When I look at the image at the head of the linked page ("Disney.jpg" - curiously) what I see is a VERY low resolution image.
Look at the red trail behind it. There are a bunch of little raster-aligned four-pointed star shapes. (The one on the extreme left is a prime example).
This is what you get if you take a VERY low resolution image an blow it up with simple bilinear blending between the pixels. Taking this as evidence of the original image resolution, we can see that the 'spaceship' at the righthand end of the image is just about 3 pixels across - but has been false-coloured so that the bilinear blending has become magenta and yellow bands. Those are not 'real' they are just a part of the false-colouring.
Isn't it suspicious that the "UFO" is exactly aligned with the raster?
This is a fake...well, perhaps not exactly a fake - but an intentional mis-use of image manipulation to produce an image that was never really there.
You could reproduce this image in GIMP in about 3 minutes flat.
1) Create a 20x20 RGB image. 2) Using a 1 pixel brush, paint a diagonal line using bright red. 3) Fatten one end of the line slightly.
At this point, your image (if you'd gotten it from a photo of the night sky) wouldn't convince you that this was a UFO - would it? It could be any kind of a trail, meteor, military jet on afterburner, a flare, a firework, anything like that.
4) Increase the image resolution to 400x400
Notice how the 'tail' now looks EXACTLY like the one in the ufomag web site. Look at the 'star' shapes in the tail.
So, now let's do some "false-colour enhancement":
5) Choose 'select by colour' - set the threshold down to nearly
zero percent and click on a region at the center of the 'head'
of the trail. Fill it with magenta.
6) Pick a pixel close to that, fill it with a nice lemon yellow.
Notice how your image looks startlingly similar to the one on the ufomag website. All the artifacts present in their image are present in yours.
Now, I'm not saying that they painted their image in GIMP, I'm quite prepared to accept that it's a photo of a real world night-sky object. However, the pretty pink and yellow spaceship on the right - complete with spooky red glow and engine exhaust is no more than a deliberately produced artifact.
The yellow and pink regions are BOTH narrower than the original pixel resolution - no feature narrower than TWO pixels wide (Nyquist sampling limit) can ever be reconstructed from an image.
I emailed Cmdr Taco to suggest that he re-visit the idea of a META tag that says "Slashdot: Please Mirror Me" (and yes - I've now read the FAQ which contains a response to my META suggestion).
Rob Malda said:
> There are already tags in place to do exactly what you want. Research > http authentication. There are many ways to prevent people you don't > know from reading your site.
So his solution is that I should shut everyone I don't know out of my web site?!?
Well, *THAT* helps a lot!!
Sheesh!
The FAQ basically claims that the META solution won't work because not enough people use it.
1) If the story submission entry page explained the META mechanism
then people who submitted stories about their own sites could
defend themselves in advance.
2) If the editors were sensitive to the issue they could use their
good judgement to delay posting the story until the owner of the
site has had a chance to defend themselves with the META tag.
They'd only have to do that when the story is not hot breaking
news - and when it's a link to a small company or individual.
No need to mirror CNN, BBC, etc.
3) Making a big deal of the new META feature on Slashdot would get
the news out to a large percentage of the people who build
digital door knobs, beowulf clusters of TRS-80's and backyard
rollercoasters.
It's time for Slashdot to step up to the plate and not inflict DDOS attacks on the very people who provide the content that they report on.
Slashdot really, truly, utterly needs to have a local cache of the pages it references. It's getting to where Slashdotting is as bad as a denial of service attack - and that's a terrible thing to inflict on *anyone*.
Probably 50% of web sites referenced from main news items are down within an hour of Slashdot mentioning them - and they stay down until a couple of days have passed. That sucks.
They could easily implement some kind of opt-in thing where you put a META tag in your web page telling Slashdot that you grant them explicit permission to mirror the site for (say) a week after mentioning it - so Slashdot would have no legal/copyright come-backs. At the end of the week the Slashdot mirror could revert to become a redirect to the real site so you don't have problems with people bookmarking the Slashdot cache instead of the real site.
The whole process could be automated.
People who do cool things like this door lock would surely be aware that they could get Slashdotted and prepare for the event in advance by inserting the tag - and private individuals are the people who are most likely to have their server die.
Companies that want to profit from their slashdotting by advertising from their page or taking orders off of it could just leave off the META tag and handle the traffic as now.
An opt-in cache mechanism is a win-win-win solution. Slashdot wins because more people will use the service if it doesn't continually refer to dead sites. Readers will win because less sites will be dead-on-arrival - and web site operators will win (if they want to) by not having their site die from Slashdotting.
The idea is that when you create your web site - you enter the Lat/Long as a Meta tag. What you enter is entirely up to you.
Ideally, you shouldn't enter the location of the server - or of your home - you should enter the Lat/Long of the area you would LIKE people to find you hanging out at.
So - You want lots of emails from people interested in Outer Mongolia? Just add Meta tags the that lat long. You have a web site about the Eiffel tower - add Meta tags for that part of Paris. Each URL can have different tags - so you can be in many places at once.
It doesn't threaten anonymity - if you want to stay hidden - don't add the tags. If you want to lie about where you are - fine. If you want your GPS to tell your PDA to email your server to tell it to update your home page every 10 minutes - also fine...(although you'll have to ping the geosite server to tell it to update your URL in it's database).
It seems to be a good idea. Since the information is in your web page, any search engine can take advantage of this...all it takes is to define a standard lat/long tag.
IMHO, they should have included altitude and an error metric in the tag.
> Say it costs MS $250 to produce an XBox. They then sell it for $200. > If you buy it, MS loses $50. If you *don't* buy it, MS loses $250.
Stoopid. That presumes that they made a bazillion X boxen and then won't ever make any more...so if I don't buy it, they toss it into the nearest dumpster. No way!
If you buy it, they lose $50 - if you don't buy it, they sell it to someone else - and make one fewer next time around...they lose *nothing*.
The real deal is the question of how game makers pay M$. Generally, they'll be paying a per-game-sold royalty. M$ rely on the royalties from the games you buy to cover the $50 (or whatever it is) that they lost on the console sale.
So - buy an X-box and buy no games at all - M$ loses.
Buy an X-box and buy (say) 10 games - M$ breaks even.
You could argue that games won't be made for the X-box *at all* unless a certain number of them have been sold - and that's certainly true. It's not worth the game developers making games if nobody's going to buy them - but X-box is past that threshold. There are plenty of owners out there buying games to guarantee half a dozen new games every month.
Would it be more devastating to buy a Game Cube? Well, only if you are going to buy enough games for it for Nintendo to turn a profit.
I've heard that the GameCube sells at a break-even price - or perhaps a small profit - .
I absolutely agree that teachers are woefully underpaid - and in consequence, many are under-qualified. However, this teacher would have been 100% better at her job with even ONE DAY of introductory computer training. What's worse is her unwillingness to be taught - either by the kids (who know more than she does) - or by parents like myself who try to explain where she's factually incorrect.
There seems to be an attitude at the school that so long as she maintains classroom discipline, the edutainment software will somehow teach the kids what they need to know. That's a joke - but it's certainly cheap compared to hiring a computer-literate teacher.
It's certainly evident that teachers need to be paid whatever the going rate for professionals - but teachers like my son's computer class teacher wouldn't see the benefit of such an increase because the first thing it would do would be to put her out of a job as she'd be replaced by someone who had some real computing experience.
Actually, I think almost anyone who's noodled around with PC's for a few months could do this job better. It doesn't take an IT professional - it just takes someone who is enthusiastic about PC's and has played with them extensively.
I'd like to see local computer-based companies have a policy of having each staff member spend a couple of days each year helping out in computing classrooms in their local area - paid for by the company. That would have long-term payback for everyone involved.
My kid's school (a 5th/6th grade intermediate school) has a beautiful, fully equipped computer classroom - and a teacher who teaches computing only....and that's the problem. The teacher knows *nothing* about computers. Practically all the kids know far more than she does.
Because she knows nothing, she dumps 'edutainment' programs onto the machines and has the kids play them continually while she merely maintains classroom discipline.
She spent three weeks (that's 40 minutes per lesson for 10 lessons) having the kids run some kind of 'typing tutor' program. Since all the kids learned to type in 3rd grade (at least as well as a typing tutor program *can* teach), they were all bored to tears with the repetitive exercises.
Fortunately, my son discovered that this stoopid program doesn't disable cut and paste - so he was able to complete all the exercises insanely quickly. Since the teacher allows them to surf the web once they have finished the assignments, he was able to go off and have fun by himself the entire time.
The crowning glory came at the end of the year when the teachers were handing out class prizes - my son was awarded the prize for best EVER score on the typing tutor by the dump computer science teacher - she proudly announced that he'd scored something like 3,000 words per minute with a 0% error rate. Some of the other teachers looked a bit strangely at her - clearly realising that something had gone amok, but perhaps assuming she'd just mis-spoken the results.
This is just one of many gaffes this teacher made. She had the kids "List 10 parts of the Computer". My kid duly wrote stuff like 'CPU', 'ROM', 'Ethernet Ataptor', 'Motherboard' - and the teacher gave him zero on the "test" saying that the correct answer was 'Mouse', 'Keyboard', 'Television' (!), 'Mouse pad', etc. When my kid complained that his computer at home didn't have a mouse pad she told him that this was nonsense and that ALL computers have mouse pads - this dissuaded him from telling her that the monitor is not, in fact, a TV set.
Similarly, she had the kids write down the 10 good things and 10 bad things about computers. My son complained that he couldn't think of 10 bad things. His teacher gave as an example: "They crash a lot" - well, since we only run Linux at home, my son knows that this isn't necessarily true and that it's not the COMPUTER that crashes - it's the SOFTWARE. Inevitably, when he complained he got in trouble.
I've written several letters to the teacher in question (she doesn't appear to read her email - even though it's provided by the school) - with poor results. I wrote and even visited with the Principal to try to get something done - but of course she just says that qualified staff are hard to get - and the State doesn't require that teachers are trained in the subject they are teaching.
So, can we conclude that teaching with computers is "A Bad Thing" ?
No!
Not unless we've carefully checked that the teachers and curriculum are sensibly chosen. Clearly, if my son's school had spent the money that went into that computer lab in some other way, they'd have gotten more value for money and the kid's grades would have been better...but that doesn't prove that teaching computers are bad - just that they are ineptly managed.
Here in Texas, police stop people for speeding because they are funded in part from the fines they collect. It is very common to see the speed limit drop by 5mph when you cross the border from one county to another despite no other differences in road conditions. Why? Because people don't change speed as they cross the boundary - and the cops can get them on a more serious charge and raise more money.
...well, at least mine does. I have a MINI Cooper S (the new model from BMW).
When we did the 'walk through' of the controls, the dealer told me, that when you turn on the ignition, it displays the ESTIMATED number of miles to your next service. I asked the guy - "What do you mean 'estimated'? Surely it just counts down the miles? Since the odometer and the miles-to-next-service indicators are both digital, why would they ever disagree?"
Aparrently not. It monitors how 'agressively' you drive and counts the miles down faster if you redline it a lot (very tempting with the MINI BTW). This makes sense - a car that's driven hard needs servicing more often. The onboard computer knows the RPM - the number of times the traction control and dynamic steering controls kick in and everything else there is to know about how hard the poor beast is being thrashed - it's in a good position to know when a service is likely to be needed.
Does this happen in practice? Yes!
When I took delivery, there was 20 miles on the odometer and since the first service is nominally at 10,000 miles, the miles-to-next-service indicator was reading 9980 as you'd expect.
After I'd clocked up ~500 miles, driving it fairly agressively (because it's my new toy) the service indicator was saying ~9440 to go - suggesting it needed servicing 60 miles before it 'nominally' should. In the past few weeks, my driving style has returned somewhat to 'normal' and when I hit 1000 miles, the service indicator was showing ~8930 to go - so my better driving style had only cost me an additional 10 miles of 'penalty'.
So - there is no doubt that the car monitors my driving style and makes that readily apparent to the dealership - requiring me to undergo more frequent services in order to stay withing warranty if I drive agressively - and rewarding me with fewer services if I'm a good person.
Whether that's a good thing or not depends on your perspective. Yes, it's a slight invasion of privacy because your car dealer now knows you are a bad person. But if you intend to keep the car beyond the end of the warranty then you are better off for knowing that you need to service it more often in order to avoid it crapping out on you. I guess it also allows the manufacturers to set the service intervals nominally further apart - so they don't penalise good drivers by requiring more frequent services.
Even a Z80 had a 4MHz clock - if you couldn't time the engine firing to within a millisecond or so, you'd better get a better programmer! I programmed telephone exchange software on Z80's - and we did tons of stuff with millisecond precision. A 200MHz x86 would be overkill in a missile.
Does the engine firing have to be more accurate than a millisecond? I doubt very much that the engine could respond reliably at higher rates than that - it's a big mechanical/chemical contraption and those things don't move that fast.
Don't forget, you aren't talking about running some big complex operating system here. The time-critical code can be in assembler - and there won't be an OS running in there.
The only time I could imagine needing more CPU horsepower than a Z80 in a missile would be if it had some kind of a camera or radar image to analyze as a part of targetting.
Hmmm - he promised his wife he wouldn't email porn...now I wonder if someone could get him in deep trouble with his missus by faking porn from HIS address - now THAT would be punishment enough:-)
I was always impressed by Steve Job's comment to the guy who was writing the Mac bootstrap code. The guy was complaining that it wasn't worth optimising the bootstrap loader any more because it was fast enough already. I don't Steve's exact words - but it was something like: We will sell 100 million of these machines - if each of those people boots their machine once a day for five years - then that's 15 billion reboots. If you can save just one second from the reboot time of the Mac then that's 480 YEARS saved.
So shaving one second of the boot time is like saving the lives of 50 people. What could be a more noble activity than saving human lives like this.
So - applying that math to this spammer: If he sends out ten million spams a day and it takes 1 second to delete each one - and if this guy does that every day for five years - then that's morally equivelent to murdering 50 people.
Just because the damage he does to each individual is small, the cumulative damage is huge.
There is another story (probably apochryphal) about the guy writing the banking system software who changed the code to take the roundoff error (less than a half cent) from every interest calculation and direct it into his personal account. The story goes that he made tens of thousands of dollars a week. This story probably isn't true - but should such a person be considered any less a criminal because the money he stole was spread so thinly? Obviously not - he stole those thousands of dollars and that's that.
This spammer deprived the people of the world of 50 human lives - he should be considered a mass murderer and treated accordingly.
Dunno where you live - but here in sunny Texas all the toy stores carry the bulk Lego buckets. They cost around $10 and contain around 200 parts. Some of them are *mildly* themed - so the StarWars bucket has some interesting things like radar dishes and fins and stuff - but it's still basically bulk Lego.
For younger kids you can also get bulk Duplo - which is twice-normal-size Lego that is reverse-compatible with 'real Lego' but is less of a choking hazard. Because Duplo works with Lego, they can still use it when they grow up enough to have the real thing.
Don't be tempted into buying any of the 'clone' brands of brick - only the genuine Lego parts are made of the softer plastic that enables the bricks to be re-used indefinitely. The cheaper brands are made with hard polystyrene and the wear out extremely quickly. Build about three things with the same set of bricks and you'll start to notice that they don't hold together anymore.
So - get genuine Lego bulk buckets. My son has about a dozen of them - and it's never enough.
The anti-cirumvention software was working just fine at the time that the alleged offense was committed. It was only after it was cracked that it became (effectively) obsolete.
That's like saying "But Officer - I didn't break into this house - the front door was wide open...right after I smashed the lock."
Dimitiri's defense (IMHO) is twofold - firstly, he did all this in Russia where it isn't illegal - secondly that he did it to aid disabled people to read eBooks and not to help the pirating of eBooks.
The whole "to help blind people" thing seems to me to be the linchpin here. If Adobe had picked some cracker who lived in the USA - and who had personally pirated a bunch of books using their own cracking tools - Adobe would have a much stronger case.
> If you buy a gun legally (ie from a liscenced firearms dealer) you > DO have to undergo a background check....but if you buy a gun legally from a private individual (eg at a gun show), you DO NOT have to undergo a background check. That's a ridiculous loophole that makes a mockery of the flimsy US gun laws.
> I doubt TV and radio from 60 light years away are going to rate.
I agree - your local TV station transmitter is at most (say) 100km from your TV antenna. The NEAREST star is three times ten to the 13 kilometers away. Since the strength of the signal drops as the SQUARE of the distance, the receiver needed to pick up TV from the nearest star would need to be about ten to the 25 times more sensitive than your TV set. I don't think we have anything that good yet - and even if we did, it would be swamped by junk from our local system.
Start looking at stars beyond the nearest handful and the sensitivity has to be MANY orders of magnitude higher than that.
I don't think SETI is sensitive enough to pick up TV signals from even the nearest stars.
To pick up their TV signals, I think we need a dedicated SETI radio telescope on the far side of the moon - something a couple of hundred miles across maybe.
So we are listening for a definite "Hello Earthlings!" type of signal from a pretty powerful transmitter. Something containing the prime numbers, the first 100 binary digits of PI, something like that.
My question is whether any aliens would send such a signal. You'd be taking one heck of a chance that it won't get picked up by more advanced civilisations with a penchant for destroying upstart planets.
It seems to me that most civilisations will be sitting - quietly listening just like we are.
1) Will the shortness of the lives of civilisations reduce the
probability of our detecting a message by so much that we won't
ever see one?
2) If we recieve a signal from a long dead civilisation - then
how will we ever talk with them?
In response to (1): The Drake equation (which estimates the number of alien civilisations that ought to be out there) takes this into account - and taking our best guess at that number, we should still expect to see a significant number of civilisations out there at the right stage in their life-span to talk to us. Of course there are a huge number of wild-ass-guesses in that equation - so making any concrete statements about the result is dangerous.
However we can never know what the typical lifespan of a civilisation is - because the only planetary civilisation we have any data on hasn't died out yet!
In response to (2), I have to say that if we could ONLY detect signals from long-dead civilisations, it would still be worth listening.
Firstly because the mere knowledge of the existance of intelligent life elsewhere in the universe would justify the search.
Secondly, it's also possible that the transmission would include the entire Encyclopedia Galactica - so even though the civilisation is dead, it might pass on knowledge that would pay for SETI a million times over.
So, whilst the shortness of the lifespan of civilisations is a concern, it's not a reason not to search.
Slashdot Mirror
When I look at the image at the head of the linked page ("Disney.jpg" - curiously)
what I see is a VERY low resolution image.
Look at the red trail behind it. There are a bunch of little raster-aligned
four-pointed star shapes. (The one on the extreme left is a prime example).
This is what you get if you take a VERY low resolution image an blow it up
with simple bilinear blending between the pixels. Taking this as evidence
of the original image resolution, we can see that the 'spaceship' at the
righthand end of the image is just about 3 pixels across - but has been
false-coloured so that the bilinear blending has become magenta and yellow
bands. Those are not 'real' they are just a part of the false-colouring.
Isn't it suspicious that the "UFO" is exactly aligned with the raster?
This is a fake...well, perhaps not exactly a fake - but an intentional
mis-use of image manipulation to produce an image that was never really
there.
You could reproduce this image in GIMP in about 3 minutes flat.
1) Create a 20x20 RGB image.
2) Using a 1 pixel brush, paint a diagonal line using bright red.
3) Fatten one end of the line slightly.
At this point, your image (if you'd gotten it from a photo of the
night sky) wouldn't convince you that this was a UFO - would it?
It could be any kind of a trail, meteor, military jet on afterburner,
a flare, a firework, anything like that.
4) Increase the image resolution to 400x400
Notice how the 'tail' now looks EXACTLY like the one in the
ufomag web site. Look at the 'star' shapes in the tail.
So, now let's do some "false-colour enhancement":
5) Choose 'select by colour' - set the threshold down to nearly
zero percent and click on a region at the center of the 'head'
of the trail. Fill it with magenta.
6) Pick a pixel close to that, fill it with a nice lemon yellow.
Notice how your image looks startlingly similar to the one
on the ufomag website. All the artifacts present in their
image are present in yours.
Now, I'm not saying that they painted their image in GIMP,
I'm quite prepared to accept that it's a photo of a real
world night-sky object. However, the pretty pink and yellow
spaceship on the right - complete with spooky red glow and
engine exhaust is no more than a deliberately produced
artifact.
The yellow and pink regions are BOTH narrower than the original
pixel resolution - no feature narrower than TWO pixels wide
(Nyquist sampling limit) can ever be reconstructed from an
image.
Bah. BULLSHIT!!
I emailed Cmdr Taco to suggest that he re-visit the idea of
a META tag that says "Slashdot: Please Mirror Me" (and yes - I've now read
the FAQ which contains a response to my META suggestion).
Rob Malda said:
> There are already tags in place to do exactly what you want. Research
> http authentication. There are many ways to prevent people you don't
> know from reading your site.
So his solution is that I should shut everyone I don't know out of
my web site?!?
Well, *THAT* helps a lot!!
Sheesh!
The FAQ basically claims that the META solution won't work because
not enough people use it.
1) If the story submission entry page explained the META mechanism
then people who submitted stories about their own sites could
defend themselves in advance.
2) If the editors were sensitive to the issue they could use their
good judgement to delay posting the story until the owner of the
site has had a chance to defend themselves with the META tag.
They'd only have to do that when the story is not hot breaking
news - and when it's a link to a small company or individual.
No need to mirror CNN, BBC, etc.
3) Making a big deal of the new META feature on Slashdot would get
the news out to a large percentage of the people who build
digital door knobs, beowulf clusters of TRS-80's and backyard
rollercoasters.
It's time for Slashdot to step up to the plate and not inflict DDOS
attacks on the very people who provide the content that they
report on.
I agree.
Slashdot really, truly, utterly needs to have a local cache of the
pages it references. It's getting to where Slashdotting is as bad as a
denial of service attack - and that's a terrible thing to inflict
on *anyone*.
Probably 50% of web sites referenced from main news items are down within
an hour of Slashdot mentioning them - and they stay down until a couple
of days have passed. That sucks.
They could easily implement some kind of opt-in thing where you put a META tag
in your web page telling Slashdot that you grant them explicit permission
to mirror the site for (say) a week after mentioning it - so Slashdot would
have no legal/copyright come-backs. At the end of the week the Slashdot
mirror could revert to become a redirect to the real site so you don't have
problems with people bookmarking the Slashdot cache instead of the real
site.
The whole process could be automated.
People who do cool things like this door lock would surely be aware that
they could get Slashdotted and prepare for the event in advance by
inserting the tag - and private individuals are the people who are
most likely to have their server die.
Companies that want to profit from their slashdotting by advertising from
their page or taking orders off of it could just leave off the META tag
and handle the traffic as now.
An opt-in cache mechanism is a win-win-win solution. Slashdot wins because
more people will use the service if it doesn't continually refer to dead
sites. Readers will win because less sites will be dead-on-arrival - and
web site operators will win (if they want to) by not having their site
die from Slashdotting.
Now Slashdot has crashed his server and he probably can't get :-)
indoors to reboot it.
Peer review - without the peer:
http://www.sjbaker.org/humor/cardboard_dog.html
The idea is that when you create your web site - you enter the Lat/Long
as a Meta tag. What you enter is entirely up to you.
Ideally, you shouldn't enter the location of the server - or of your home -
you should enter the Lat/Long of the area you would LIKE people to find
you hanging out at.
So - You want lots of emails from people interested in Outer Mongolia?
Just add Meta tags the that lat long. You have a web site about the
Eiffel tower - add Meta tags for that part of Paris. Each URL can
have different tags - so you can be in many places at once.
It doesn't threaten anonymity - if you want to stay hidden - don't add
the tags. If you want to lie about where you are - fine. If you want
your GPS to tell your PDA to email your server to tell it to update
your home page every 10 minutes - also fine...(although you'll have to
ping the geosite server to tell it to update your URL in it's database).
It seems to be a good idea. Since the information is in your web page,
any search engine can take advantage of this...all it takes is to define
a standard lat/long tag.
IMHO, they should have included altitude and an error metric in the tag.
That's bogus.
> Say it costs MS $250 to produce an XBox. They then sell it for $200.
> If you buy it, MS loses $50. If you *don't* buy it, MS loses $250.
Stoopid. That presumes that they made a bazillion X boxen and then
won't ever make any more...so if I don't buy it, they toss it into
the nearest dumpster. No way!
If you buy it, they lose $50 - if you don't buy it, they sell it to
someone else - and make one fewer next time around...they lose *nothing*.
The real deal is the question of how game makers pay M$. Generally,
they'll be paying a per-game-sold royalty. M$ rely on the royalties
from the games you buy to cover the $50 (or whatever it is) that they
lost on the console sale.
So - buy an X-box and buy no games at all - M$ loses.
Buy an X-box and buy (say) 10 games - M$ breaks even.
You could argue that games won't be made for the X-box *at all*
unless a certain number of them have been sold - and that's certainly
true. It's not worth the game developers making games if nobody's
going to buy them - but X-box is past that threshold. There are
plenty of owners out there buying games to guarantee half a dozen
new games every month.
Would it be more devastating to buy a Game Cube? Well, only if
you are going to buy enough games for it for Nintendo to turn a
profit.
I've heard that the GameCube sells at a break-even price - or perhaps
a small profit - .
I absolutely agree that teachers are woefully underpaid - and in consequence, many are under-qualified. However, this teacher would have been 100% better at her job with even ONE DAY of introductory computer training. What's worse is her unwillingness to be taught - either by the kids (who know more than she does) - or by parents like myself who try to explain where she's factually incorrect.
There seems to be an attitude at the school that so long as she maintains classroom discipline, the edutainment software will somehow teach the kids what they need to know. That's a joke - but it's certainly cheap compared to hiring a computer-literate teacher.
It's certainly evident that teachers need to be paid whatever the going rate for professionals - but teachers like my son's computer class teacher wouldn't see the benefit of such an increase because the first thing it would do would be to put her out of a job as she'd be replaced by someone who had some real computing experience.
Actually, I think almost anyone who's noodled around with PC's for a few months could do this job better. It doesn't take an IT professional - it just takes someone who is enthusiastic about PC's and has played with them extensively.
I'd like to see local computer-based companies have a policy of having each staff member spend a couple of days each year helping out in computing classrooms in their local area - paid for by the company. That would have long-term payback for everyone involved.
My kid's school (a 5th/6th grade intermediate school) has a beautiful, fully equipped computer classroom - and a teacher who teaches computing only. ...and that's the problem. The teacher knows *nothing* about computers. Practically all the kids know far more than she does.
Because she knows nothing, she dumps 'edutainment' programs onto the machines and has the kids play them continually while she merely maintains classroom discipline.
She spent three weeks (that's 40 minutes per lesson for 10 lessons) having the kids run some kind of 'typing tutor' program. Since all the kids learned to type in 3rd grade (at least as well as a typing tutor program *can* teach), they were all bored to tears with the repetitive exercises.
Fortunately, my son discovered that this stoopid program doesn't disable cut and paste - so he was able to complete all the exercises insanely quickly. Since the teacher allows them to surf the web once they have finished the assignments, he was able to go off and have fun by himself the entire time.
The crowning glory came at the end of the year when the teachers were handing out class prizes - my son was awarded the prize for best EVER score on the typing tutor by the dump computer science teacher - she proudly announced that he'd scored something like 3,000 words per minute with a 0% error rate. Some of the other teachers looked a bit strangely at her - clearly realising that something had gone amok, but perhaps assuming she'd just mis-spoken the results.
This is just one of many gaffes this teacher made. She had the kids "List 10 parts of the Computer". My kid duly wrote stuff like 'CPU', 'ROM', 'Ethernet Ataptor', 'Motherboard' - and the teacher gave him zero on the "test" saying that the correct answer was 'Mouse', 'Keyboard', 'Television' (!), 'Mouse pad', etc. When my kid complained that his computer at home didn't have a mouse pad she told him that this was nonsense and that ALL computers have mouse pads - this dissuaded him from telling her that the monitor is not, in fact, a TV set.
Similarly, she had the kids write down the 10 good things and 10 bad things about computers. My son complained that he couldn't think of 10 bad things. His teacher gave as an example: "They crash a lot" - well, since we only run Linux at home, my son knows that this isn't necessarily true and that it's not the COMPUTER that crashes - it's the SOFTWARE. Inevitably, when he complained he got in trouble.
I've written several letters to the teacher in question (she doesn't appear to read her email - even though it's provided by the school) - with poor results. I wrote and even visited with the Principal to try to get something done - but of course she just says that qualified staff are hard to get - and the State doesn't require that teachers are trained in the subject they are teaching.
So, can we conclude that teaching with computers is "A Bad Thing" ?
No!
Not unless we've carefully checked that the teachers and curriculum are sensibly chosen. Clearly, if my son's school had spent the money that went
into that computer lab in some other way, they'd have gotten more value for money and the kid's grades would have been better...but that doesn't prove that teaching computers are bad - just that they are ineptly managed.
Here in Texas, police stop people for speeding because they are funded in part from the fines they collect. It is very common to see the speed limit drop by 5mph when you cross the border from one county to another despite no other differences in road conditions. Why? Because people don't change speed as they cross the boundary - and the cops can get them on a more serious charge and raise more money.
...well, at least mine does. I have a MINI Cooper S (the new model from BMW).
When we did the 'walk through' of the controls, the dealer told me, that when you turn on the ignition, it displays the ESTIMATED number of miles to your next service. I asked the guy - "What do you mean 'estimated'? Surely it just counts down the miles? Since the odometer and the miles-to-next-service indicators are both digital, why would they ever disagree?"
Aparrently not. It monitors how 'agressively' you drive and counts the miles down faster if you redline it a lot (very tempting with the MINI BTW). This makes sense - a car that's driven hard needs servicing more often. The onboard computer knows the RPM - the number of times the traction control and dynamic steering controls kick in and everything else there is to know about how hard the poor beast is being thrashed - it's in a good position to know when a service is likely to be needed.
Does this happen in practice? Yes!
When I took delivery, there was 20 miles on the odometer and since the first service is nominally at 10,000 miles, the miles-to-next-service indicator was reading 9980 as you'd expect.
After I'd clocked up ~500 miles, driving it fairly agressively (because it's my new toy) the service indicator was saying ~9440 to go - suggesting it needed servicing 60 miles before it 'nominally' should. In the past few weeks, my driving style has returned somewhat to 'normal' and when I hit 1000 miles, the service indicator was showing ~8930 to go - so my better driving style had only cost me an additional 10 miles of 'penalty'.
So - there is no doubt that the car monitors my driving style and makes that readily apparent to the dealership - requiring me to undergo more frequent services in order to stay withing warranty if I drive agressively - and rewarding me with fewer services if I'm a good person.
Whether that's a good thing or not depends on your perspective. Yes, it's a slight invasion of privacy because your car dealer now knows you are a bad person. But if you intend to keep the car beyond the end of the warranty then you are better off for knowing that you need to service it more often in order to avoid it crapping out on you. I guess it also allows the manufacturers to set the service intervals nominally further apart - so they don't penalise good drivers by requiring more frequent services.
Even a Z80 had a 4MHz clock - if you couldn't time the engine firing to
within a millisecond or so, you'd better get a better programmer! I programmed
telephone exchange software on Z80's - and we did tons of stuff with millisecond
precision. A 200MHz x86 would be overkill in a missile.
Does the engine firing have to be more accurate than a millisecond? I doubt
very much that the engine could respond reliably at higher rates than that -
it's a big mechanical/chemical contraption and those things don't move that
fast.
Don't forget, you aren't talking about running some big complex operating
system here. The time-critical code can be in assembler - and there won't
be an OS running in there.
The only time I could imagine needing more CPU horsepower than a Z80 in a missile
would be if it had some kind of a camera or radar image to analyze as a part
of targetting.
> Microsoft will have the power to turn off whole countries by remote control.
Through "The Great Firewall of China" ? I doubt it.
But in any case, China is avoiding this by switching to Linux - avoiding
Microsoft doesn't require designing your own CPU.
The concern about Palladium is quite possibly the reason though.
So, when all US PC's are crippled with Palladium DRM technology,
us Linux users will be using Chinese CPU's?
Scarey scenario!
Yes - it is a waste of time - but in a geeky/cool way.
What it truly is - is a waste of bandwidth.
Hmmm - he promised his wife he wouldn't email porn...now I wonder if someone :-)
could get him in deep trouble with his missus by faking porn from HIS address
- now THAT would be punishment enough
I was always impressed by Steve Job's comment to the guy who was writing the Mac bootstrap code. The guy was complaining that it wasn't worth optimising the bootstrap loader any more because it was fast enough already. I don't Steve's exact words - but it was something like: We will sell 100 million of these machines - if each of those people boots their machine once a day for five years - then that's 15 billion reboots. If you can save just one second from the reboot time of the Mac then that's 480 YEARS saved.
So shaving one second of the boot time is like saving the lives of 50 people. What
could be a more noble activity than saving human lives like this.
So - applying that math to this spammer: If he sends out ten million spams a day and it takes 1 second to delete each one - and if this guy does that every day for five years - then that's morally equivelent to murdering 50 people.
Just because the damage he does to each individual is small, the cumulative damage is huge.
There is another story (probably apochryphal) about the guy writing the banking system software who changed the code to take the roundoff error (less than a half cent) from every interest calculation and direct it into his personal account. The story goes that he made tens of thousands of dollars a week. This story probably isn't true - but should such a person be considered any less a criminal because the money he stole was spread so thinly? Obviously not - he stole those thousands of dollars and that's that.
This spammer deprived the people of the world of 50 human lives - he should be considered a mass murderer and treated accordingly.
Dunno where you live - but here in sunny Texas all the toy stores
carry the bulk Lego buckets. They cost around $10 and contain around 200
parts. Some of them are *mildly* themed - so the StarWars bucket has
some interesting things like radar dishes and fins and stuff - but
it's still basically bulk Lego.
For younger kids you can also get bulk Duplo - which is twice-normal-size
Lego that is reverse-compatible with 'real Lego' but is less of a choking
hazard. Because Duplo works with Lego, they can still use it when they
grow up enough to have the real thing.
Don't be tempted into buying any of the 'clone' brands of brick - only
the genuine Lego parts are made of the softer plastic that enables the
bricks to be re-used indefinitely. The cheaper brands are made with
hard polystyrene and the wear out extremely quickly. Build about
three things with the same set of bricks and you'll start to notice
that they don't hold together anymore.
So - get genuine Lego bulk buckets. My son has about a dozen of them
- and it's never enough.
Possibly because voting in American Idol actually makes a difference.
With virtually all US politicians being bankrolled by big business - in
the case of laws like DMCA, it make zero difference who you vote for.
I don't think that is a defence:
The anti-cirumvention software was working just fine at the time that the
alleged offense was committed. It was only after it was cracked that it became
(effectively) obsolete.
That's like saying "But Officer - I didn't break into this house - the
front door was wide open...right after I smashed the lock."
Dimitiri's defense (IMHO) is twofold - firstly, he did all this in Russia
where it isn't illegal - secondly that he did it to aid disabled people to
read eBooks and not to help the pirating of eBooks.
The whole "to help blind people" thing seems to me to be the linchpin here.
If Adobe had picked some cracker who lived in the USA - and who had personally
pirated a bunch of books using their own cracking tools - Adobe would have
a much stronger case.
> If you buy a gun legally (ie from a liscenced firearms dealer) you ...but if you buy a gun legally from a private individual (eg at a gun show),
> DO have to undergo a background check.
you DO NOT have to undergo a background check. That's a ridiculous loophole
that makes a mockery of the flimsy US gun laws.
Slashdot needs to cache the pages it references for the first day or two and
then revert to a genuine hyper-link to the real site.
> I doubt TV and radio from 60 light years away are going to rate.
I agree - your local TV station transmitter is at most (say) 100km
from your TV antenna. The NEAREST star is three times ten to the
13 kilometers away. Since the strength of the signal drops as
the SQUARE of the distance, the receiver needed to pick up
TV from the nearest star would need to be about ten to the
25 times more sensitive than your TV set. I don't think we
have anything that good yet - and even if we did, it would be
swamped by junk from our local system.
Start looking at stars beyond the nearest handful and the
sensitivity has to be MANY orders of magnitude higher than
that.
I don't think SETI is sensitive enough to pick up TV signals from even
the nearest stars.
To pick up their TV signals, I think we need a dedicated SETI radio
telescope on the far side of the moon - something a couple of hundred
miles across maybe.
So we are listening for a definite "Hello Earthlings!"
type of signal from a pretty powerful transmitter. Something
containing the prime numbers, the first 100 binary digits of PI,
something like that.
My question is whether any aliens would send such a signal. You'd
be taking one heck of a chance that it won't get picked up by more
advanced civilisations with a penchant for destroying upstart
planets.
It seems to me that most civilisations will be sitting - quietly
listening just like we are.
There are two aspects to this:
1) Will the shortness of the lives of civilisations reduce the
probability of our detecting a message by so much that we won't
ever see one?
2) If we recieve a signal from a long dead civilisation - then
how will we ever talk with them?
In response to (1): The Drake equation (which estimates the number
of alien civilisations that ought to be out there) takes this into
account - and taking our best guess at that number, we should still
expect to see a significant number of civilisations out there at the
right stage in their life-span to talk to us. Of course there are a
huge number of wild-ass-guesses in that equation - so making any
concrete statements about the result is dangerous.
However we can never know what the typical lifespan of a civilisation
is - because the only planetary civilisation we have any data on hasn't
died out yet!
In response to (2), I have to say that if we could ONLY detect signals
from long-dead civilisations, it would still be worth listening.
Firstly because the mere knowledge of the existance of intelligent
life elsewhere in the universe would justify the search.
Secondly, it's also possible that the transmission would include the
entire Encyclopedia Galactica - so even though the civilisation is
dead, it might pass on knowledge that would pay for SETI a million
times over.
So, whilst the shortness of the lifespan of civilisations is a concern,
it's not a reason not to search.