Paradigms per se have zilch to do with memory mangement. A fully garbage-collecting, memory-safe environment can be implemented beneath any language.
That's debatable. OO is a model that promotes a very simple dynamic memory allocation mechanism: the 'new' operator. This operator was designed to be combined with an automatic garbage collector. [This is based on observing Smalltalk - I'm not sure if Simula used "new"/had a garbage collector]
That C++ doesn't do this is somewhat irrelevant as C++ is clearly a "multi-paradigm" language, not a strictly OO one.
It is certainly an improvement upon top-down, procedural design in terms of flexibility and maintainability.
The jury is out on that.
Perhaps... I think that empirical evidence shows that software productivity and quality has been modestly improving over the last 20 years. [Given that the whole economy's productivity has been improving around 3% per year, mainly caused by information technology, according to Alan Greenspan]. The improvement in programmer productivity may not be caused by OO... but there certainly would be a correlation, just as there would be a correletion among other "paradigm waves", such as the structured programming movement in the 70's.
On the contrary, few programmers are actually versed enough with how to _handle_ the power of a multi-paradigm language that they often wind up being more productive in a more constrained environment.
This notion has its roots in philosophy where "true freedom" is actually a misnomer: we require some constraints in order to make sense of the world and to be productive in it.
So yes, there are situations where OO is not the ideal paradigm. Functional languages are a great alternative and are very expressive.
I think it's wrong to conclude that because OO is not a panacea it will be laughed out of the office. It is certainly an improvement upon top-down, procedural design in terms of flexibility and maintainability.
OO doesn't replace the need for good programmers. Shoddy programmers will ALWAYS make shoddy programs, no matter the paradigm.
HOWEVER, I would say that the constraints imposed on an OO language like Java or Smalltalk would result in a program that is less shoddy in the area of memory leaks and segmentation faults.
This, at the very least, is an quantitative improvement, by enforcing a paradigm of dynamic memory allocation (the "new" operator) and removing raw pointers from the language.
OO is not the perfect paradigm, nor should it be, or we might as well stop all software engineering research right now. It would be hard to argue that there is any ONE TRUE paradigm at all. I don't think anyone is claiming that it is the end of all software paradigms. It is, however, a useful and productive paradigm, and is worthy of widespread use. Which, as a matter of fact, is the case.
I would say the reason such opinions are read are because the legal field is, above all, an ongoing conversation among rational people.
Just because one is not a lawyer does not mean one cannot have a rational argument about a legal issue. To assume otherwise is just as preposterous as to assume that only programmers can have an opinion on whether a piece of software is of high quality or not.
The more detailed and low-level the argument, the more professional expertise becomes an issue. Hence "IANAL" becomes a necessary disclaimer.
Re:Everybody take a breather
on
Microsoft Loses
·
· Score: 1
To me, the prospect that all this activity would have vanished if Microsoft had not established dominance is ludicrous.
I wouldn't go so far as to say the activity would have "vanished". All I would say is that it is extremely difficult to claim that it would be better without Microsoft. My reasoning for this is that new ventures are under tremendous duress during their growth years: they have to perform, they have to compete with the big boys, and they have to do it without a ton of money. Very often many small companies poised for hypergrowth are in a vicious circle: they don't have enough time to get the management talent needed to handle this growth, and hence they can't grow without major screwups. Netscape is a good example of this problem in effect.
Regarding your last paragraph: I don't know how many times this needs to be said before it sinks in. This whole case was brought up because Microsoft illegally removed true choice from the consumer. PEOPLE may have bought Microsoft, but they did so because they didn't have a real alternative. So the "market" didn't decide anything. I believe that, if it had, we would have a for more stable economy with our far more stable PC's.
I disagree.
The court found that Microsoft was guilty of propagating its monopoly through anticompetitive means, and attempting to gain monopoly power in the web browser industry through anti-competitive means. It found that Microsoft was NOT guilty of anti-competitve marketing agreements.
This ruling, to me, suggests that Microsoft's abuse of monopoly power really began around 1996. The consent decree of 1994/95 was designed to address the problems with Microsoft's former licencing practices. These practices were *not* deemed anti-competitive at the time.
Furthermore, Microsoft's licencing practices in the 80's were done at a time when there WAS market choice. You still could choose to get a Mac, or an Amiga in 1985. There was no 90% monopoly then. That the market decided to go with the herd, as it usually does, is disappointing, but not anti-competitive.
Even more so in 1995 - the Windows 95 launch was a moderate success, despite a substantial marketing effort on the part of IBM's OS/2 since 1994. The market did swing towards OS/2 for a while, before the Win95 release. But within one and a half years after Win95 came out, shrink-wrapped sales, not pre-loaded ones, overcame the entire OS/2 install base. The market decided again.
This case was about punishing Microsoft for attempting to remove choice in the market through the browser bundling, through its threats of dropping the Macintosh MS Office, and through its exclusive arrangements with ISP's. It is not punishing Microsoft for making shoddy, but popular operating systems.
Re:Everybody take a breather
on
Microsoft Loses
·
· Score: 1
Microsoft is responsible for U.S. economic growth in one important way: profits. Lots of them, and also profits from companies that support or are complements to Microsoft's products.
That totals up to tens of billions of dollars in GDP every year.
Another way: wealth. Almost every pension fund or mutual fund owns some sort of position in Microsoft. MANY ordinary people have gotten wealthier because of their success.
Another way: providing a de facto operating system standard. We may be outgrowing this now, but it was indispensible in allowing the software industry to grow. Stable infrastructure is a necessity for innovation.
Yet another way: Minor innovation. No, Microsoft is not a big innovator. They do lots of little things, however. Little things that are customer requests. Or are features the competition has. Incorporating any of these ideas into their products is an innovation because it exposes those ideas to a wider audience, continuing the positive feedback loop by attracting more customers to Microsoft.
Side note: If you don't think this is innovation, you probably have a double standard here. (Think about your definition of innovation as applied to MIcrosoft, and then apply it to Gnome or RedHat, or any other Linux distro vendor. And then tell me with a straight face that these companies are better innovators.)
The damage Microsoft has caused is quite intangible, at best. It would be difficult to claim that the amount of profits Microsoft has would be small compared to its potential competitors if they were allowed to thrive, given at how risky a business venture really is. However, the damage is real, so the remedy should at least be preventive, if not retributive (how do you compensate for what you can't measure?).
I think in the end some balance is required with Microsoft. They have done some good for the computer industry in a business sence, if not a technological sence by providing a stable infrastructure. They have also abused their power and done much damage to some parts of the industry. Hence, they should be punished and prevented from doing so again.
This does NOT make them the great satan who forced their wares upon us. Never forget that it is PEOPLE who bought Microsoft products. They could have bought Apple products or Commodore products, but they chose not to. Later they could have bought OS/2 during the Win95 upgrade but they chose not to. It's all about choice, and the market decided that Microsoft should win. And they did win, which is good. But they kept winning by illegally perpetuating their position, which is bad.
If the company is a monopoly, then I agree exclusive contracts should be a legal no-no. [And at 5pm today I think we'll find out what one judge thinks about this matter]
Otherwise, I think they should be free to do it. It's a relatively short term measure, at best.
Yes, the Star Wars trailer was only released in QuickTime. Some third parties converted it to AVI/MPEG, but the quality often wasn't kept during the conversion...
Sorenson has an exclusive licence with Apple; they can't create an AVI version of their codec.
Sorenson has MPEG-1 quality with better sound support and compression. Take a look at the Star Wars trailers, for instance.
There hasn't been any empircal study contrasting the two approaches AFAIK, but just observing the quality contrasted to file size is an educational experience.
This probably would be a textbook example of a "well-behaved" monopoly... remember.. monopoly's can be good for consumers, it just usually happens that they aren't.
one would think that IBM learned their lesson on what it means to be a "poor behaved" monopoly.
sorry folks, but I don't buy the "Star wars' CGI was same-old, same-old" argument. It was ground breaking work, and was BY FAR the most visually extraordinary movie of the year. ILM pulled off major innovations in order to get Phantom Menace out the door, and deserved the oscar.
The academy most likely wanted to give Lucas the finger. Which is somewhat understandable.
- let them be self important. we know the truth. - actors affiliate themselves with pet causes because it attracts attention to the cause and to the actor. it's a win/win situation. Yes it's a sad state of affairs when we don't notice "great causes" unless a shining smile/body/face is attached to them, but can you really blame us? It's an information glut... - sometimes it's easy to get jaded, but it's important to remember that WE are the reason hollywood is an empire. intellectual property isn't going to save them in the long term - you don't make that much money off of re-runs. It's up to us if we want to watch shlock movies.
if there's something to get jaded about, it's about wide spread cynicism, double standards and attention deficit in the general populace.
That no matter what system you devise, you're going to have the "tyranny of the ruling class". In a democracy, it's the "tyranny of the majority". Might does not make right.
The founding fathers of the U.S. knew this: no single political principle can work without a counterbalance.
(which is arguably why we have private property and free markets.)
I agree with what you're saying, so I'm not quite sure what your point is. My point was primarily that "pay per use" is not exactly the ideal pricing structure for most forms of information, so that a lot of the worrying going on about how we'll have to pay so much extra $ in the future to watch movies or listen to music is somewhat overzealous.
This is not to say that pay per use couldn't work, just that pay per use with extreme fair use restrictions would have a hard time working.
Have you been in a position where you're looking to hire people? I have. We resorted to foreigners because it is _VERY_ hard to find great programmers out there.
Cheap H1B my ass. The INS requires you to prove that you've looked in the local market for someone to fill the position first, and even then you have to pay slightly above the going average salary for the area.
I just recently accepted an H1B job offer in New York (I'm from Canada). The offer was more than generous, and the employer definitely does not have me by the balls. If I don't like it, I'll pack up and drive back to Canada. That's key when you're an H1-B: you can't ever presume that you'll stay in the US. You need a saftey net that will allow you to return home.
And sure, getting out of a lease is annoying, but the signing bonus at my next position would more than make up for that.
Beyond the law, there's a fundamental economic problem with the whole "pay per use" universe that the critics say the media moguls want to force upon us.
Information has to be priced in terms of value to the customer, not cost, since it has negligible reproduction costs. When you restrict usage of information, you lower that value for your customer. Similarily, if you allow broad usage of information, you raise the value of the product for your customer, but at the risk of selling fewer copies.
This is the dilemma RedHat has to deal with: by offering their product on the 'net, they're making fewer sales, even though the product is more valuable because the source is open. It's a tradeoff between price ($70 now) and amount sold.
So, I think that in the end, companies will have to find a "sweet spot" of how many rights to restrict. Pay per use is not a sweet spot, it's an extreme.
While I appreciate the references to the labor bureau's definition, please understand that my definition wasn't taking into account the statistical meaning of the term. I was drawing my definition of "knowledge work" mainly from Peter Drucker's definition of the term from his books during the 50's and 60's.
I find it odd that Slashdot would disclaim this article saying that "few jobs are knowledge work" today.
Blue collar work has been on a steady downward decline since the 1960's. A significant portion of work today IS some form of knowledge work.
Knowledge work is not just "working with computers" - it is an occupation where your education, experience and knowledge are what allow you to do your job.
E.g.: Auto mechanics Electricians Programmers IT workers Hardware technicians Radiologists Accountants Clerks Managers Analysts Hairdressers Nurses Doctors Lawyers Engineers Scientists Teachers Consultants Economists Marketing people
to name a few.
The bigger and better the knowledge, the more power one has over their employers. With regards to NDAs or NCAs or whatever agreement one may have, I don't really worry much about them. I have the right to reject an overly restrictive contract, and for narrowly-defined NCA's, finding a job at a non-competitor isn't much of a problem.
Slashdot Mirror
That's debatable. OO is a model that promotes a very simple dynamic memory allocation mechanism: the 'new' operator. This operator was designed to be combined with an automatic garbage collector. [This is based on observing Smalltalk - I'm not sure if Simula used "new"/had a garbage collector]
That C++ doesn't do this is somewhat irrelevant as C++ is clearly a "multi-paradigm" language, not a strictly OO one.
It is certainly an improvement upon top-down, procedural design in terms of flexibility and maintainability.
The jury is out on that.
Perhaps... I think that empirical evidence shows that software productivity and quality has been modestly improving over the last 20 years. [Given that the whole economy's productivity has been improving around 3% per year, mainly caused by information technology, according to Alan Greenspan]. The improvement in programmer productivity may not be caused by OO... but there certainly would be a correlation, just as there would be a correletion among other "paradigm waves", such as the structured programming movement in the 70's.
Go to Alphaworks and download "DirectDOM". It's there. And it kicks ass.
On the contrary, few programmers are actually versed enough with how to _handle_ the power of a multi-paradigm language that they often wind up being more productive in a more constrained environment.
This notion has its roots in philosophy where "true freedom" is actually a misnomer: we require some constraints in order to make sense of the world and to be productive in it.
So yes, there are situations where OO is not the ideal paradigm. Functional languages are a great alternative and are very expressive.
I think it's wrong to conclude that because OO is not a panacea it will be laughed out of the office. It is certainly an improvement upon top-down, procedural design in terms of flexibility and maintainability.
OO doesn't replace the need for good programmers. Shoddy programmers will ALWAYS make shoddy programs, no matter the paradigm.
HOWEVER, I would say that the constraints imposed on an OO language like Java or Smalltalk would result in a program that is less shoddy in the area of memory leaks and segmentation faults.
This, at the very least, is an quantitative improvement, by enforcing a paradigm of dynamic memory allocation (the "new" operator) and removing raw pointers from the language.
OO is not the perfect paradigm, nor should it be, or we might as well stop all software engineering research right now. It would be hard to argue that there is any ONE TRUE paradigm at all. I don't think anyone is claiming that it is the end of all software paradigms. It is, however, a useful and productive paradigm, and is worthy of widespread use. Which, as a matter of fact, is the case.
I would say the reason such opinions are read are because the legal field is, above all, an ongoing conversation among rational people.
Just because one is not a lawyer does not mean one cannot have a rational argument about a legal issue. To assume otherwise is just as preposterous as to assume that only programmers can have an opinion on whether a piece of software is of high quality or not.
The more detailed and low-level the argument, the more professional expertise becomes an issue. Hence "IANAL" becomes a necessary disclaimer.
I wouldn't go so far as to say the activity would have "vanished". All I would say is that it is extremely difficult to claim that it would be better without Microsoft. My reasoning for this is that new ventures are under tremendous duress during their growth years: they have to perform, they have to compete with the big boys, and they have to do it without a ton of money. Very often many small companies poised for hypergrowth are in a vicious circle: they don't have enough time to get the management talent needed to handle this growth, and hence they can't grow without major screwups. Netscape is a good example of this problem in effect.
Regarding your last paragraph: I don't know how many times this needs to be said before it sinks in. This whole case was brought up because Microsoft illegally removed true choice from the consumer. PEOPLE may have bought Microsoft, but they did so because they didn't have a real alternative. So the "market" didn't decide anything. I believe that, if it had, we would have a for more stable economy with our far more stable PC's.
I disagree.
The court found that Microsoft was guilty of propagating its monopoly through anticompetitive means, and attempting to gain monopoly power in the web browser industry through anti-competitive means. It found that Microsoft was NOT guilty of anti-competitve marketing agreements.
This ruling, to me, suggests that Microsoft's abuse of monopoly power really began around 1996. The consent decree of 1994/95 was designed to address the problems with Microsoft's former licencing practices. These practices were *not* deemed anti-competitive at the time.
Furthermore, Microsoft's licencing practices in the 80's were done at a time when there WAS market choice. You still could choose to get a Mac, or an Amiga in 1985. There was no 90% monopoly then. That the market decided to go with the herd, as it usually does, is disappointing, but not anti-competitive.
Even more so in 1995 - the Windows 95 launch was a moderate success, despite a substantial marketing effort on the part of IBM's OS/2 since 1994. The market did swing towards OS/2 for a while, before the Win95 release. But within one and a half years after Win95 came out, shrink-wrapped sales, not pre-loaded ones, overcame the entire OS/2 install base. The market decided again.
This case was about punishing Microsoft for attempting to remove choice in the market through the browser bundling, through its threats of dropping the Macintosh MS Office, and through its exclusive arrangements with ISP's. It is not punishing Microsoft for making shoddy, but popular operating systems.
Microsoft is responsible for U.S. economic growth in one important way: profits. Lots of them, and also profits from companies that support or are complements to Microsoft's products.
That totals up to tens of billions of dollars in GDP every year.
Another way: wealth. Almost every pension fund or mutual fund owns some sort of position in Microsoft. MANY ordinary people have gotten wealthier because of their success.
Another way: providing a de facto operating system standard. We may be outgrowing this now, but it was indispensible in allowing the software industry to grow. Stable infrastructure is a necessity for innovation.
Yet another way: Minor innovation. No, Microsoft is not a big innovator. They do lots of little things, however. Little things that are customer requests. Or are features the competition has. Incorporating any of these ideas into their products is an innovation because it exposes those ideas to a wider audience, continuing the positive feedback loop by attracting more customers to Microsoft.
Side note: If you don't think this is innovation, you probably have a double standard here. (Think about your definition of innovation as applied to MIcrosoft, and then apply it to Gnome or RedHat, or any other Linux distro vendor. And then tell me with a straight face that these companies are better innovators.)
The damage Microsoft has caused is quite intangible, at best. It would be difficult to claim that the amount of profits Microsoft has would be small compared to its potential competitors if they were allowed to thrive, given at how risky a business venture really is. However, the damage is real, so the remedy should at least be preventive, if not retributive (how do you compensate for what you can't measure?).
I think in the end some balance is required with Microsoft. They have done some good for the computer industry in a business sence, if not a technological sence by providing a stable infrastructure. They have also abused their power and done much damage to some parts of the industry. Hence, they should be punished and prevented from doing so again.
This does NOT make them the great satan who forced their wares upon us. Never forget that it is PEOPLE who bought Microsoft products. They could have bought Apple products or Commodore products, but they chose not to. Later they could have bought OS/2 during the Win95 upgrade but they chose not to. It's all about choice, and the market decided that Microsoft should win. And they did win, which is good. But they kept winning by illegally perpetuating their position, which is bad.
Sorry dude. I can second the experience. Netscape takes takes out X around 1 out of 5 times it crashes.
So, yes there's an exagerration that it happens "all the time", but it does happen.
The text of the ruling is available here: http://usvms.gpo.gov/
If the company is a monopoly, then I agree exclusive contracts should be a legal no-no. [And at 5pm today I think we'll find out what one judge thinks about this matter]
Otherwise, I think they should be free to do it. It's a relatively short term measure, at best.
Yes, the Star Wars trailer was only released in QuickTime. Some third parties converted it to AVI/MPEG, but the quality often wasn't kept during the conversion...
Sorenson has an exclusive licence with Apple; they can't create an AVI version of their codec.
www.sorenson.com
Sorenson has MPEG-1 quality with better sound support and compression. Take a look at the Star Wars trailers, for instance.
There hasn't been any empircal study contrasting the two approaches AFAIK, but just observing the quality contrasted to file size is an educational experience.
This probably would be a textbook example of a "well-behaved" monopoly... remember.. monopoly's can be good for consumers, it just usually happens that they aren't.
one would think that IBM learned their lesson on what it means to be a "poor behaved" monopoly.
I was going to reply to the parent article, but you pretty much covered all the bases. I second your opinion.
sorry folks, but I don't buy the "Star wars' CGI was same-old, same-old" argument. It was ground breaking work, and was BY FAR the most visually extraordinary movie of the year. ILM pulled off major innovations in order to get Phantom Menace out the door, and deserved the oscar.
The academy most likely wanted to give Lucas the finger. Which is somewhat understandable.
- let them be self important. we know the truth.
- actors affiliate themselves with pet causes because it attracts attention to the cause and to the actor. it's a win/win situation. Yes it's a sad state of affairs when we don't notice "great causes" unless a shining smile/body/face is attached to them, but can you really blame us? It's an information glut...
- sometimes it's easy to get jaded, but it's important to remember that WE are the reason hollywood is an empire. intellectual property isn't going to save them in the long term - you don't make that much money off of re-runs. It's up to us if we want to watch shlock movies.
if there's something to get jaded about, it's about wide spread cynicism, double standards and attention deficit in the general populace.
IIRC, that was the way it was supposed to happen in the original 1977 version. The effects at the time just didn't quite show it.
I actually liked his speech.
The drooling masses who watch any shlock hollywood throws at them do suck.
at the same time, the prententious art folk who watch obscure indie films just because it's "hard to understand" also suck.
the name of the game is "straddling the line" between art & sensibility. which is hard to do.
That no matter what system you devise, you're going to have the "tyranny of the ruling class". In a democracy, it's the "tyranny of the majority". Might does not make right.
The founding fathers of the U.S. knew this: no single political principle can work without a counterbalance.
(which is arguably why we have private property and free markets.)
I agree with what you're saying, so I'm not quite sure what your point is. My point was primarily that "pay per use" is not exactly the ideal pricing structure for most forms of information, so that a lot of the worrying going on about how we'll have to pay so much extra $ in the future to watch movies or listen to music is somewhat overzealous.
This is not to say that pay per use couldn't work, just that pay per use with extreme fair use restrictions would have a hard time working.
Uhh, he would fly back. Given the salaries most programmers make, this shouldn't prove to be a significant financial burden.
Have you been in a position where you're looking to hire people? I have. We resorted to foreigners because it is _VERY_ hard to find great programmers out there.
Cheap H1B my ass. The INS requires you to prove that you've looked in the local market for someone to fill the position first, and even then you have to pay slightly above the going average salary for the area.
I just recently accepted an H1B job offer in New York (I'm from Canada). The offer was more than generous, and the employer definitely does not have me by the balls. If I don't like it, I'll pack up and drive back to Canada. That's key when you're an H1-B: you can't ever presume that you'll stay in the US. You need a saftey net that will allow you to return home.
And sure, getting out of a lease is annoying, but the signing bonus at my next position would more than make up for that.
Beyond the law, there's a fundamental economic problem with the whole "pay per use" universe that the critics say the media moguls want to force upon us.
Information has to be priced in terms of value to the customer, not cost, since it has negligible reproduction costs. When you restrict usage of information, you lower that value for your customer. Similarily, if you allow broad usage of information, you raise the value of the product for your customer, but at the risk of selling fewer copies.
This is the dilemma RedHat has to deal with: by offering their product on the 'net, they're making fewer sales, even though the product is more valuable because the source is open. It's a tradeoff between price ($70 now) and amount sold.
So, I think that in the end, companies will have to find a "sweet spot" of how many rights to restrict. Pay per use is not a sweet spot, it's an extreme.
While I appreciate the references to the labor bureau's definition, please understand that my definition wasn't taking into account the statistical meaning of the term. I was drawing my definition of "knowledge work" mainly from Peter Drucker's definition of the term from his books during the 50's and 60's.
There's a wise saying: "Success means having the humility to work with people that are smarter than you."
There are smarter people than you out there. Seek them out, and be successful. Bill Gates did.
I find it odd that Slashdot would disclaim this article saying that "few jobs are knowledge work" today.
Blue collar work has been on a steady downward decline since the 1960's. A significant portion of work today IS some form of knowledge work.
Knowledge work is not just "working with computers" - it is an occupation where your education, experience and knowledge are what allow you to do your job.
E.g.:
Auto mechanics
Electricians
Programmers
IT workers
Hardware technicians
Radiologists
Accountants
Clerks
Managers
Analysts
Hairdressers
Nurses
Doctors
Lawyers
Engineers
Scientists
Teachers
Consultants
Economists
Marketing people
to name a few.
The bigger and better the knowledge, the more power one has over their employers. With regards to NDAs or NCAs or whatever agreement one may have, I don't really worry much about them. I have the right to reject an overly restrictive contract, and for narrowly-defined NCA's, finding a job at a non-competitor isn't much of a problem.