After all, the epoch will no longer end at 03:14:08 UTC on January 19, 2038; it will now end at 03:14:07 UTC on January 19, 2038.
But more seriously, I think time should be monotonic. Most Unix people let the hardware clock run UTC, and let the OS add time zone and daylight savings only when presenting local time to the user. All file stamps are saved in UTC.
Changing UTC is as stupid as changing your hardware clock to daylight savings and back.
It's a 'do not contact'-flag, it's not an opt-out. Even with the 'do not contact'-flag, they will keep all your information (grade points etc.), continue to gather new information, and even pass it on for purposes outside the military.
[blockquote]For me, having the 'e' correspond to my middle finger on my left hand seems like a very poor choice since it is most commonly pressed. I'd say assigning the 'e' to the pointer finger on either hand would make much more sense.[/blockquote]
The problem with the pointer finger is, that it has to deal with 2 keys on the home row, where the other fingers have only one. So your pointer finger already does a lot of work.
In the current Dvorak layout, the left pointer finger does 'u' and 'i'. This is good, because these two vowels are not often adjacent to each other. Having to type 'e' and 'i' with the same finger makes the combinations 'ie' and 'ei' harder to type, in my experience. And then there's also the consonants above and below the vowels.
Where 'learning' is any typing where you don't pause to think all the time. (copying stuff, dictation, typing tutors, etc.) If you cram lots of typing hours in very little days, the effect will be less per hour (but more per day).
It took me 2 weeks to get to 30 wpm, and I spent about two hours a day learning. After that, I stopt typing for the purpose of learning to type, and the speed gradually kept rising.
I think I can type 'comfortably' at 70 wpm - I don't usually measure my relaxed, untimed typing speed. On a short (few minutes) typing test, I can score 90-100 wpm, but it is mentally strenuous.
My 'before' qwerty score was about 60 wpm. But that is not reliably, because I typed with 6 fingers then, and with 9 fingers now.
If you're already typing really fast, the mental process rather than the motorical process can become the bottleneck. Switching layouts will not help in such a case. Even then, typing Dvorak might be more comfortable.
My 'learning' from an English typing course easily transferred to other languages, to unix commands, etc. So once I had learned to type 30wpm in English, I could do the same in Dutch without studying. But programming punctiation, the one-letter commands in mutt and emacs, and all my passwords, had to be learnt all over again. That might take some extra time, depending on which applications you use.
and I found the Dvorak layout still a great improvement over Qwerty, but slightly less efficient than it would be for English.
If I would only type in Dutch, I would switch the 'j' and the 'y' key. But I type so much English that it's not worth it.
OTOH, when I type really fast in English, I tend to make mistakes with the vowels that don't happen in Dutch. I guess it's a combination of all the vowels being next to each other, and typing in a second language with these really weird un-roman vowel sounds.
However, there are many languages which are much more different from English than Dutch is. And luckily, Dutch doesn't have Umlauts. (Well, it doesn't have Umlauts to extent it's vowel repertoire. It does sometimes use them to mark syllable boundaries, but that's very rare. And even then, I usually need to ä or \"a)
[blockquote]wait a minute.. decimals to fractions??? what kind of useless skill is that. It's trivially based on the definition of decimals:.34523452 = 34523452/100000000[/blockquote]
Hmm, I don't know where you got that decimal from, but are you sure it isn't meant to be repeating?
[pre]
x = 0.3452345234... 10000x = 3452.3452345234...
9999x = 3452
x = 3452/9999 [/pre]
Of course, the whole problem is with people typing in fractions and 'guessing' from their tiny calculator display whether it's repeating or not. Which can lead to school teachers proclaiming that they discovered 5/23 to be irrational, because it doesn't repeat (at least not on their display).
[blockquote]If anybody else needs to use my computer for whatever reason I find myself having to continually switch it to QWERTY (also, in Dvorak, QWERTY is much harder to type:) )[/blockquote]
Duhh... I type 'aoeu' to switch to quer^H^H^Hwerty, 'asdf' to switch to dvorak.
And BTW, the XHTML + MathML part of this has been implemented in Firefox for a long time, and I love it. No hassle with putting every formula in a separate MathML document.
Of course people can't say or write anything. If it's slander, or lies, they can be held responsible for their writings afterwards. If you don't want other things to be published, make it a crime so that people can be sued afterwards.
Censorship means checking everything before it is published. That is draconian. I'm not sure it's what you want. I think the problem with the questionnair is probably that they suggest that censorship is the only solution. That is bullshit. The judicial system is about punishing people for their crimes, not making it impossible to permit crimes.
How bad can it be if your computer is one hour off? If you're really bothered by it, you change the hardware clock manually (since Windows assumes the hardware clock is in local, season dependent time anyway, and PC Un*xes can conform to this). The DST season used to be different between European countries, and a few years ago this changed. I don't remember any problems over it.
This is not comparable to the Y2K problem. Your computer will not suddenly think that your grandgrandmother is only 2 years old. It will not think all you're files have changed. Most applications care about days, years, and about the relations 'earlier / later / younger / older'. These things are not affected much by a one hour time shift, whenever it may take place.
Sure, time ought to perfectly monotonous, and therefore all timestamps should be in GMT (local time should only be calculated when presenting it to the user), but this has never been the case on Windows systems, are we seem to be getting away with it.
Both braille and speech have a sequential information flow, while most software (even vi, emacs, lynx) present information in a two-dimensional way.
So it depends a lot on the user. Maybe for somebody born blind, or for an experienced Unix guy, the command line (not even emacspeak) would be great. This guy actually wrote an application that let you browse websites through ed. But I can imagine that someone used two Windows would like to stick with Windows, especially if he became blind at a later age, so has little trouble with navigating through two-dimensional information.
Most blind users seem to find speech interfaces easier than braille. Especially since with braille, you can't read and type at the same time (unless you use a one handed keyboard). Braille interfaces are better for programming and other syntax-heavy tasks.
But in the end, to be attractive to the 'average' blind user, software should be - similar in use and functionaly to current mainstream apps, i.e. not require remembering a lot of commands and syntax - present information as sequentially as possible and provide for easy navigation
Try running your app in a 1 line X-terminal, to get an idea of how blind users have to navigate.
social security non financed throught taxes
on
The Jobs Crunch
·
· Score: 1
How can they afford to do that? THey tax the rich and upper income earners more than we do (50% at least)
That's not true. Unemployment benefits are not financed throught (progressive) taxes, but through a collective (that is: non-optional) insurance for which workers pay a fixed percentage of their income. The same goes for other social securities (like workers' disabilities etc.).
Public services, welfare, 'defence', etc., are of course being financed by progressive taxes.
At least LOTR came out in cinemas all over the world at the same time. Most films come out in Europe a year later than in the US, and in Asia sometimes three years later.
Of course, real fans will always go to the movie (*and* attempt to download it to see it even before official release). But the LOTR movies were seen by a lot of people who hadn't even read the books, and these are the people that, in Europe and Asia, wouldn't have gone if they had had to wait for years and they're friends (the real fans) had long ago got a DVD and shown it to them.
Regio encoding is evil (like export subsidies, international trade restrictions etc.), and completely unnecessary if you release something worldwide simultaneously. If movies as big as LOTR and all the Bond movies can do it, it should be possible, right?
When Linus "used" Minix and GCC, he used them as tools. Is this so hard for Mr. Brown to get through his skull? Apparently so.
And I suppose Ken Brown started writing his book in Word (and the publisher manually converted it to somethinh better); does that mean that he owes Microsoft something? It seems to be his implication...
For instance, to have older programs recognize the error, but next generation programs (web browsers mainly) be able to return useful information like possible alternatives?
How is that better than web browsers receiving a normal domain not found response, and then having the browser itself take responsibility for gathering useful information e.g. from Microsoft Search?
By putting the functionality at the client level, at least you're giving the users a choice.
He praises Mac for not having open ports and Red Hat for having a firewall. Never does he mentions that a lot of Linux distros *do* ship with open ports (maybe through the firewall). 'Between Blaster and Sobig' is not a much longer period than the time between the ssh (root) exploit and the apache worm last year.
I still think Windows could be a lot more secure by default. But this article is unbalanced and therefore does not seem like a very fine analysis.
And is in every conspiracy theory, lack of evidence is evidence (since it wouldn't be a conspiracy if everything weren't secret):
"Even though IBM looks like they're not really involved in it, they're very involved," he said. "From a PR standpoint, they're able to extract themselves from (the dispute), and so they throw Red Hat at us, they throw Novell at us, they have (Open Source Initiative President) Eric Raymond on their payroll. They have all these guys that they fund and then they just step back and watch the fracas go on."
This is ridiculous. It only sends out the message: "We don't think our teachers are able to enforce discipline by themselves." And it teaches childrine that conflicts can only be resolved by hard evidence, not by taking reports. The teacher's word should be good enough, and management should back him up (but also audit classes once in a while). In classrooms are really physicially unsafe (and I think this is rare), you need to do a lot more than just install cameras.
Cameras in other places
I think this good be good, but especially for protecting property (getting evidence). Bullies will always find another place to bully, and school personnel doesn't really care anyway.
Other uses
Sometimes, cameras in classrooms (with the consent of everybody involved) could actually be useful. E.g. when part of the class is missing lessons because of an extracurricular activity, or an epidemic. Of course, you would need good sound, so it would be a bit more expensive than surveillance cameras. But watching these lessons in your own time would be a great advantage. Of course I'm assuming the lessons are actually worth watching, but if they're not, you can always surf the web and simultaneously listen in case something interesting happens in the recording.
But, rather than telling parents how to check on their childrens actions, they should learn to teach their children basic internet safety. Like: how to be invisible to strangers while chatting, how to deal with strangers, how to deal with personal information etc.
More importantly, don't just deal with the dangers, but give parents some pointers to useful resources for their children. Especially if you target at paranoid parents who have no idea what this Internet thing is all about, show them that's it's more than just the horror stories.
You might want to consider RSI problems. School furniture is usually not suited for working with laptops. In the Netherlands, regulations forbid employees from using a laptop more than two hours a day (that means that the employer is obliged to provide a docking station or PC). Recently, the Inspection has concluded that (college) student computer equipment should meet the same standards as personnel equipment. In short, if there are regulations about laptop use for employees, why shouldn't they apply to pupils? (And if that doesn't work, you can still frighten the parents.)
At least that's what it claims: you can enable an option to protect against boot sector viruses. When I have it enabled, I get a pop-up when trying to OS'es or boot loaders. But I don't know how reliable this option is.
Your main goal should be to get them interested, and to accommodate a broad audience with very different interests. Focus on books with practical use (either to catch their interest, or to allow them to practice what they learn), and on books that link different disciplines.
About computing science: a little programming is good, but I don't think most high school kids would pick up Knuth, let alone understand it. Make sure they have a chance to use those languages at school, even it those languages are Javascript and Basic. *Do* buy books on networking and computer architecture; I would have been very happy to learn more about this black box, the Internet, in high school. Something on making webpages is an easy way to get them interested.
There are plenty of interesting books on minds, brains and computers that link psychology, neuroscience, philosophy and artificial intelligence. This will be completely new to most students and a lot more interesting than some advanced A.I. algorithms. You want them to discover new things; if they want to specialize, they can always find a real library.
assuming that your default foreground is black...
If you're gonna overwrite the users default color, at least overwrite all of them. Some people actually *like* green text on a black background (don't ask me why), but if you think their 'default' text will look fine on your nice green background image, just because your default text does, think again.
Slashdot Mirror
So, when can we expect an update of localtime()?
After all, the epoch will no longer end at 03:14:08 UTC on January 19, 2038; it will now end at 03:14:07 UTC on January 19, 2038.
But more seriously, I think time should be monotonic. Most Unix people let the hardware clock run UTC, and let the OS add time zone and daylight savings only when presenting local time to the user. All file stamps are saved in UTC.
Changing UTC is as stupid as changing your hardware clock to daylight savings and back.
It's a 'do not contact'-flag, it's not an opt-out. Even with the 'do not contact'-flag, they will keep all your information (grade points etc.), continue to gather new information, and even pass it on for purposes outside the military.
[blockquote]For me, having the 'e' correspond to my middle finger on my left hand seems like a very poor choice since it is most commonly pressed. I'd say assigning the 'e' to the pointer finger on either hand would make much more sense.[/blockquote]
The problem with the pointer finger is, that it has to deal with 2 keys on the home row, where the other fingers have only one. So your pointer finger already does a lot of work.
In the current Dvorak layout, the left pointer finger does 'u' and 'i'. This is good, because these two vowels are not often adjacent to each other. Having to type 'e' and 'i' with the same finger makes the combinations 'ie' and 'ei' harder to type, in my experience. And then there's also the consonants above and below the vowels.
Where 'learning' is any typing where you don't pause to think all the time. (copying stuff, dictation, typing tutors, etc.) If you cram lots of typing hours in very little days, the effect will be less per hour (but more per day).
It took me 2 weeks to get to 30 wpm, and I spent about two hours a day learning. After that, I stopt typing for the purpose of learning to type, and the speed gradually kept rising.
I think I can type 'comfortably' at 70 wpm - I don't usually measure my relaxed, untimed typing speed. On a short (few minutes) typing test, I can score 90-100 wpm, but it is mentally strenuous.
My 'before' qwerty score was about 60 wpm. But that is not reliably, because I typed with 6 fingers then, and with 9 fingers now.
If you're already typing really fast, the mental process rather than the motorical process can become the bottleneck. Switching layouts will not help in such a case. Even then, typing Dvorak might be more comfortable.
My 'learning' from an English typing course easily transferred to other languages, to unix commands, etc. So once I had learned to type 30wpm in English, I could do the same in Dutch without studying. But programming punctiation, the one-letter commands in mutt and emacs, and all my passwords, had to be learnt all over again. That might take some extra time, depending on which applications you use.
and I found the Dvorak layout still a great improvement over Qwerty, but slightly less efficient than it would be for English.
If I would only type in Dutch, I would switch the 'j' and the 'y' key. But I type so much English that it's not worth it.
OTOH, when I type really fast in English, I tend to make mistakes with the vowels that don't happen in Dutch. I guess it's a combination of all the vowels being next to each other, and typing in a second language with these really weird un-roman vowel sounds.
However, there are many languages which are much more different from English than Dutch is. And luckily, Dutch doesn't have Umlauts. (Well, it doesn't have Umlauts to extent it's vowel repertoire. It does sometimes use them to mark syllable boundaries, but that's very rare. And even then, I usually need to ä or \"a)
[blockquote]wait a minute.. decimals to fractions??? what kind of useless skill is that. It's trivially based on the definition of decimals: .34523452 = 34523452/100000000[/blockquote]
Hmm, I don't know where you got that decimal from, but are you sure it isn't meant to be repeating?
[pre]
x = 0.3452345234...
10000x = 3452.3452345234...
9999x = 3452
x = 3452/9999
[/pre]
Of course, the whole problem is with people typing in fractions and 'guessing' from their tiny calculator display whether it's repeating or not. Which can lead to school teachers proclaiming that they discovered 5/23 to be irrational, because it doesn't repeat (at least not on their display).
[blockquote]If anybody else needs to use my computer for whatever reason I find myself having to continually switch it to QWERTY (also, in Dvorak, QWERTY is much harder to type :) )[/blockquote]
Duhh... I type 'aoeu' to switch to quer^H^H^Hwerty, 'asdf' to switch to dvorak.
Well, I'm not sure if it can count as a standard already, but at least the w3 is working on it:
V G-20020809/
http://www.w3.org/TR/2002/WD-XHTMLplusMathMLplusS
And BTW, the XHTML + MathML part of this has been implemented in Firefox for a long time, and I love it. No hassle with putting every formula in a separate MathML document.
Of course people can't say or write anything. If it's slander, or lies, they can be held responsible for their writings afterwards. If you don't want other things to be published, make it a crime so that people can be sued afterwards.
Censorship means checking everything before it is published. That is draconian. I'm not sure it's what you want. I think the problem with the questionnair is probably that they suggest that censorship is the only solution. That is bullshit. The judicial system is about punishing people for their crimes, not making it impossible to permit crimes.
How bad can it be if your computer is one hour off? If you're really bothered by it, you change the hardware clock manually (since Windows assumes the hardware clock is in local, season dependent time anyway, and PC Un*xes can conform to this). The DST season used to be different between European countries, and a few years ago this changed. I don't remember any problems over it.
This is not comparable to the Y2K problem.
Your computer will not suddenly think that your grandgrandmother is only 2 years old. It will not think all you're files have changed. Most applications care about days, years, and about the relations 'earlier / later / younger / older'. These things are not affected much by a one hour time shift, whenever it may take place.
Sure, time ought to perfectly monotonous, and therefore all timestamps should be in GMT (local time should only be calculated when presenting it to the user), but this has never been the case on Windows systems, are we seem to be getting away with it.
Both braille and speech have a sequential information flow, while most software (even vi, emacs, lynx) present information in a two-dimensional way.
So it depends a lot on the user. Maybe for somebody born blind, or for an experienced Unix guy, the command line (not even emacspeak) would be great. This guy actually wrote an application that let you browse websites through ed.
But I can imagine that someone used two Windows would like to stick with Windows, especially if he became blind at a later age, so has little trouble with navigating through two-dimensional information.
Most blind users seem to find speech interfaces easier than braille. Especially since with braille, you can't read and type at the same time (unless you use a one handed keyboard). Braille interfaces are better for programming and other syntax-heavy tasks.
But in the end, to be attractive to the 'average' blind user, software should be
- similar in use and functionaly to current mainstream apps, i.e. not require remembering a lot of commands and syntax
- present information as sequentially as possible and provide for easy navigation
Try running your app in a 1 line X-terminal, to get an idea of how blind users have to navigate.
That's not true. Unemployment benefits are not financed throught (progressive) taxes, but through a collective (that is: non-optional) insurance for which workers pay a fixed percentage of their income. The same goes for other social securities (like workers' disabilities etc.).
Public services, welfare, 'defence', etc., are of course being financed by progressive taxes.
At least LOTR came out in cinemas all over the world at the same time. Most films come out in Europe a year later than in the US, and in Asia sometimes three years later.
Of course, real fans will always go to the movie (*and* attempt to download it to see it even before official release). But the LOTR movies were seen by a lot of people who hadn't even read the books, and these are the people that, in Europe and Asia, wouldn't have gone if they had had to wait for years and they're friends (the real fans) had long ago got a DVD and shown it to them.
Regio encoding is evil (like export subsidies, international trade restrictions etc.), and completely unnecessary if you release something worldwide simultaneously. If movies as big as LOTR and all the Bond movies can do it, it should be possible, right?
And I suppose Ken Brown started writing his book in Word (and the publisher manually converted it to somethinh better); does that mean that he owes Microsoft something? It seems to be his implication...
How to Design Programs
How is that better than web browsers receiving a normal domain not found response, and then having the browser itself take responsibility for gathering useful information e.g. from Microsoft Search?
By putting the functionality at the client level, at least you're giving the users a choice.
He praises Mac for not having open ports and Red Hat for having a firewall. Never does he mentions that a lot of Linux distros *do* ship with open ports (maybe through the firewall). 'Between Blaster and Sobig' is not a much longer period than the time between the ssh (root) exploit and the apache worm last year.
I still think Windows could be a lot more secure by default. But this article is unbalanced and therefore does not seem like a very fine analysis.
Cameras in classrooms?
This is ridiculous. It only sends out the message: "We don't think our teachers are able to enforce discipline by themselves." And it teaches childrine that conflicts can only be resolved by hard evidence, not by taking reports. The teacher's word should be good enough, and management should back him up (but also audit classes once in a while). In classrooms are really physicially unsafe (and I think this is rare), you need to do a lot more than just install cameras.
Cameras in other places
I think this good be good, but especially for protecting property (getting evidence). Bullies will always find another place to bully, and school personnel doesn't really care anyway.
Other uses
Sometimes, cameras in classrooms (with the consent of everybody involved) could actually be useful. E.g. when part of the class is missing lessons because of an extracurricular activity, or an epidemic. Of course, you would need good sound, so it would be a bit more expensive than surveillance cameras. But watching these lessons in your own time would be a great advantage. Of course I'm assuming the lessons are actually worth watching, but if they're not, you can always surf the web and simultaneously listen in case something interesting happens in the recording.
Indeed, if you spell it like that, you're not using a qwerty keyboard.
This is actually a good idea.
But, rather than telling parents how to check on their childrens actions, they should learn to teach their children basic internet safety. Like: how to be invisible to strangers while chatting, how to deal with strangers, how to deal with personal information etc.
More importantly, don't just deal with the dangers, but give parents some pointers to useful resources for their children. Especially if you target at paranoid parents who have no idea what this Internet thing is all about, show them that's it's more than just the horror stories.
You might want to consider RSI problems. School furniture is usually not suited for working with laptops. In the Netherlands, regulations forbid employees from using a laptop more than two hours a day (that means that the employer is obliged to provide a docking station or PC). Recently, the Inspection has concluded that (college) student computer equipment should meet the same standards as personnel equipment. In short, if there are regulations about laptop use for employees, why shouldn't they apply to pupils? (And if that doesn't work, you can still frighten the parents.)
At least that's what it claims: you can enable an option to protect against boot sector viruses. When I have it enabled, I get a pop-up when trying to OS'es or boot loaders. But I don't know how reliable this option is.
Your main goal should be to get them interested, and to accommodate a broad audience with very different interests. Focus on books with practical use (either to catch their interest, or to allow them to practice what they learn), and on books that link different disciplines.
About computing science: a little programming is good, but I don't think most high school kids would pick up Knuth, let alone understand it. Make sure they have a chance to use those languages at school, even it those languages are Javascript and Basic. *Do* buy books on networking and computer architecture; I would have been very happy to learn more about this black box, the Internet, in high school. Something on making webpages is an easy way to get them interested.
There are plenty of interesting books on minds, brains and computers that link psychology, neuroscience, philosophy and artificial intelligence. This will be completely new to most students and a lot more interesting than some advanced A.I. algorithms. You want them to discover new things; if they want to specialize, they can always find a real library.
assuming that your default foreground is black... If you're gonna overwrite the users default color, at least overwrite all of them. Some people actually *like* green text on a black background (don't ask me why), but if you think their 'default' text will look fine on your nice green background image, just because your default text does, think again.