Typing hasn't affected my spelling. I have not been able to spell much longer than I have been using a computer.
The problem with autocorrection is they frequently autocorrect incorrectly and often the "corrected" sentence is less understandable than the mistyped version since the corrected word is no longer recognizable as a mangling of the intended word.
I bet it can't tell the difference between someone watching the TV and someone sleeping in front of the TV.
I bet it can't tell the difference between me, sitting at the kitchen table watching the Football and my wife sitting at the breakfast table with her back turned.
I bet it can't tell that I am reading, not watching.
How does it distinguish a large dog from a small child?
If it uses infra red it can at least distinguish a human from a cardboard cut-out of the Duke of Edinburgh! I have seen award ceremonies have trouble with that one, so I guess that makes it smarter than some humans.
This seems rather like the position musicians are in. For every dollar you spend on a large music download site, the people who run the computers trowser around 82 cents while the creators of the content get around 18. That's a terrible division of the money. The cost of delivery of that content is a few cents; the rest is profit.
We see something similar with TicketMonster; charges are often around a quarter to a third of the total cost of tickets even though the preferred method of delivery is electronic; they even tack on a bit extra for the privelege of printing them on your printer, and they tack on a bunch of advertising while they do it, costing you excess ink.
Content can be delivered electronically cheaply, but its delivery is currently expensive because a handful of companies have cornered the delivery.
This is definitely a model that content providers need to break up, so Murdoch clearly does have some understanding of the way the net works and why it is bad. Unfortunately he does not seem to have worked out how to combat it. His answer that "we don't make hardware" is exactly the wrong answer - he should be leading the effort to either open up Kindle or come up with an open replacement.
My laptop has shock detection on the harddrive and I am almost certain I have had it since before April 2008. Seems like this is prior art.
In any case, devices with cases that cannot be opened undetectably have been with us a long time as have those devices that you stick on the inside of a box to detect rough handling.
Because its free, so no contract is formed between the user and the supplier.
In any case, for a lot of open source software, the bug database is also open, so making sure any bug you find is reported in a timely manner should be a good defense. Putting it in the database discloses it while making sure it is timely means you cannot be accused of keeping it secret.
It does create an incentive for projects to keep open bug databases.
Not from where I am sitting, unless they can also see through walls.
Anyway, you can pry my progressives from my cold dead fingers. No way would I switch to having to spend the day thumbing my nose. Can you imagine trying to walk down a staircase in these things?
No they're not. Although perhaps implementations differ. Qwest redirects you to their own web page and changes the URL so you have to go back and retype the whole thing. Its very annoying.
This is just one data point I know, but my scrawl was illegible long before I ever used a computer. You should see what used to come back from the keypunch operators even when I was trying to write clearly in capital block letters.
This is one of those works that everybody should have read at least by the time they leave high school. So how many high schoolers are using Kindles? - I expect its very few - while everyone else should have this on their bookshelves already, together with a bunch of other of Eric's works.
Now if it had been Misner, Wheeler and Thorne, you could understand why people would need an additional copy that was easier to read in bed.
You don't need to be able to play the violin to make money playing one on the streets. In fact its a positive advantage not to - shopkeepers come out and pay you to move away.
The problem with this line of argument is that it takes a result as a premise. Why should not some mathematics be patentable? I don't really see much difference between inventing a working sewing machine and finding a way to factor large numbers in logarithmic time (say).
The answer of course is that publishing mathematics benefits the art more than patenting it would. So allowing it to be patented in the US would be unconstitutional.
On the other hand, software patents as currently being issued only require to disclose what is done, not the total details of how it is done. If to get a software patent you had to forgo copyright and disclose all your code, there might actually be a public good to awarding such patents. If I patent a sewing machine I am patenting the machine not sewing. The problem with software patents is they are patenting sewing not the machine.
Its not their file system - its from CP/M80. The only bit they added was the long file name support, and its hard to see exactly where the originality is in that claim. Didn't Burroughs B system use variable numbers of fixed length slots for variable length file names? The idea of having multpile names for one file has, of course, been part of Unux for as long as I have used it (1984-ish). Actually I think you could do that on DOS360.
What are the operations for which this is homomorphic?
It has to be quite limited. Otherwise for example, lets suppose I have an integer (encrypted of course) and I have comparison and addition/subtraction and multiply/divide.
I can very easily find the encrypted values of both 0 (a-a for any a) and 1 (a/a)
I can now decrypt the data with repeated additions (or subtractions) of 1 and equality comparisons.
And, I don't see how you can prevent equality tests in the encrypted domain. You might have to calculate a Kernel but surely there is no way to prevent that.
So I don't see how the operations available can be as much as the usual operators on reals.
It also falls under the classification of "artifical task avoidance device". Schemes which might be legal if they served an actual commercial purpose are not if their sole purpose is to avoid paying taxes.
The devil is always in the details. If the team is simply doing support tasks, time to close is one reasonable part of a metric, but you need to at least measure customer satisfaction since just time to fix will tend to drive customer satisfaction down.
If they are responsible as well for implementing or administrating the system generating the tickets, then arrival rates need to also be measured. You don't want the team creating easy to fix outages so that they can bias the metrics.
In general you need to try to work out what behaviors a given metric will tend to, or could, produce, and you need to combine the metric with elements that measure unwanted behaviors. You want the resulting metrics to be scale free, so that the metric cannot be gamed simply by changing some parameter.
Simple example, suppsoe you measure a maintenance team by how many new regressions they create (per month say) in the maintained system. The team can get zero (which in this case is good) by never fixing any existing bugs). So as a metric this is useless. Instead of course, you should be measuring regressions per bug fixed. This is scale free because you can measure over one month or one year and the size of the values will not change just because the period changed.
If you have got a covert cable, surely you would have covert surveillance too. I simply can't imagine obvious spy types rolling up and saying "you just cut our super-secret cable". At the very least I would expect a fake comms company as cover.
But you have to care about it. You need to test it but that is not enough.
The problem is performance creep. Its easy to find big slowdowns with regression analysis, but you get perhaps as much a 5% variations in timings just depending on the phase of the moon. So any slowdown less than, say, 5% is not discernable from noise. As a result your performance can deteriorate by a few percent per checkin. Over a year that can mount up.
So you have to combat this by actually tuning the software every so often - say once per release - to recover the creep. And, of course, after you do this a couple of times, it gets harder to knock hot spots on the head and you have to do it early in the release cycle as you have to start rearchitecting to actually make a difference.
The information density of postings to Slashdot is completely random on P in [-1,1], but the distribution is not uniform - E(P)0.
(-1, of course, represents a post that is 100% wrong)
Slashdot Mirror
The problem with autocorrection is they frequently autocorrect incorrectly and often the "corrected" sentence is less understandable than the mistyped version since the corrected word is no longer recognizable as a mangling of the intended word.
I bet it can't tell the difference between me, sitting at the kitchen table watching the Football and my wife sitting at the breakfast table with her back turned.
I bet it can't tell that I am reading, not watching.
How does it distinguish a large dog from a small child?
If it uses infra red it can at least distinguish a human from a cardboard cut-out of the Duke of Edinburgh! I have seen award ceremonies have trouble with that one, so I guess that makes it smarter than some humans.
We see something similar with TicketMonster; charges are often around a quarter to a third of the total cost of tickets even though the preferred method of delivery is electronic; they even tack on a bit extra for the privelege of printing them on your printer, and they tack on a bunch of advertising while they do it, costing you excess ink.
Content can be delivered electronically cheaply, but its delivery is currently expensive because a handful of companies have cornered the delivery.
This is definitely a model that content providers need to break up, so Murdoch clearly does have some understanding of the way the net works and why it is bad. Unfortunately he does not seem to have worked out how to combat it. His answer that "we don't make hardware" is exactly the wrong answer - he should be leading the effort to either open up Kindle or come up with an open replacement.
In any case, devices with cases that cannot be opened undetectably have been with us a long time as have those devices that you stick on the inside of a box to detect rough handling.
In any case, for a lot of open source software, the bug database is also open, so making sure any bug you find is reported in a timely manner should be a good defense. Putting it in the database discloses it while making sure it is timely means you cannot be accused of keeping it secret.
It does create an incentive for projects to keep open bug databases.
Anyway, you can pry my progressives from my cold dead fingers. No way would I switch to having to spend the day thumbing my nose. Can you imagine trying to walk down a staircase in these things?
No they're not. Although perhaps implementations differ. Qwest redirects you to their own web page and changes the URL so you have to go back and retype the whole thing. Its very annoying.
She rubbed her hands together and yelled "Out out, Damn, Stop"
Is telling bad bus jokes also a felony?
What enquiring minds want to know is when they made talking on transit frequencies a felony, did they give a list of the frequencies we were to avoid?
Or perhaps he's hiking the Appalachian Trail with Lord Lucan.
If we're lucky he'll leave his clothes on a beach comewhere so we'll know he's still alive.
This is just one data point I know, but my scrawl was illegible long before I ever used a computer. You should see what used to come back from the keypunch operators even when I was trying to write clearly in capital block letters.
Now if it had been Misner, Wheeler and Thorne, you could understand why people would need an additional copy that was easier to read in bed.
The language that eventually became Fortran 90 was known as Fortran 8x during its development period.
Given that its now the latter half of 2009, the chances of C++ 0x actually having x==9 have been vanishingly small for quite a long time already.
They should put in a night soil closet where they remove the can once a week and deorbit it over some random totalitarian regime.
You don't need to be able to play the violin to make money playing one on the streets. In fact its a positive advantage not to - shopkeepers come out and pay you to move away.
The problem with this line of argument is that it takes a result as a premise. Why should not some mathematics be patentable? I don't really see much difference between inventing a working sewing machine and finding a way to factor large numbers in logarithmic time (say).
The answer of course is that publishing mathematics benefits the art more than patenting it would. So allowing it to be patented in the US would be unconstitutional.
On the other hand, software patents as currently being issued only require to disclose what is done, not the total details of how it is done. If to get a software patent you had to forgo copyright and disclose all your code, there might actually be a public good to awarding such patents. If I patent a sewing machine I am patenting the machine not sewing. The problem with software patents is they are patenting sewing not the machine.
Its not their file system - its from CP/M80. The only bit they added was the long file name support, and its hard to see exactly where the originality is in that claim. Didn't Burroughs B system use variable numbers of fixed length slots for variable length file names? The idea of having multpile names for one file has, of course, been part of Unux for as long as I have used it (1984-ish). Actually I think you could do that on DOS360.
Isn't that a good thing? Surely his system can make good use of large numbers of assholes.
It has to be quite limited. Otherwise for example, lets suppose I have an integer (encrypted of course) and I have comparison and addition/subtraction and multiply/divide.
I can very easily find the encrypted values of both 0 (a-a for any a) and 1 (a/a)
I can now decrypt the data with repeated additions (or subtractions) of 1 and equality comparisons.
And, I don't see how you can prevent equality tests in the encrypted domain. You might have to calculate a Kernel but surely there is no way to prevent that.
So I don't see how the operations available can be as much as the usual operators on reals.
It also falls under the classification of "artifical task avoidance device". Schemes which might be legal if they served an actual commercial purpose are not if their sole purpose is to avoid paying taxes.
Why is NASA using non-SI units again? Didn't they learn that lesson from the last time it caused a screw up?
If they are responsible as well for implementing or administrating the system generating the tickets, then arrival rates need to also be measured. You don't want the team creating easy to fix outages so that they can bias the metrics.
In general you need to try to work out what behaviors a given metric will tend to, or could, produce, and you need to combine the metric with elements that measure unwanted behaviors. You want the resulting metrics to be scale free, so that the metric cannot be gamed simply by changing some parameter.
Simple example, suppsoe you measure a maintenance team by how many new regressions they create (per month say) in the maintained system. The team can get zero (which in this case is good) by never fixing any existing bugs). So as a metric this is useless. Instead of course, you should be measuring regressions per bug fixed. This is scale free because you can measure over one month or one year and the size of the values will not change just because the period changed.
So this sounds like an urban myth to me.
The problem is performance creep. Its easy to find big slowdowns with regression analysis, but you get perhaps as much a 5% variations in timings just depending on the phase of the moon. So any slowdown less than, say, 5% is not discernable from noise. As a result your performance can deteriorate by a few percent per checkin. Over a year that can mount up.
So you have to combat this by actually tuning the software every so often - say once per release - to recover the creep. And, of course, after you do this a couple of times, it gets harder to knock hot spots on the head and you have to do it early in the release cycle as you have to start rearchitecting to actually make a difference.
I wonder if you can still get thermal paper for a silent 700?
The information density of postings to Slashdot is completely random on P in [-1,1], but the distribution is not uniform - E(P)0. (-1, of course, represents a post that is 100% wrong)