Re:That's only partly true
on
Debugging
·
· Score: 1
Testers get bored and become dumb if you make them do it two dozen times, thus they miss the higher-level bugs.
Why do you use humans to do what computers should be doing? Tests should be automated as far as possible, so that mindlessly repetitive testing is cheap and accurate. Obviously humans are still useful for creative prodding of the UI, but any problems they find that way should get automated tests.
Another thing that's totally wrong with testing is breaking down the application by areas in any kind of strict fashion, especially with things like security or performance. This fucks up the most important thing in the app - namely, how well the pieces work together. Individual features may be OK, but there's a lot of unexplored garbage at the seams, and sooner or later seams start to break revealing fundamental problems and total lack of testing in feature interaction.
That's why you do system testing too - unless what you're saying is that system testing still tends to test one area at a time.
If the emotial trauma is too painful for the victim to bear thinking about the incident, how can they come to terms with it until they have been somewhat desensitised?
To clarify further: pricing as a form of communication, and by pricing the feature of digital TV support at $300 the manufacturers are saying "digital TV is worth an extra $300 and costs a substantial fraction of that to produce". That implicit statement is what I was calling "bullshit".
The TGI of Paris has sentenced Mandrake, the French publisher of the Linux distribution of the same name, to pay &euro70,000 to the American companies Hearst Holdings and King Feature Syndicate, owners of the trademark "Mandrake le magicien" and publishers of the comic of the same name. The latter had brought this to court in France for trademark dilution. The court also forbade the French company to continue using its name and demanded that Mandrake transfer its domain names to the two American companies - a verdict which could be a deadly blow to the French company whose business resides solely on its eponymous distribution.
For the moment, Mandrake has appealed, thus suspending the judgement and retaining its trademark and domain names.
Let us remember that an earlier judgement - concerning the logo - had been in favour of the American companies. The French company has already had to revise its copy.
This kind of installation process may be acceptable for the tech-savvy user but how is Aunt Tilly supposed to deal with this? The average user would never be able to install Windows with the current installer. Windows will never be successful in the marketplace until it gets an easy-to-use installer.
When parts of the disk are found to be bad at the factory, they are marked as such. You're talking about dynamic defect management, which is not what the page you linked to describes. Magneto-optical drives and the new Mount Rainier CD/DVD re-writers do this, but I've not heard of it being implemented in hard drives.
$300 for a tuner is bullshit. Maybe it costs that much extra if the development cost is spread over a few tens of thousand of sets. If it's spread across millions of sets (as it will be) the cost goes down a long way. Here in the UK an external digital STB (with its own case, power supply, remote, etc.) costs 50 ($90). The cost of the tuner itself must be a tiny fraction of that. Now I know you have a different digital system in the US but I doubt that it's so fundamentally different that it will require much more expensive electronics.
Telcos lie. Those "taxes and fees" are inflated on the bill (just like the extra charge for touch-tone dialling I hear some are being charged) to deflect criticism of the high price of local telephone service in the US. Every business has to pay some taxes; they're just a cost of doing business.
libjpeg has a quite explicit requirement for credit in documentation:
If only executable code is distributed, then the accompanying
documentation must state that "this software is based in part on the work of
the Independent JPEG Group".
Why is it OK to use this in GPL software? (libjpeg is ubiquitous and it appears that all GNOME applications are linked indirectly with libjpeg.)
Cinemas project *the entire image* during the whole frame-time.
No, they blink out once or twice per frame so each frame is shown 2 or 3 times over.
TV only projects a very small part of the image at the same time, and relies on afterglow of the projected area to make the image appear.
Normal CRTs rely on persistence of vision, the same as cinema projectors. The image on the screen fades very quickly. It is possible to make CRTs which fade slowly but this is generally undesirable since it results in a kind of motion blur (commonly seen on LCDs).
Your eyes are only capable of seeing 20-25 fps.
Nonsense. That's just a minimal rate to achieve the illusion of smooth motion. (Cheaply made cartoons often use a frame rate of 12.5 or 15 Hz but they do tend to look a bit jerky.)
They are? Then I don't know what the problem is. (The specific problematic class I am aware of is Point.) Maybe it's the fact that you need to call methods to get to the attributes because Java doesn't allow for read-only attributes. Again, that's not a fault of OO but a limitation of Java.
It's not an either/or situation. A colour is logically both a value and an object (it has attributes of its own). A bignum needn't necessarily be an object but I don't see why it shouldn't be. I suspect what you really dislike is the existence of mutators on some of these classes. I agree that colours and bignums should be immutable values but I don't think that means they shouldn't be objects.
UIQ and Series 60 are basically GUI layers so it's not at all correct to call them OSes. You're right that SymbianOS isn't real-time so it does need to run alongside or under an RTOS. SymbianOS can run as a task under NOS (Nokia OS) or on a separate processor from the GSM (or other protocol) stack. I don't remember which handsets do which.
Most parts of SymbianOS can work as a layer over Windows (the "WINS" build), which allows you to do most of your development and debugging on a PC using a special version of CodeWarrior. (Previously Symbian supported Visual C++; I don't know whether they still do.) Of course the real thing works a bit differently; for example, it runs applications in separate processes with memory protection.
In some ways it should be easier to debug for PalmOS since you can just run the app in POSE and attach gdb to it.
In Windows all anyone has to do is double-click the attachment to execute it,
This was once true, but recent versions of Outlook, Outlook Express and Eudora do not allow this. The problem is the huge installed base of older, more trusting versions.
Another way to take over a Windows machine from an unprivileged account is to insert code into a privileged processe using window messages. Microsoft advises against creating windows in service processes and provides the "allow service to interact with desktop" flag to force any windows they do create (e.g. error message-boxes) to be walled-off from the real desktop. However, some of the default services break this rule, the messenger service being a prime example. This may have been fixed in Windows Server 2003 though.
Slashdot Mirror
Why do you use humans to do what computers should be doing? Tests should be automated as far as possible, so that mindlessly repetitive testing is cheap and accurate. Obviously humans are still useful for creative prodding of the UI, but any problems they find that way should get automated tests.
That's why you do system testing too - unless what you're saying is that system testing still tends to test one area at a time.
If the emotial trauma is too painful for the victim to bear thinking about the incident, how can they come to terms with it until they have been somewhat desensitised?
VirtualProtect()
No, he wasn't looking at the slush pile. He was giving a talk and responding to the common belief that 90% of science fiction is crud.
To clarify further: pricing as a form of communication, and by pricing the feature of digital TV support at $300 the manufacturers are saying "digital TV is worth an extra $300 and costs a substantial fraction of that to produce". That implicit statement is what I was calling "bullshit".
I meant that the price is unreasonable, not that it's incorrect.
Mandrake has to change its name
The TGI of Paris has sentenced Mandrake, the French publisher of the Linux distribution of the same name, to pay &euro70,000 to the American companies Hearst Holdings and King Feature Syndicate, owners of the trademark "Mandrake le magicien" and publishers of the comic of the same name. The latter had brought this to court in France for trademark dilution. The court also forbade the French company to continue using its name and demanded that Mandrake transfer its domain names to the two American companies - a verdict which could be a deadly blow to the French company whose business resides solely on its eponymous distribution.
For the moment, Mandrake has appealed, thus suspending the judgement and retaining its trademark and domain names.
Let us remember that an earlier judgement - concerning the logo - had been in favour of the American companies. The French company has already had to revise its copy.
No, dogmatic people hate that, and there are plenty of them among liberals, conservatives and any other political group.
This kind of installation process may be acceptable for the tech-savvy user but how is Aunt Tilly supposed to deal with this? The average user would never be able to install Windows with the current installer. Windows will never be successful in the marketplace until it gets an easy-to-use installer.
No, hang on, that's not right is it?
When parts of the disk are found to be bad at the factory, they are marked as such. You're talking about dynamic defect management, which is not what the page you linked to describes. Magneto-optical drives and the new Mount Rainier CD/DVD re-writers do this, but I've not heard of it being implemented in hard drives.
$300 for a tuner is bullshit. Maybe it costs that much extra if the development cost is spread over a few tens of thousand of sets. If it's spread across millions of sets (as it will be) the cost goes down a long way. Here in the UK an external digital STB (with its own case, power supply, remote, etc.) costs 50 ($90). The cost of the tuner itself must be a tiny fraction of that. Now I know you have a different digital system in the US but I doubt that it's so fundamentally different that it will require much more expensive electronics.
Telcos lie. Those "taxes and fees" are inflated on the bill (just like the extra charge for touch-tone dialling I hear some are being charged) to deflect criticism of the high price of local telephone service in the US. Every business has to pay some taxes; they're just a cost of doing business.
libjpeg has a quite explicit requirement for credit in documentation:
Why is it OK to use this in GPL software? (libjpeg is ubiquitous and it appears that all GNOME applications are linked indirectly with libjpeg.)
No, they blink out once or twice per frame so each frame is shown 2 or 3 times over.
Normal CRTs rely on persistence of vision, the same as cinema projectors. The image on the screen fades very quickly. It is possible to make CRTs which fade slowly but this is generally undesirable since it results in a kind of motion blur (commonly seen on LCDs).
Nonsense. That's just a minimal rate to achieve the illusion of smooth motion. (Cheaply made cartoons often use a frame rate of 12.5 or 15 Hz but they do tend to look a bit jerky.)
They are? Then I don't know what the problem is. (The specific problematic class I am aware of is Point.) Maybe it's the fact that you need to call methods to get to the attributes because Java doesn't allow for read-only attributes. Again, that's not a fault of OO but a limitation of Java.
It's not an either/or situation. A colour is logically both a value and an object (it has attributes of its own). A bignum needn't necessarily be an object but I don't see why it shouldn't be. I suspect what you really dislike is the existence of mutators on some of these classes. I agree that colours and bignums should be immutable values but I don't think that means they shouldn't be objects.
Britain adopted the American system a while ago.
UIQ and Series 60 are basically GUI layers so it's not at all correct to call them OSes. You're right that SymbianOS isn't real-time so it does need to run alongside or under an RTOS. SymbianOS can run as a task under NOS (Nokia OS) or on a separate processor from the GSM (or other protocol) stack. I don't remember which handsets do which.
Maybe some of them are dial-up users who keep redialling and getting new addresses because "my connection's slow!".
Those are examples of successful Indian branches of US companies, not examples of outsourcing.
Most parts of SymbianOS can work as a layer over Windows (the "WINS" build), which allows you to do most of your development and debugging on a PC using a special version of CodeWarrior. (Previously Symbian supported Visual C++; I don't know whether they still do.) Of course the real thing works a bit differently; for example, it runs applications in separate processes with memory protection.
In some ways it should be easier to debug for PalmOS since you can just run the app in POSE and attach gdb to it.
My impression from watching the news is that sensationalism is spiralling out of control.
This was once true, but recent versions of Outlook, Outlook Express and Eudora do not allow this. The problem is the huge installed base of older, more trusting versions.
That doesn't mean the local root vulnerability isn't there.
Another way to take over a Windows machine from an unprivileged account is to insert code into a privileged processe using window messages. Microsoft advises against creating windows in service processes and provides the "allow service to interact with desktop" flag to force any windows they do create (e.g. error message-boxes) to be walled-off from the real desktop. However, some of the default services break this rule, the messenger service being a prime example. This may have been fixed in Windows Server 2003 though.