanism they are using to allow the "hover over the link to see the link" doesn't work, so I cannot check the link.
Of course, they also don't show you the full message headers, and the messages are shown as HTML messages - something I also have turned off.
So most of the first cut tests I use to check a message are disallowed - this would be like taking a test on electromagnetic theory without being allowed to use math symbols.
Does anybody else get the feeling that the writers from Star Dreck:Voyager have moved into the particle physics business?
This just reeks of the "Particle of the Week" writing that ST:V indulged in so frequently.
What next - the hypothesis that the universe will undergo a "Big Rip", but then the interaction of the accelerons and the whetions will reset the timeline and everything will be back to normal?
OK Ms. Portman, stand very still right here, the scanning process will only take a minute...