Then I suggest you write a single regression test for all of your projects since any arbitrary regression test can meet your minimum criteria of "regardless of... whether the tests reflect what the code is supposed to do".
I'm not sure what your point is. Linux uses different abstractions then Windows but they're still abstractions and have little to do with anything outside the OS itself.
"guess I should have saved it for Troll Tuesday, but really, experience with today's version Microsoft Office is irrelent to what most students will be using in the workforce."
You have to admit, however, that today's version of MS Office is far more relevent to future versions of MS Office than any of today's OSS tools will be.
The real issue is whether computers are relevent at all to the future of the average student. POS systems are really just electronic cash registers from the operators point of view. They just allow cashiers to have fewer skills then they would need using the old systems.
".. which thanks to the gross missmanagement by the Clintoon administration..."
Now I see where you are coming from. I can't take anyone seriously who blames 9/11 on Clinton without at least assigning some responsibility to the man who was in office when the attack occurred, so I'll just stop here.
Well, Linux isn't going to do a better job of exposing kids to "the realities behind how things work" then Windows would, unless you believe there's a command prompt blinking inside your microprocessor.
"But what we're talking about is legacy code, where you probably don't even have requirements, or the requirements are so out of date that they're useless."
I don't see any reason why legacy code is less likely to have requirements than new code. In any case, if the requirements are not updated with the code, than you have a flaw in your development process.
You can certainly peform regression tests even if your process is flawed in other ways, I'm not disputing that.
"Again - are you being purposefully obtuse?"
Perhaps being obtuse isn't really what you want to accuse me of. Why don't you double-check the definition.
"How about, "Apple bought some ideas from Xerox for millions in cash and stock?""
This is a common myth on Slashdot. Xerox never licensed their GUI technology to Apple. That's why they sued Apple for violation of copyright.
As I've described before, it was common in those days for a company to sue another over the "look and feel" of the product. The theory was that copying the behavior of a product violated the copyright.
This was exactly the same argument Apple made when trying to sue MS. Neither Xerox nor Apple could make their case.
"Are you aware that there was a national id during WWII, which although not linked thru an electronic database did exactly the same function albeit with days not milliseconds as the cycle time."
I assume your aware of the illegal imprisonment of japanese men, woman, and children in WWII as well (including George Takei and his family).
Lots of stuff happened during WWII and not all of it was good. So I don't see how the fact that a national ID was used in WWII has anything to do with whether it's a good or bad idea.
"Looking at the code to see what it is supposed to do is silly."
Well, there are times when looking at the code is useful but I never claimed you needed to do that to test it. All I'm saying is that a test should be devised to determine if its requirements are met.
"What if it segfaults for some input? Was it supposed too? I've seen code that was supposed to seg fault!"
As a former Atari 2600 programmer, I can appreciate the idea that correct behavior may be quite unconventional (such as performing an indexed write to a scratch location in order to maintain consistent scan line timing without triggering a repositioning of a "player"), but the point (as I think you'll agree) is to confirm the appropriate behavior regardless of how it is achieved.
I'm not a big fan of working around bugs instead of fixing them, however. So I think the old behavior should also be the correct behavior.
No. It still does exactly what it "says" it does, it just "says" something different after the code has been modified.
Regression testing can be very handy as long as the tests reflect what the code is supposed to do. Otherwise passing the regression simply means that your code is consistently failing to achieve its requirements.
At least some of the hijackers were known by the government to be in the US and known to be associates of Bin Laden and yet they were able to do things like take flying lessons. Even when the instructors notified the government of their suspicions, nothing was done.
The idea that a high-tech license would have prevented these individuals from carrying out their objective is pure speculation and assumes a level of government competancy that all the actual evidence suggests did not exist.
"OpenOffice.org is both an office suite and an open-source project. Compatible with all other major office suites, the product is free to download, use, and distribute."
If you claim your product is compatable and it isn't, the fault is 100% your own.
"The mainframe makers had their lunch eaten by the minicomputer makers. The minicomputer makers had their lunch eaten by the PC makers. Now the PC makers are going to see their lunch eaten by the consumer electronic makers."
I used to believe that too, but nearly 25 years after the introduction of the PC, mainframes continue to sell.
It may be true for chip makers that the winner will be the one who can produce a processor that works over the widest range of applications, but at the system level you can't ignore ergonomics and suitability.
A small hand-held consumer computer will never replace a general purpose machine like the PC. The size of the human hand, the need for a large display and other factors aren't going to change.
"and analogous situations (e.g. A Private Little War) to help put current issues into perspective."
"I don't want to watch Sci-fi to put current events into perspective. That is what The Daily Show is for"
Yes, I really enjoyed those early Daily Show programs in the late 1960s. Those episodes where Stuart was joking about farts with his other little friends really put the Vietnam War into perspective.
Well, if you want to talk about doing a musical within a non-musical series you can go back at least as far as "Moonlighting". It's not as if Xena was breaking any new ground there.
I really enjoyed watching Xena, but it's really not in the same class as Buffy or Angel.
Well you can certainly reverse some jokes as you did the one I told, but if the punch line doesn't ring true (as was the case with your reversal IMHO) the result is neither funny or insightful.
Actually, Sun first sued MS before MS stopped updating Java.
Sun decided that allowing Windows programmers to create non-portable Java applications optimized for Windows was more onerous than losing the opportunity to have Java be a significant development language for Windows. IMHO that was a strategic error.
Slashdot Mirror
Then I suggest you write a single regression test for all of your projects since any arbitrary regression test can meet your minimum criteria of "regardless of ... whether the tests reflect what the code is supposed to do".
I'm not sure what your point is. Linux uses different abstractions then Windows but they're still abstractions and have little to do with anything outside the OS itself.
"guess I should have saved it for Troll Tuesday, but really, experience with today's version Microsoft Office is irrelent to what most students will be using in the workforce."
You have to admit, however, that today's version of MS Office is far more relevent to future versions of MS Office than any of today's OSS tools will be.
The real issue is whether computers are relevent at all to the future of the average student. POS systems are really just electronic cash registers from the operators point of view. They just allow cashiers to have fewer skills then they would need using the old systems.
"Besides, how much "Microsoft Experience" did *you* need to be able to say "Do you want fries with that?"
I love the smell of an Ad Hominem argument in the morning.
MS never does anything out of spite. There's no profit in it. (Isn't that the 300th Rule of Acquistion?)
" .. which thanks to the gross missmanagement by the Clintoon administration ..."
Now I see where you are coming from. I can't take anyone seriously who blames 9/11 on Clinton without at least assigning some responsibility to the man who was in office when the attack occurred, so I'll just stop here.
Well, Linux isn't going to do a better job of exposing kids to "the realities behind how things work" then Windows would, unless you believe there's a command prompt blinking inside your microprocessor.
All OS's hide complexity, that's their purpose.
I've never objected to regression tests in this thread, apparently you haven't been paying attention.
"But what we're talking about is legacy code, where you probably don't even have requirements, or the requirements are so out of date that they're useless."
I don't see any reason why legacy code is less likely to have requirements than new code. In any case, if the requirements are not updated with the code, than you have a flaw in your development process.
You can certainly peform regression tests even if your process is flawed in other ways, I'm not disputing that.
"Again - are you being purposefully obtuse?"
Perhaps being obtuse isn't really what you want to accuse me of. Why don't you double-check the definition.
"How about, "Apple bought some ideas from Xerox for millions in cash and stock?""
This is a common myth on Slashdot. Xerox never licensed their GUI technology to Apple. That's why they sued Apple for violation of copyright.
As I've described before, it was common in those days for a company to sue another over the "look and feel" of the product. The theory was that copying the behavior of a product violated the copyright.
This was exactly the same argument Apple made when trying to sue MS. Neither Xerox nor Apple could make their case.
"Are you aware that there was a national id during WWII, which although not linked thru an electronic database did exactly the same function albeit with days not milliseconds as the cycle time."
I assume your aware of the illegal imprisonment of japanese men, woman, and children in WWII as well (including George Takei and his family).
Lots of stuff happened during WWII and not all of it was good. So I don't see how the fact that a national ID was used in WWII has anything to do with whether it's a good or bad idea.
"Looking at the code to see what it is supposed to do is silly."
Well, there are times when looking at the code is useful but I never claimed you needed to do that to test it. All I'm saying is that a test should be devised to determine if its requirements are met.
"What if it segfaults for some input? Was it supposed too? I've seen code that was supposed to seg fault!"
As a former Atari 2600 programmer, I can appreciate the idea that correct behavior may be quite unconventional (such as performing an indexed write to a scratch location in order to maintain consistent scan line timing without triggering a repositioning of a "player"), but the point (as I think you'll agree) is to confirm the appropriate behavior regardless of how it is achieved.
I'm not a big fan of working around bugs instead of fixing them, however. So I think the old behavior should also be the correct behavior.
If the old behavior was what you "wanted" but different behavior then what your requirements state, you have problems that testing will not solve.
"Until someone changes it."
No. It still does exactly what it "says" it does, it just "says" something different after the code has been modified.
Regression testing can be very handy as long as the tests reflect what the code is supposed to do. Otherwise passing the regression simply means that your code is consistently failing to achieve its requirements.
At least some of the hijackers were known by the government to be in the US and known to be associates of Bin Laden and yet they were able to do things like take flying lessons. Even when the instructors notified the government of their suspicions, nothing was done.
The idea that a high-tech license would have prevented these individuals from carrying out their objective is pure speculation and assumes a level of government competancy that all the actual evidence suggests did not exist.
From the top of the OpenOffice.org website:
"OpenOffice.org is both an office suite and an open-source project. Compatible with all other major office suites, the product is free to download, use, and distribute."
If you claim your product is compatable and it isn't, the fault is 100% your own.
That might be a resonable trade-off if it really worked.
It's interesting to me that most of the security changes that have been made since 9/11 have a common property: they wouldn't have prevented it.
Code always does what it says it does. The problem is determining what the code should be doing and testing for that.
Rotoscoping
"The mainframe makers had their lunch eaten by the minicomputer makers. The minicomputer makers had their lunch eaten by the PC makers. Now the PC makers are going to see their lunch eaten by the consumer electronic makers."
I used to believe that too, but nearly 25 years after the introduction of the PC, mainframes continue to sell.
It may be true for chip makers that the winner will be the one who can produce a processor that works over the widest range of applications, but at the system level you can't ignore ergonomics and suitability.
A small hand-held consumer computer will never replace a general purpose machine like the PC. The size of the human hand, the need for a large display and other factors aren't going to change.
"and analogous situations (e.g. A Private Little War) to help put current issues into perspective."
"I don't want to watch Sci-fi to put current events into perspective. That is what The Daily Show is for"
Yes, I really enjoyed those early Daily Show programs in the late 1960s. Those episodes where Stuart was joking about farts with his other little friends really put the Vietnam War into perspective.
Well, if you want to talk about doing a musical within a non-musical series you can go back at least as far as "Moonlighting". It's not as if Xena was breaking any new ground there.
I really enjoyed watching Xena, but it's really not in the same class as Buffy or Angel.
"the joke works both ways as you see."
Well you can certainly reverse some jokes as you did the one I told, but if the punch line doesn't ring true (as was the case with your reversal IMHO) the result is neither funny or insightful.
I considered the possibility that you made it up, but I was giving you the benefit of the doubt.
As to why make a point with a joke, well, why not? The fact that it's an old joke implies that people have thought about this issue for a long time.
Actually, Sun first sued MS before MS stopped updating Java.
Sun decided that allowing Windows programmers to create non-portable Java applications optimized for Windows was more onerous than losing the opportunity to have Java be a significant development language for Windows. IMHO that was a strategic error.