"combatants" and "illegal combatants" or "bandits".
They are apparently using the phrase "illegal combatants" (and I doubt they said it in English) as a synonym for "violent criminal". Bandits are criminals, and must be turned over to civilian law-enforcement as soon as possible. However, nobody who defends his home town against a surprise aerial bombardment can be called a bandit.
The prisoners in Guantamo are either POWs or kidnap victims, depending on whether the US agrees to label them "combatants" under international law.
You can walk down the corridor of any prison in the world and hear 'I didn't do it' called out in the dark.
That's rare, at least in the US. Someone who protests his innocence sets himself up to be assaulted by the other prisoners. They quickly learn not to shout that stuff randomly.
Since it never (formally) started, I guess we're in for a bit of a wait.
and according to the laws of war rules and laws are not applicable on illegal combatants;
Nope. Looks like you're repeating an invention of Donald Rumsfeld.
Laws of war cover combatants. The idea of an "illegal combatant" was only recently invented by the US. There is no such distinction in any Geneva or similar conventions.
Re:Extent of orbital normality
on
Eating in Space
·
· Score: 1
For example, there was no "table" on ISS
A "table" is a machine which prevents objects from responding to gravity's pull. In a situation with no noticable gravitational force, tables are meaningless... unless augmented with magnets or velcro.
Please don't give this nonsense about the 'science' that is being conducted in space. There is no science in space, there's nothing in space.
While you're basically correct, there is plenty of scientifically valuable tasks to do in space. Humans aren't needed to do them... only robots. That bit of hyperbole gets you modded down.
Heroes are people who things that are extraordinary, inspiring, and, above all, useful.
Sadly, no. A 'hero' is anyone whom a mass of people call a hero; it's a democratically-defined honorific. That's why Jessica Lynch became a hero for her military failure.
Although the US NASA had hero-building as a main goal, it's failing at even that minor task. Few in the public can name a current astronaut... until he explodes, that is!
A download off your website easily qualfies here. YOU are responsible for providing the download though, you can't just point them at a different copy someone else maintains.
Fine, go explain to a Yale Law professor about how you're better at reading a license than he is.
That's completely different from what is being proposed now. Taxing phone calls "per se" is exactly what they're considering.
If they must tax the internet, then a percentage tax taken from the bill collected by the ISP would be a much better idea. That, at least, would be fair and wouldn't discriminate amoung one internet protocol or another. I don't want to see (for example) people prefering NetMeeting over VoIP or AIM over email just because one of them isn't taxed.
Taxes are much less onerous when they are attached to an existing monentary transaction. Sales tax, hotel tax, income tax... the hurt that those things do to the payer is mainly from the actual money taken away. But taxes on something free impose many more costs- you've now got to go through all the paperwork of making a transaction that hadn't been necessary at all. (Like how the biggest irritation of tollbooths is not the money itself, but the traffic congestion from having to sit in line digging for coins)
But if you buy the argument that web and e-mail access should be universal, then it suggests the need for some form of tax & subsidy scheme to provide that access to everyone.
Again, that is completely unrelated to the discussion at hand. The government certainly isn't proposing a "Rural Internetization Project". There's no specific internet service they'd be funding- the tax would go towards the general fund.
for instance, the make manual; ignore the fact that there is probably a more suitable for printing PDF version available directly from GNU
Gnu.org would never host a newfangled non-standard* format like PDF when Postscript is perfectly able to present "as printed" data.
However, if I wanted the manual printed, I'd certainly go to the HTML, not PDF. Since I can read significantly smaller text than the average person, the PDF would use up 25 times as much paper as the HTML. (People with impaired vision would suffer the opposite problem).
If the file was only available in one format, it should be HTML, not PDF. Converting an HTML to a decent PDF is easy for an end-user, but the reverse translation is much more difficult.
(Note that to optimally print an HTML file requires smarter software than is typically used. For example, links to targets in the same file should be replaced with parenthetical page numbers)
* Adobe has a PDF specification document online, but it is incomplete. There are some PDF files Acrobat can read, but those specs don't describe. Whenever someone publishes how to read those files, Adobe has him arrested.
Re:Am not sure 3-click rule was really *debunked*
on
Web 'Rules' Changing?
·
· Score: 1
But for a regular user of a site, it might not make a difference if they need 3 or 25 clicks to reach something. They might be happily exploring the site and feel they are learning something that will be useful next time.
Yes, exactly. The 3-click rule is more an advertising strategy than anything else. It's meant to ensure that new visitors can find something they enjoy quickly.
Once users have decided they like a site, it's still important for the content to be easily accessible... but counting the times a visitor clicks is not a valid measure of whether the 3-click rule was obeyed.
One of the study's unjustified assumptions is that one website-visit = one piece of desired information. In reality, a site visitor may want to see several different things. For example, 12 clicks spent on The Onion probably means that 7+ different articles were read, since most everything is accessible in less than 1-2 clicks.
I find that annoying because, assuming I'm interested, the first thing I end up doing is printing the HTML page to a PDF file so I can archive it.
This I can't believe. How can a PDF ever be better than HTML for digital archiving? HTML was meant to be read on computer; PDF is intended to be printed out.
Unless you really meant "ugly HTML" instead of merely "HTML". Stupid web pages with colorful toolbars, formatting, background pics, tables-for-layour, ad banners, 'related content' links and 'click here for page 3/21' on the bottom... they're a tough way to read documents, and I suppose a PDF could be an improvement.
But the best way for publishers to present documentation is as simple, usable HTML. Then, if the reader wants a PDF, she can print it herself, and it'll take whatever font and pagination she prefers. (PDFs created by publishers are greatly flawed in that the layout is frozen, instead of being dependent on the qualities of the output device. If I'm reading on a computer, there should be no page breaks.)
Most of the price goes to the retail channel which multiplies the price several times before putting on the shelf.
No... most of the price goes back to the publisher. Of course, much of that money gets sent back to the retailer, because game publishers must rent shelf-space in stores.
If anything, that - and experimenting with user created GR scenarious with abyssmal voice-acting and scenarios - taught me that sometimes it just is worth paying for quality game software.
A self-contradictory statement. First it notes that scenarios are more important than the software, then reverts to praising software.
It is notable that the elements of videogames least amenable to opensource-style production are the non-software parts: maps, models, pictures, and sounds.
Putting source code on a website is orthogonal to GPL compliance.
huh. you must have read an entirely different GPL than i did.
No. Guess I just understood it better. Let's go through this. The GPL line you quote is from the preamble, which is less precise (and less legally-binding) than the actual license requirements which follow. The requirements really are in section 3, and they say you must either include the whole source code, or an offer to get that source code. "A written offer, valid for at least 3 years, to any third party"
Putting code on a webserver does not satisfy that requirement. It may sound strange, and it's fairly rare, but not everyone has web access. You still need a way to give code to non-web users who contact you. And since the system which works for non-web-users can also serve web-users, putting code on the web doesn't make you more compliant to GPL.
Downloadable source isn't enough to be compliant. Therefore a penny-pinching corporation, who only wants to do the minimum required effort to obey the license, won't do this.
hm. sounds to me like including a web-site address, along with the GPL license in the rest of your legal license statements (you do buy stuff from stores, and not on the grey/black market, correct?) would fulfill the license agreement nicely.
Absolutely not. A URL is not an offer. It can be part of an offer, but it's not enough.
And, "Putting the code on the web" doesn't mean that the website address was included with the product.
along with the GPL license in the rest of your legal license statements
That's a good idea, and completely separate from whether the code is up for download. The company in this case didn't do that. Arguably, that is their biggest violation.
Germany was NOT a democracy when they started their aggression. Have you ever heard of Kristallenacht? That was the night the MINORITY Nazi party seized power in a coup and quickly killed and imprisoned their enemies.
What a joke. A supreme joke. Your line about "do you actually study any history?" applies 5000x stronger to yourself.
Your definition of Kristallnacht is so completely wrong that I'm laughing too hard to read the post any further. Even with the sorry state of US education, I'd expect a typical 7th grader to know that Kristallnacht was an anti-semitic pogrom. It was a popular event- a grass-roots riot, not an imposition from above. It cement Hitler's popular esteem. It was an effect, not a cause. (Maybe you were thinking of this?)
The GPL states that those that distribute GPLed software must provide the source to the recipients of the distributions upon request.
It says more than that. They must TELL customers the source is GPL. They can't leave it as a secret for people to guess. Lying about the license status is a license violation.
It means that people you receive the ditribution, in the form of the DVD player, must receive the source upon request. No one else is entitled to the source, only those that have received the distribution.
Why, why, why can't people PRETEND to READ the GPL before going on and lecturing about it?
Here, I will help you. I will paste, and you may read.
Accompany it with a written offer, valid for at least three years, to
give any third party, for a charge no more than your cost of physically performing source distribution, a complete machine-readable copy of the corresponding source code
Had I ever accepted contributions to my GPL projects without copyright assignment, pieces of those projects would be GPL *and* copyright to someone else, and I wouldn't be able to use my own projects at work!
Ahem. "Use". If they were GPL, You could use them at work. *I* could use them at work.
What you can't do if they're GPL is republish the code in proprietary software. And it's quite likely that people who might send you patches wouldn't want you to be able to re-sell the things they sent you for free. That's exactly why the GPL was created- to prevent an author's code from being re-used where he's not allowed to see it.
you ship 300 products, all based on a GPL'd product. your cost is a web server with source, let's say $100-500/month.
A company interesting in minimizing expenses wouldn't want to host put the GPL on a $100/month webserver. Having the source on a webserver doesn't comply with the GPL... it's just a convenience for the public.
To comply, they can either include the source code with the product, or include an offer for the source. Including with the product is cheap&easy if they are already including a CD-ROM for other reasons. Otherwise, adding a coupon for the source is the cheapest way... because they can require anyone needing the code to pay $30 for the priviledge (enough to cover the time of the employee who responds)
Peace happened in Germany when we killed those who wanted war; we didn't come to an understanding with them.
Wrongo. Germany was a democracy. War was favored by a majority of the population.
The majority of the population was not killed. They came to an understanding while staring into machinegun barrels.
Peace happened in Japan when we killed those who wanted war; we didn't come to an understanding with them.
Wrongo again. The majority of the nation wanted war. They weren't killed; they came to an understand while glancing upwards for more atom bombs.
it ISN'T the Israelis who want war.
It also WASN'T the Germans who wanted WWII. Conquerors never want war; they prefer the invaded nation to roll over and give up the territory without a fight.
How do you explain to a robot the difference between an enemy and a civilian........ In the middle east a shepard has a beard, a turban and a kalashnikov. Enemy troops has a beard, a turban and a kalashnikov.
Human infantry will have exactly the same problem- the difficulty of recognizing enemies isn't unique to robotics. (In fact, the currently active and near-future battlefield robots all use remotely-viewed cameras to detect targets, so its still a human who makes the decision)
A battlefield robot may be LESS likely to kill the wrong target for several reasons: 1) US soldiers can rarely speak Arabic. They can't talk to strangers they meet. One translator could stand by to assist the operators of 40 RC robots whenever they need to listen to someone.
2) The most important reason, by far: An AK47 cannot kill a robot. Neither can a Stinger, or a hydrogen bomb. There's no lives at risk. An infantry platoon that comes upon an armed stranger will have an inclination to shoot him out of self defense. A robot has no fear of death, and won't open fire on a dark flicker that might be a sniper, or a farmer who made a rapid grab into the back of a truck. (In jargon, a robot would operate at a lower force protection posture than DIs)
For the US military, losing millions of dollars in high-tech hardware is perfectly fine if it protects the troops. (And hey, then they need to buy more robots, which feeds the defense contractors and spurs the economy!)
He said "Real programmers don't have to test the whole world". You said real programmers will "re-test your product". Which is exactly in agreement, so why are you attacking?
"Don't test the whole world" means "test your product, but not every single thing that runs on the same system as your product".
Things like test scrips for perl and bzip2 are testing your product, not the whole world. Microsoft(tm) design methodology is bad because unit testing is insufficient to predict how many dependent apps a patch will break.
A critical bug in Konqueror and all of KDE becomes useless.
This gets back to the terms sproketboy used: no "commingling" in a "properly written application".
I won't go into a 10-page lecture on software engineering. But just because an application is depended on by any others doesn't mean they're comingled, or improperly written. A good component app will have a limited number of interfaces to the rest of the system (on the order of 10-200, and hopefully towards the low side).
Testing the program's correctness on those interfaces gives you a high trust that it'll work correctly in the larger system.
Microsoft(tm) IE(r) isn't like that. It doesn't have defined interfaces to the rest of the system. Its not an application which runs on the OS kernel and talks with other apps. It's source code is intermixed with much of the rest of the Windows OS. Testing every interface isn't enough to show that a new version is working right... you'd have to go through every line of code and see how it might possibly perturb Windows itself.
Compared to component-interface testing, that's a prohitably lengthy task; a combinatorical explosion of places to check.
no Kate working no editors
Again, Kate is one component, and testing that component's agreement with each of its public interfaces should be enough to verify there are no critical bugs. That only works if the components are well-separated enough. But separation leads to slowness, and Microsoft wants to be fast.
You're a bitch for thinking the USA is terrorists.
The US, like every nation with a victorious army, uses fear and terror to get what they want. Scaring enemies into cooperation by killing some of them is considered morally superior to the alternative of shooting them all.
stopping a bitch from killing thousands of people in a single day isn't terror its good
Nobody has killed more people (130,000) with a single bomb than a US pilot.
Slashdot Mirror
"combatants" and "illegal combatants" or "bandits".
They are apparently using the phrase "illegal combatants" (and I doubt they said it in English) as a synonym for "violent criminal". Bandits are criminals, and must be turned over to civilian law-enforcement as soon as possible. However, nobody who defends his home town against a surprise aerial bombardment can be called a bandit.
The prisoners in Guantamo are either POWs or kidnap victims, depending on whether the US agrees to label them "combatants" under international law.
You can walk down the corridor of any prison in the world and hear 'I didn't do it' called out in the dark.
That's rare, at least in the US. Someone who protests his innocence sets himself up to be assaulted by the other prisoners. They quickly learn not to shout that stuff randomly.
the war is not (formally) over in Afganistan yet.
Since it never (formally) started, I guess we're in for a bit of a wait.
and according to the laws of war rules and laws are not applicable on illegal combatants;
Nope. Looks like you're repeating an invention of Donald Rumsfeld.
Laws of war cover combatants. The idea of an "illegal combatant" was only recently invented by the US. There is no such distinction in any Geneva or similar conventions.
For example, there was no "table" on ISS
A "table" is a machine which prevents objects from responding to gravity's pull. In a situation with no noticable gravitational force, tables are meaningless... unless augmented with magnets or velcro.
Please don't give this nonsense about the 'science' that is being conducted in space. There is no science in space, there's nothing in space.
While you're basically correct, there is plenty of scientifically valuable tasks to do in space. Humans aren't needed to do them... only robots. That bit of hyperbole gets you modded down.
Heroes are people who things that are extraordinary, inspiring, and, above all, useful.
Sadly, no. A 'hero' is anyone whom a mass of people call a hero; it's a democratically-defined honorific. That's why Jessica Lynch became a hero for her military failure.
Although the US NASA had hero-building as a main goal, it's failing at even that minor task. Few in the public can name a current astronaut... until he explodes, that is!
A download off your website easily qualfies here. YOU are responsible for providing the download though, you can't just point them at a different copy someone else maintains.
Fine, go explain to a Yale Law professor about how you're better at reading a license than he is.
The source has to be as easy to get to as the binaries.
No, it has to be "with" the binaries. Read the GPL people!
If you distribute the binaries on CD, I would consiter web accessable sources at least as easy to get to.
That's completely wrong. If the binaries are on CD, then source on the CD will be just as easy to reach.
Source on the web will be harder to reach for some people, and easier for others, but never equal.
They don't need to tax the phone call per se.
That's completely different from what is being proposed now. Taxing phone calls "per se" is exactly what they're considering.
If they must tax the internet, then a percentage tax taken from the bill collected by the ISP would be a much better idea. That, at least, would be fair and wouldn't discriminate amoung one internet protocol or another. I don't want to see (for example) people prefering NetMeeting over VoIP or AIM over email just because one of them isn't taxed.
Taxes are much less onerous when they are attached to an existing monentary transaction. Sales tax, hotel tax, income tax... the hurt that those things do to the payer is mainly from the actual money taken away. But taxes on something free impose many more costs- you've now got to go through all the paperwork of making a transaction that hadn't been necessary at all. (Like how the biggest irritation of tollbooths is not the money itself, but the traffic congestion from having to sit in line digging for coins)
But if you buy the argument that web and e-mail access should be universal, then it suggests the need for some form of tax & subsidy scheme to provide that access to everyone.
Again, that is completely unrelated to the discussion at hand. The government certainly isn't proposing a "Rural Internetization Project". There's no specific internet service they'd be funding- the tax would go towards the general fund.
for instance, the make manual; ignore the fact that there is probably a more suitable for printing PDF version available directly from GNU
Gnu.org would never host a newfangled non-standard* format like PDF when Postscript is perfectly able to present "as printed" data.
However, if I wanted the manual printed, I'd certainly go to the HTML, not PDF. Since I can read significantly smaller text than the average person, the PDF would use up 25 times as much paper as the HTML. (People with impaired vision would suffer the opposite problem).
If the file was only available in one format, it should be HTML, not PDF. Converting an HTML to a decent PDF is easy for an end-user, but the reverse translation is much more difficult.
(Note that to optimally print an HTML file requires smarter software than is typically used. For example, links to targets in the same file should be replaced with parenthetical page numbers)
* Adobe has a PDF specification document online, but it is incomplete. There are some PDF files Acrobat can read, but those specs don't describe. Whenever someone publishes how to read those files, Adobe has him arrested.
there must be atleast one rule.
Do what thou whilt shall be the whole of the law
But for a regular user of a site, it might not make a difference if they need 3 or 25 clicks to reach something. They might be happily exploring the site and feel they are learning something that will be useful next time.
Yes, exactly. The 3-click rule is more an advertising strategy than anything else. It's meant to ensure that new visitors can find something they enjoy quickly.
Once users have decided they like a site, it's still important for the content to be easily accessible... but counting the times a visitor clicks is not a valid measure of whether the 3-click rule was obeyed.
One of the study's unjustified assumptions is that one website-visit = one piece of desired information. In reality, a site visitor may want to see several different things. For example, 12 clicks spent on The Onion probably means that 7+ different articles were read, since most everything is accessible in less than 1-2 clicks.
I find that annoying because, assuming I'm interested, the first thing I end up doing is printing the HTML page to a PDF file so I can archive it.
This I can't believe. How can a PDF ever be better than HTML for digital archiving? HTML was meant to be read on computer; PDF is intended to be printed out.
Unless you really meant "ugly HTML" instead of merely "HTML". Stupid web pages with colorful toolbars, formatting, background pics, tables-for-layour, ad banners, 'related content' links and 'click here for page 3/21' on the bottom... they're a tough way to read documents, and I suppose a PDF could be an improvement.
But the best way for publishers to present documentation is as simple, usable HTML. Then, if the reader wants a PDF, she can print it herself, and it'll take whatever font and pagination she prefers. (PDFs created by publishers are greatly flawed in that the layout is frozen, instead of being dependent on the qualities of the output device. If I'm reading on a computer, there should be no page breaks.)
Most of the price goes to the retail channel which multiplies the price several times before putting on the shelf.
No... most of the price goes back to the publisher. Of course, much of that money gets sent back to the retailer, because game publishers must rent shelf-space in stores.
If anything, that - and experimenting with user created GR scenarious with abyssmal voice-acting and scenarios - taught me that sometimes it just is worth paying for quality game software.
A self-contradictory statement. First it notes that scenarios are more important than the software, then reverts to praising software.
It is notable that the elements of videogames least amenable to opensource-style production are the non-software parts: maps, models, pictures, and sounds.
The GPL states that if the people who bought my boxed version requests me for the source I have to provide it.
Actually he is wrong. And you are wrong. The GPL is easy to download, try reading it next time before making a fool of yourself.
Hint: search for "any third party"
Putting source code on a website is orthogonal to GPL compliance.
huh. you must have read an entirely different GPL than i did.
No. Guess I just understood it better. Let's go through this. The GPL line you quote is from the preamble, which is less precise (and less legally-binding) than the actual license requirements which follow. The requirements really are in section 3, and they say you must either include the whole source code, or an offer to get that source code. "A written offer, valid for at least 3 years, to any third party"
Putting code on a webserver does not satisfy that requirement. It may sound strange, and it's fairly rare, but not everyone has web access. You still need a way to give code to non-web users who contact you. And since the system which works for non-web-users can also serve web-users, putting code on the web doesn't make you more compliant to GPL.
Downloadable source isn't enough to be compliant. Therefore a penny-pinching corporation, who only wants to do the minimum required effort to obey the license, won't do this.
hm. sounds to me like including a web-site address, along with the GPL license in the rest of your legal license statements (you do buy stuff from stores, and not on the grey/black market, correct?) would fulfill the license agreement nicely.
Absolutely not. A URL is not an offer. It can be part of an offer, but it's not enough.
And, "Putting the code on the web" doesn't mean that the website address was included with the product.
along with the GPL license in the rest of your legal license statements
That's a good idea, and completely separate from whether the code is up for download. The company in this case didn't do that. Arguably, that is their biggest violation.
Germany was NOT a democracy when they started their aggression. Have you ever heard of Kristallenacht? That was the night the MINORITY Nazi party seized power in a coup and quickly killed and imprisoned their enemies.
What a joke. A supreme joke. Your line about "do you actually study any history?" applies 5000x stronger to yourself.
Your definition of Kristallnacht is so completely wrong that I'm laughing too hard to read the post any further. Even with the sorry state of US education, I'd expect a typical 7th grader to know that Kristallnacht was an anti-semitic pogrom. It was a popular event- a grass-roots riot, not an imposition from above. It cement Hitler's popular esteem. It was an effect, not a cause. (Maybe you were thinking of this?)
For exhaustive details on why Nazi Germany was emphatically a democracy, see Daneil Goldhagen's compiled evidence.
It says more than that. They must TELL customers the source is GPL. They can't leave it as a secret for people to guess. Lying about the license status is a license violation.
It means that people you receive the ditribution, in the form of the DVD player, must receive the source upon request. No one else is entitled to the source, only those that have received the distribution.
Why, why, why can't people PRETEND to READ the GPL before going on and lecturing about it?
Here, I will help you. I will paste, and you may read.
Next time before you say anything about GPL, check the original first.
Had I ever accepted contributions to my GPL projects without copyright assignment, pieces of those projects would be GPL *and* copyright to someone else, and I wouldn't be able to use my own projects at work!
Ahem. "Use". If they were GPL, You could use them at work. *I* could use them at work.
What you can't do if they're GPL is republish the code in proprietary software. And it's quite likely that people who might send you patches wouldn't want you to be able to re-sell the things they sent you for free. That's exactly why the GPL was created- to prevent an author's code from being re-used where he's not allowed to see it.
you ship 300 products, all based on a GPL'd product. your cost is a web server with source, let's say $100-500/month.
A company interesting in minimizing expenses wouldn't want to host put the GPL on a $100/month webserver. Having the source on a webserver doesn't comply with the GPL... it's just a convenience for the public.
To comply, they can either include the source code with the product, or include an offer for the source. Including with the product is cheap&easy if they are already including a CD-ROM for other reasons. Otherwise, adding a coupon for the source is the cheapest way... because they can require anyone needing the code to pay $30 for the priviledge (enough to cover the time of the employee who responds)
Peace happened in Germany when we killed those who wanted war; we didn't come to an understanding with them.
Wrongo. Germany was a democracy. War was favored by a majority of the population.
The majority of the population was not killed. They came to an understanding while staring into machinegun barrels.
Peace happened in Japan when we killed those who wanted war; we didn't come to an understanding with them.
Wrongo again. The majority of the nation wanted war. They weren't killed; they came to an understand while glancing upwards for more atom bombs.
it ISN'T the Israelis who want war.
It also WASN'T the Germans who wanted WWII. Conquerors never want war; they prefer the invaded nation to roll over and give up the territory without a fight.
How do you explain to a robot the difference between an enemy and a civilian........ In the middle east a shepard has a beard, a turban and a kalashnikov. Enemy troops has a beard, a turban and a kalashnikov.
Human infantry will have exactly the same problem- the difficulty of recognizing enemies isn't unique to robotics. (In fact, the currently active and near-future battlefield robots all use remotely-viewed cameras to detect targets, so its still a human who makes the decision)
A battlefield robot may be LESS likely to kill the wrong target for several reasons:
1) US soldiers can rarely speak Arabic. They can't talk to strangers they meet. One translator could stand by to assist the operators of 40 RC robots whenever they need to listen to someone.
2) The most important reason, by far: An AK47 cannot kill a robot. Neither can a Stinger, or a hydrogen bomb. There's no lives at risk. An infantry platoon that comes upon an armed stranger will have an inclination to shoot him out of self defense. A robot has no fear of death, and won't open fire on a dark flicker that might be a sniper, or a farmer who made a rapid grab into the back of a truck. (In jargon, a robot would operate at a lower force protection posture than DIs)
For the US military, losing millions of dollars in high-tech hardware is perfectly fine if it protects the troops. (And hey, then they need to buy more robots, which feeds the defense contractors and spurs the economy!)
Let me guess. You're not a "real programmer"?
Let me guess. You don't like to read comments?
He said "Real programmers don't have to test the whole world". You said real programmers will "re-test your product". Which is exactly in agreement, so why are you attacking?
"Don't test the whole world" means "test your product, but not every single thing that runs on the same system as your product".
Things like test scrips for perl and bzip2 are testing your product, not the whole world. Microsoft(tm) design methodology is bad because unit testing is insufficient to predict how many dependent apps a patch will break.
A critical bug in Konqueror and all of KDE becomes useless.
This gets back to the terms sproketboy used: no "commingling" in a "properly written application".
I won't go into a 10-page lecture on software engineering. But just because an application is depended on by any others doesn't mean they're comingled, or improperly written. A good component app will have a limited number of interfaces to the rest of the system (on the order of 10-200, and hopefully towards the low side).
Testing the program's correctness on those interfaces gives you a high trust that it'll work correctly in the larger system.
Microsoft(tm) IE(r) isn't like that. It doesn't have defined interfaces to the rest of the system. Its not an application which runs on the OS kernel and talks with other apps. It's source code is intermixed with much of the rest of the Windows OS. Testing every interface isn't enough to show that a new version is working right... you'd have to go through every line of code and see how it might possibly perturb Windows itself.
Compared to component-interface testing, that's a prohitably lengthy task; a combinatorical explosion of places to check.
no Kate working no editors
Again, Kate is one component, and testing that component's agreement with each of its public interfaces should be enough to verify there are no critical bugs. That only works if the components are well-separated enough. But separation leads to slowness, and Microsoft wants to be fast.
You're a bitch for thinking the USA is terrorists.
The US, like every nation with a victorious army, uses fear and terror to get what they want. Scaring enemies into cooperation by killing some of them is considered morally superior to the alternative of shooting them all.
stopping a bitch from killing thousands of people in a single day isn't terror its good
Nobody has killed more people (130,000) with a single bomb than a US pilot.