Australia was not at risk until just over two years into the Second World War (it started in 1939 remember, not in 1941). I guess they and Canada went in early because they still had close ties with the UK. India was still part of the British Empire so it didn't have a choice.
You could argue that Britain and France went in before they were directly at risk. They only declared war on Germany because Germany invaded Poland.
That's a complete lie. Word loads when you click it. It then calls the (undocumented) Win32 API function "void EmulateWordEx (DWORD WordVersion, UINT CrashesPerHour)".
The COM interface standard is open. Anybody can write COM objects, the interface is well documented. The thing that isn't open is the implementation and that's what needs to be created for the Open Source community or something similar. I'm going to write it and call it the "COM Open Reusable Blob Application" or for short by it's acronym.... no err wait....
The Windows API is not a COM interface. The windows API is probably no more or less comlex than any other windowing API for Unix or otherwise. It's complexity results from the complexity of the object you are trying to manage with it. You do not need to remember thousands of numeric codes because the documentation is exceptionally good.
The X protocol proves you can manage a graphical user interface with two one dimensional byte streams a la stdin/stdout, but you can't tell me its not a complex one dimension byte stream.
I don't think inheritance is that important a feature of object oriented programming. Some of the poorest code I've seen was caused by inappropriate overuse of inheritance - "I must figure out a way of defining this class in terms of that one because I've got inheritance and I must use it." I don't say inheritance is evil, just overused.
Yes, it's sad because it means a group of programmers spent some time (hopefully their own, not the company's) writing a piece of code designed only to pump up their own egos.
It's bad because it is at least partially responsible for the famous M$ code bloat. It's bad because it's a sub-optimal solution to the spreadsheet problem (that doesn't sound like good hacking to me). It's bad because it shows a lack of responsibility on the part of the M$ engineers (what else might they have snuck in there?). It's bad because more code means more bugs. It's bad because it means their project management lacked control over the process.
The problem with your argument is the assertion that this tragedy would not have happened if everybody's communications were monitored the whole time.
Probably some well directed increases in security may have prevented this. In the UK they are reporting that security on US domestic flights is not tight enough, but the wholesale introduction of the police state is not justified IMHO.
You can make money on services. I have worked for several consultancy companies that did that quite successfully. The difficulty with open source s/w is that the customer has unrealistically low expectations of what some services cost. If you say "here's three days consulting, it'll cost you $1,000 / day" they'll accept that, but a support contract has to be dirt cheap. For instance Red Hat tries to make money by providing support services to Linux, but there's no way that they can provide a *useful* suport service for $99 per customer. You need that $1,000/day consultant to provide a significant level of expertise so if the customer makes one call lasting two hours, there goes your margin.
Re:Not this stupid 'programming is art' BS again!
on
Software Aesthetics
·
· Score: 1
Programming is not an art, cabinet designing is not an art, architecture is not an art. I'm defining the term "artist" quite narrowly as somebody who makes objects who's primary purpose is to inspire some sort of emotional response from people. e.g. a Henry Moore sculpture has no other purpose.
Software, cabinets and buildings have a different primary purpose and subjugating that purpose to the goal of making it look nice often results in less than optimum performance in the primary purpose. e.g. an architect may decide to make his building entirely of glass for aesthetic reasons, but that might make it too hot in sunny weather. Another example would be the original design for Quicktime 4 which looked great with its brushed metal graphics, but was terrible to use (review here http://www.iarchitect.com/qtime.htm).
It depends on your criteria for beauty. Speaking as somebody who has spent a significant proportion of his career maintaining other people's code, to me, beauty and maintainability are pretty much the same thing.
Also, maintainability / beauty is not a property independent of the other properties of the code. It is an *effect* that arises from the software being well designed ad well coded.
I guess I'm agreeing with you.
One final point. Often the excuse for writing ugly over complex code is that it is optimised for speed, but ugly over complex code is often harder for the compiler optimiser to understand and turns out slower.
I bet you could find a few DC3s still in service out there, probably not with their original airlines though.
Re:Hate to say, sounds like a dot-bomb strategy...
on
HP Buys Compaq
·
· Score: 1
Digital decided VMS was dead (or at least not worth developing further) long before they were bought by Compaq. When they brought out the Alpha architecture and OSF1 (which was renamed Digital Unix which was renamed True 64) they decided to rationalise all their server products down to Alpha + OSF or NT.
Sad to say, I think this merger is the last ditch attempt by these two companies to stay afloat. Without it they are both finished. With it, they may still be finished.
It's something you pay for, so you have a contract with the ISP to provide an internet connection. If the ISP arbitrarily cuts you off, it is in violation of the contract and can therefore be sued. If there is a clause in the contract which says you can't spam, and you do spam, then you are in violation of the contract and the ISP is within its rights to terminate it in which case it can cut you off straight away.
In this case Monster Hut disputes that it has violated the anti-spamming condition so there is a difference of opinion on whether the contract has been violated which will get resolved through the courts. In the meantime, they should not be penalised for an unproven allegation. Personnally, I reckon they will lose.
No he did believe in quantum mechanics. He didn't believe in the Copenhagen Interpretation of the theory which said that quantum mechanics means that a particle is in a "superposition" of all possible states until you try to measure it.
That's why he invented the Schrodinger's cat experiment where you put a cat in a box with some poison gas which gets released if a certain quantum mechanical event takes place (e.g. radioactive decay of an atom). Thus the cat is in a superposition of "alive" and "dead" which is (according to Schrodinger) nonsense.
When Drumbrovsky and Brown finally performed the experiment in 1965, they opened the box to find that the cat was in a state known as "extremely angry and upset".
Interestingy, Einstein did not believe in quantum mechanics or at least some parts of it ("God does not play dice") in spite of the fact that he received the Nobel prize for his paper on the photo-electric effect which is recognised as one of the early contributions to the subject of quantum mechanics.
I don't think Darwin "found" religion on his deathbed. He was a committed Christian throughout his life and didn't see any incompatibility between his theory and Christianity. He may have got frightened at the end and said "Hey what if God really is mad at me?" in which case recanting is the logical thing to do.
I take exception to putting Newtonian Gravity in the same class as alchemy, phlogiston and creationism. Those other ideas were/are not good scientific theories. General relativity only applies a very small correction to Newtonian gravity, (so small that it is virtually impossible to observe on human scales). For example, Newtonian physics was good enough to get several men to the moon and back safely.
OS X has a great user interface, I agree, so why does the OS X version of Mozilla not follow the Aqua standards? Because the programmers are too f***ing lazy to tailor their UI to the environment in which people will be using it.
</rant> Sorry
Re:Not only the net. THe article mentions CPRM als
on
Taming the Web
·
· Score: 1
Information is *not* free (as in beer). There is always a cost involed in gathering it and publishing it even if it's only somebody else's time.
If I expend money, time and effort in writing a book, recording music, writing a computer program etc, why shouldn't I expect other people to give me something in return for using it? Of course if I choose to give it away for free, then that's fine, but it's *my* decision, not yours.
It's easy. The Linux community doesn't expect to pay for software. You can't build a profitable business on something you have to give away for free.
Focussing on the games market, there are some additional issues. There are fewer Linux users than Windows users by at least an order of magnitude, so the market is smaller. On average, Linux users are more technical than Windows users. Their idea of relaxing in front of the PC might be to *write* the game, not play it. Linux users have the ability to boot their PC into Windows if they want to play a game (not really any different from using a console). This is what I do, whatever else you say about Windows, the variety and quality of games is infinitely better than Linux.
All of the above makes the Linux games market miniscule which means that at the same unit price it's harder to cover your dev costs.
SUVs are unsafe for the drivers as well. The reason for this is that some of the safety features modern cars have are impractical for offroaders which are therefore exempt.
Loud noise from the CD into the amplifier causes the amplifier to overload the speakers. I imagine that with any properly matched set of speakers and amplifier, it's not possible. The loudness of the noise and distortion would cause you to turn down the volume before you got close to destroying anything.
We did once connect a small transistor radio speaker to the output of a guitar amplifier. We got a great distorted sound for about 1 minute until the speaker cone split.
1 If you have a document which starts with a list of 95 things, people will read the first five, get bored and return to Slashdot to complain about it.
If the engineer at XYZ tells the press that they are having problems with the Linux drivers for their latest video cards, the headline will come out:
XYZ cannot make Linux work with its video cards. Microsoft spokesman states "well that's the problem with these two bit amateur operating systems. Windows is the one true OS capable of supporting all of todays hardware."
Yes, you're right, they just moved the front end servers to Win2K and IIS. Event then they are running some software to emulate a FreeBSD environment so that their cgi scripts etc still work.
Unfortunately, the whole point of Sircam is to attack the web front end servers, so the back end is irrelevant.
More seriously, people say "just apply the patch" but for a site like Hotmail that is a non-trivial exercise. You have to test the patch extensively in your environment to make sure it doesn't break anything. Can you imagine the smugness of the Slashdot community if they had applied the patch and some subtle bug in it brought down every web server in Hotmail?
Slashdot Mirror
Err no. They don't have accents at all. It's you Americans that have accents and can't spell proper:)
Australia was not at risk until just over two years into the Second World War (it started in 1939 remember, not in 1941). I guess they and Canada went in early because they still had close ties with the UK. India was still part of the British Empire so it didn't have a choice.
You could argue that Britain and France went in before they were directly at risk. They only declared war on Germany because Germany invaded Poland.
That's a complete lie. Word loads when you click it. It then calls the (undocumented) Win32 API function "void EmulateWordEx (DWORD WordVersion, UINT CrashesPerHour)".
The COM interface standard is open. Anybody can write COM objects, the interface is well documented. The thing that isn't open is the implementation and that's what needs to be created for the Open Source community or something similar. I'm going to write it and call it the "COM Open Reusable Blob Application" or for short by it's acronym.... no err wait....
The Windows API is not a COM interface. The windows API is probably no more or less comlex than any other windowing API for Unix or otherwise. It's complexity results from the complexity of the object you are trying to manage with it. You do not need to remember thousands of numeric codes because the documentation is exceptionally good.
The X protocol proves you can manage a graphical user interface with two one dimensional byte streams a la stdin/stdout, but you can't tell me its not a complex one dimension byte stream.
I don't think inheritance is that important a feature of object oriented programming. Some of the poorest code I've seen was caused by inappropriate overuse of inheritance - "I must figure out a way of defining this class in terms of that one because I've got inheritance and I must use it." I don't say inheritance is evil, just overused.
Yes, it's sad because it means a group of programmers spent some time (hopefully their own, not the company's) writing a piece of code designed only to pump up their own egos.
It's bad because it is at least partially responsible for the famous M$ code bloat. It's bad because it's a sub-optimal solution to the spreadsheet problem (that doesn't sound like good hacking to me). It's bad because it shows a lack of responsibility on the part of the M$ engineers (what else might they have snuck in there?). It's bad because more code means more bugs. It's bad because it means their project management lacked control over the process.
The problem with your argument is the assertion that this tragedy would not have happened if everybody's communications were monitored the whole time.
Probably some well directed increases in security may have prevented this. In the UK they are reporting that security on US domestic flights is not tight enough, but the wholesale introduction of the police state is not justified IMHO.
You can make money on services. I have worked for several consultancy companies that did that quite successfully. The difficulty with open source s/w is that the customer has unrealistically low expectations of what some services cost. If you say "here's three days consulting, it'll cost you $1,000 / day" they'll accept that, but a support contract has to be dirt cheap. For instance Red Hat tries to make money by providing support services to Linux, but there's no way that they can provide a *useful* suport service for $99 per customer. You need that $1,000/day consultant to provide a significant level of expertise so if the customer makes one call lasting two hours, there goes your margin.
Programming is not an art, cabinet designing is not an art, architecture is not an art. I'm defining the term "artist" quite narrowly as somebody who makes objects who's primary purpose is to inspire some sort of emotional response from people. e.g. a Henry Moore sculpture has no other purpose.
Software, cabinets and buildings have a different primary purpose and subjugating that purpose to the goal of making it look nice often results in less than optimum performance in the primary purpose. e.g. an architect may decide to make his building entirely of glass for aesthetic reasons, but that might make it too hot in sunny weather. Another example would be the original design for Quicktime 4 which looked great with its brushed metal graphics, but was terrible to use (review here http://www.iarchitect.com/qtime.htm).
It depends on your criteria for beauty. Speaking as somebody who has spent a significant proportion of his career maintaining other people's code, to me, beauty and maintainability are pretty much the same thing.
Also, maintainability / beauty is not a property independent of the other properties of the code. It is an *effect* that arises from the software being well designed ad well coded.
I guess I'm agreeing with you.
One final point. Often the excuse for writing ugly over complex code is that it is optimised for speed, but ugly over complex code is often harder for the compiler optimiser to understand and turns out slower.
I bet you could find a few DC3s still in service out there, probably not with their original airlines though.
Digital decided VMS was dead (or at least not worth developing further) long before they were bought by Compaq. When they brought out the Alpha architecture and OSF1 (which was renamed Digital Unix which was renamed True 64) they decided to rationalise all their server products down to Alpha + OSF or NT.
Sad to say, I think this merger is the last ditch attempt by these two companies to stay afloat. Without it they are both finished. With it, they may still be finished.
It's something you pay for, so you have a contract with the ISP to provide an internet connection. If the ISP arbitrarily cuts you off, it is in violation of the contract and can therefore be sued. If there is a clause in the contract which says you can't spam, and you do spam, then you are in violation of the contract and the ISP is within its rights to terminate it in which case it can cut you off straight away.
In this case Monster Hut disputes that it has violated the anti-spamming condition so there is a difference of opinion on whether the contract has been violated which will get resolved through the courts. In the meantime, they should not be penalised for an unproven allegation. Personnally, I reckon they will lose.
It's like that Monty Python Sketch.
One entrant to the "Invent a derrogatory term for the Belgians" competition said he couldn't come up with anything more derrogatory than "Belgians".
Apologies to any Belgians. It wasn't me, it was Michael Palin.
No he did believe in quantum mechanics. He didn't believe in the Copenhagen Interpretation of the theory which said that quantum mechanics means that a particle is in a "superposition" of all possible states until you try to measure it.
That's why he invented the Schrodinger's cat experiment where you put a cat in a box with some poison gas which gets released if a certain quantum mechanical event takes place (e.g. radioactive decay of an atom). Thus the cat is in a superposition of "alive" and "dead" which is (according to Schrodinger) nonsense.
When Drumbrovsky and Brown finally performed the experiment in 1965, they opened the box to find that the cat was in a state known as "extremely angry and upset".
Interestingy, Einstein did not believe in quantum mechanics or at least some parts of it ("God does not play dice") in spite of the fact that he received the Nobel prize for his paper on the photo-electric effect which is recognised as one of the early contributions to the subject of quantum mechanics.
I don't think Darwin "found" religion on his deathbed. He was a committed Christian throughout his life and didn't see any incompatibility between his theory and Christianity. He may have got frightened at the end and said "Hey what if God really is mad at me?" in which case recanting is the logical thing to do.
I take exception to putting Newtonian Gravity in the same class as alchemy, phlogiston and creationism. Those other ideas were/are not good scientific theories. General relativity only applies a very small correction to Newtonian gravity, (so small that it is virtually impossible to observe on human scales). For example, Newtonian physics was good enough to get several men to the moon and back safely.
OS X has a great user interface, I agree, so why does the OS X version of Mozilla not follow the Aqua standards? Because the programmers are too f***ing lazy to tailor their UI to the environment in which people will be using it. </rant> Sorry
Information is *not* free (as in beer). There is always a cost involed in gathering it and publishing it even if it's only somebody else's time.
If I expend money, time and effort in writing a book, recording music, writing a computer program etc, why shouldn't I expect other people to give me something in return for using it? Of course if I choose to give it away for free, then that's fine, but it's *my* decision, not yours.
It's easy. The Linux community doesn't expect to pay for software. You can't build a profitable business on something you have to give away for free.
Focussing on the games market, there are some additional issues. There are fewer Linux users than Windows users by at least an order of magnitude, so the market is smaller. On average, Linux users are more technical than Windows users. Their idea of relaxing in front of the PC might be to *write* the game, not play it. Linux users have the ability to boot their PC into Windows if they want to play a game (not really any different from using a console). This is what I do, whatever else you say about Windows, the variety and quality of games is infinitely better than Linux.
All of the above makes the Linux games market miniscule which means that at the same unit price it's harder to cover your dev costs.
SUVs are unsafe for the drivers as well. The reason for this is that some of the safety features modern cars have are impractical for offroaders which are therefore exempt.
Loud noise from the CD into the amplifier causes the amplifier to overload the speakers. I imagine that with any properly matched set of speakers and amplifier, it's not possible. The loudness of the noise and distortion would cause you to turn down the volume before you got close to destroying anything.
We did once connect a small transistor radio speaker to the output of a guitar amplifier. We got a great distorted sound for about 1 minute until the speaker cone split.
Jeremy's one thesis:
1 If you have a document which starts with a list of 95 things, people will read the first five, get bored and return to Slashdot to complain about it.
If the engineer at XYZ tells the press that they are having problems with the Linux drivers for their latest video cards, the headline will come out:
XYZ cannot make Linux work with its video cards. Microsoft spokesman states "well that's the problem with these two bit amateur operating systems. Windows is the one true OS capable of supporting all of todays hardware."
Yes, you're right, they just moved the front end servers to Win2K and IIS. Event then they are running some software to emulate a FreeBSD environment so that their cgi scripts etc still work.
Unfortunately, the whole point of Sircam is to attack the web front end servers, so the back end is irrelevant.
More seriously, people say "just apply the patch" but for a site like Hotmail that is a non-trivial exercise. You have to test the patch extensively in your environment to make sure it doesn't break anything. Can you imagine the smugness of the Slashdot community if they had applied the patch and some subtle bug in it brought down every web server in Hotmail?