We must never forget that the DMCA is doing exactly what it was supposed to. There are no unintended consequences.
If this bill were passed, people would be allowed to legally circumvent copy protection. That would go completely against the whole principle of the DMCA which is to restrict access to copyrighted works.
The same people that fought to get the DMCA passed will fight to get this stopped. The problem is that these people are very powerful and have a lot of money. Don't hold your breath...
I agree and I disagree! I agree that we will probably never fully understand intelligence as it occurs in a human, but I believe that we will be (and already are) able to fake it.
When Gary Kasparov was beaten by Deep Blue he was thoroughly convinced that Deep Blue had sacrified a pawn. What actually happened was that Deep Blue had calculated that it would win the pawn back with interest later. There are already algorithms that can beat the Turing test when the man in the street is used as the judge (but not when someone who knows about such things is the judge).
So my theory is: We don't need to understand human intelligence to make something that looks like human intelligence.
Be careful. Authors are allowed to copyright writings and the like, but not patent them. Clearly the idea of patenting a mystery novel is just stupid. Inventions (physical devices and processes) are patentable, but are not subject to copyright. Copyrighting a mousetrap does not make sense.
The problem with software is that it is now possible to both copyright and patent software.
My biggest concern with something like this is: Who is going to pay for it? And the answer is: The user. So the user has to pay to have the way media can be used restricted.
Am I the only one who thinks that this doesn't make good business sense?
What is the most difficult part about starting a community ISP? Is it getting people interested, obtaining finances, finding people with the technical knowledge required to set up and maintain the system, licencing requirements, infrastructure, or something else? More importantly, how did you overcome this problem?
Your argument about Numerical Recipes using Heapsort in FORTRAN is flawed.
Numerical Recipes in C does NOT use recursion for Quicksort (to make the code faster) so there is no reason why the algorithm cannot be implemented in FORTRAN.
Heapsort is also recommended in Numerical Recipes in C because the worst case running time of Quicksort is very poor (it becomes an N^2 process), but this is not the case for Heapsort (about 20% worse than average).
So in both these cases your argument that this is a proof of fundamental limitations in FORTRAN is flawed.
Try checking the HOWTOs. I learnt half of what I know about Linux (which isn't much) by asking people and the other half just reading HOWTOs. The information is normally laid out according to tasks that must be performed which makes finding relevant information easier.
Only problem is that the HOWTOs are not always part of the default install and they can be rather well hidden from new users. Change that and I think that learning Linux would be MUCH easier!
My favourite error message
on
Pet Bugs?
·
· Score: 1
I have a piece of paper in my office that I inherited from someone who left a while back. It is a genuine Windows NT error message that he encountered trying to do something:
Here is an article from Independent Online. The Independent Newspapers group is (I think) the largest newspaper group in South Africa, so this should give a distinctly local flavour to the story.
Here in South Africa we have a statutory telecommunications monopoly. That means that by law nobody but our telecoms monopoly is allowed to provide telecoms resources. This has led to incredibly high costs with a 24% increase in local call rates earlier this year, for example. Our data rates were (and probably still are) among the top ten most expensive in the world in US dollar terms despite the fact that all expenses are in Rand (a weak currency).
This means that something like a net phone is a revelation in terms of cost. I have a friend who has been talking to his brother in Germany with a net phone for a while now. The only problem is that this is illegal because ISPs are not allowed to carry voice traffic! In fact the telecoms monopoly tried to destroy ISPs by citing a law that states that nobody is allowed to resell bandwidth. Fortunately the lost the case, but it was touch and go for a while.
My greatest sadness is that new technologies promise so much for countries like ours, but our government makes horrible mistakes like legislating a monopoly. If we can just learn to embrace new technologies and learn from trends round the world, we can rapidly pull ourselves to the front out of the mire we are in at the moment.
This might not have any legal basis, but how many open-source projects (SAMBA in this case) have the finances to prove it in court when Microsoft sues them?
It's all very well to get the contract changed, but then the employer just needs a clause that says something like "The provisions of this contract may be modified by the employer." Sometimes they add something like "in consultation with staff representatives." I know this from experience.
I work for a VERY large company (not in the USA). When I was appointed I received a document giving the regulations on doing consulting during one day of company time. This was one of the main reasons I accepted the post. It has since been retracted with no consultation with me. They did consult the staff representatives so they meet the contact requirements. Problem is that the staff representatives are either part of management or they are just ignored. And that's it... there is nothing I can do about it.
I think you make some excellent points, but you don't really take the argument far enough. The question everyone should be asking is: Why is essentially all software so poor? I think the basic answer to that is that software is written not designed.
Your analogy of a bridge is perfect. Why do the vast majority of bridges work first time and continue working essentially forever? The answer is that you only have one chance. Once construction on a bridge is started, it cannot be changed. This means that the engineers have to make sure that their bridge is properly designed long before construction ever starts.
Software is different. It is almost trivial to introduce changes at any point during the design process. This means that most people never take the time to design their software - they just start writing it because they know that they can hack it later to fix problems. A proper design process would mean that many of those problems would never arise and reduce the number of horrible hacks that exist in most software. If people started building bridges without going throught a proper design process the same thing would happen - bridges would be insecure and crash.
A large part of the answer to the problem of poor software is to force software writers to become software designers. These people would delay typing the first line of code until the design of the software is complete.
Then we can move towards a future where software is as solid and safe as bridges.
I live in a country where we do not have a particularly advanced national ID number system such as the American social security system or the system proposed in the article. I recently had an interesting conversation with my dad who works in a hospital on the accounting side. He mentioned that people overseas don't understand how he can function without a well-developed national ID system. They can't believe that it is still possible to track people and get payment (part of my dad's job) without having that person's ID. There are some problems that arise, but the basic point that I am trying to make is that something like this is not necessary - and there are countries that prove it. People are just so used to the system that they can't believe that it is possible to exist without it.
Microsoft has be found guilty of illegal practises and the decision was uphead on appeal.
So much of Microsoft's fortune has been made illegally.
Therefore punitive damages are absolutely fair.
Further, forcing Microsoft to use some of their illegally obtained fortune to encourage competition can hardly be considered unfair.
Lastly, Microsoft's definition of a fair settlement is irrelevant because they are the criminal here (proved in court and upheld on appeal).
Re:Which formats support simple batch manipulation
on
Non-MP3 Codecs?
·
· Score: 1
You could also check out Normalize. It's a utility that converts.wav (and.mp3 if you have the MAD library) files so that they sound like they have the same volume.
Check it out at http://www.cs.columbia.edu/~cvaill/normalize/
I'm an electronic engineer and a little while back I was playing around with an algorithm for designing impedance matching networks (circuits that make one impedance look like another one). The network can consist of series inductors, parallel inductors, series capacitors and parallel capacitors. I implemented each of these components with an object. So OOP can be used for some engineering programs!
On the other hand, heavy duty computations are probably not well suited to OOP because there is a very slight overhead involved with using objects. This small overhead can be multiplied a HUGE number of times for very large problem and can increase the computation time and memory requirements. That's one of the reasons FORTRAN is still so popular for heavy duty calculations: it's very simple and has no unnecessary overhead to speak of. Add to that the fact that it was designed for heavy duty mathematical work in the first place and you have a winner.
The current resurgence of left-of-centre governments in Europe could be seen as backing the argument of the emergence of a new type of society up. While these governments lean to the left, they are still much closer to the centre than the left.
Moving to Windows was initially fantastic, but I very quickly grew to miss the old command-line interface of DOS. Then I found Linux/Unix and I haven't looked back! I must be the only person in the known universe to have started using Linux/Unix because of DOS!
It's not a bad idea, but I think that there are two major problems that could prevent this from achieving widespread use.
The first based on the way cellular systems work. The major cost of the system is centred in those parts of the network that are shared between the users, namely the base stations. This makes the handsets (which each user must have) cheaper and thus lowers the system cost by sharing the cost of the expensive parts between users. Great, but why is that relevant? It is relevant because the maximum range between a cell and a base station is primarily determined by the low noise (read: expensive) receiver and high power (read: expensive) transmitter at the base station. So the range of cell-to-cell communication will be MUCH less than the base station-to-cell range. So much so that it is possible that the distance between a cell that can see a base station and a cell that can't will be no more than a few metres. This means that a large number of hops will be required in most cases clogging up the system bandwidth and meaning that each cell will need to be able to store a large number of messages. These problems will be even worse in an emergency when everyone and their dog are sending messages. Also, cell-to-cell communication will require much more power from the cell than base station-to-cell communication because the reciever in a cell is not nearly as good as that in a base station.
The second problem has to do with battery life. The battery life of cell phones is rather short when the transmitter is used. As explained previously, the power needed for cell-to-cell communication is rather high, aggravating the problem. This could cause major problems when people's batteries start going flat because the person in the next office likes sending lots of messages and is out of range of a base station (quite possible in buildings).
I think this is a great suggestion, but it assumes that the cellular environment is the same as the internet, which it is not.
If you are using large groups then there is a lot of opportunity for people to do nothing. In groups of two or three there is no way you can hide the fact that you have done nothing (or at least it is very difficult).
During my undergrad education we did all our practicals in groups of two people and you got to choose your partner. Choosing your partner was also great because it meant that you could pick someone who you knew you could work with. The slackers got known very early on and people who wanted to do good work avoided them.
Lastly, working together should not be completely removed even if students have to hand in their own work. I would rather have a student be able to do the work after asking a friend, than not be able to do the work at all. Of course, each student still has to do their own work, so work that is nearly identical is still unacceptable.
There is no perfect solution to group work, but these ideas might help.
Slashdot Mirror
If this bill were passed, people would be allowed to legally circumvent copy protection. That would go completely against the whole principle of the DMCA which is to restrict access to copyrighted works.
The same people that fought to get the DMCA passed will fight to get this stopped. The problem is that these people are very powerful and have a lot of money. Don't hold your breath...
As someone who was involved with the Warez scene: Why did you do it? What motivated you?
When Gary Kasparov was beaten by Deep Blue he was thoroughly convinced that Deep Blue had sacrified a pawn. What actually happened was that Deep Blue had calculated that it would win the pawn back with interest later. There are already algorithms that can beat the Turing test when the man in the street is used as the judge (but not when someone who knows about such things is the judge).
So my theory is: We don't need to understand human intelligence to make something that looks like human intelligence.
The problem with software is that it is now possible to both copyright and patent software.
Am I the only one who thinks that this doesn't make good business sense?
What is the most difficult part about starting a community ISP? Is it getting people interested, obtaining finances, finding people with the technical knowledge required to set up and maintain the system, licencing requirements, infrastructure, or something else? More importantly, how did you overcome this problem?
Numerical Recipes in C does NOT use recursion for Quicksort (to make the code faster) so there is no reason why the algorithm cannot be implemented in FORTRAN.
Heapsort is also recommended in Numerical Recipes in C because the worst case running time of Quicksort is very poor (it becomes an N^2 process), but this is not the case for Heapsort (about 20% worse than average).
So in both these cases your argument that this is a proof of fundamental limitations in FORTRAN is flawed.
Only problem is that the HOWTOs are not always part of the default install and they can be rather well hidden from new users. Change that and I think that learning Linux would be MUCH easier!
I have a piece of paper in my office that I inherited from someone who left a while back. It is a genuine Windows NT error message that he encountered trying to do something:
Windows NT encountered the following error:
The operation completed successfully
Here is an article from Independent Online. The Independent Newspapers group is (I think) the largest newspaper group in South Africa, so this should give a distinctly local flavour to the story.
i d= qw1023967081968B253&set_id=1
http://www.iol.co.za/index.php?click_id=31&art_
This means that something like a net phone is a revelation in terms of cost. I have a friend who has been talking to his brother in Germany with a net phone for a while now. The only problem is that this is illegal because ISPs are not allowed to carry voice traffic! In fact the telecoms monopoly tried to destroy ISPs by citing a law that states that nobody is allowed to resell bandwidth. Fortunately the lost the case, but it was touch and go for a while.
My greatest sadness is that new technologies promise so much for countries like ours, but our government makes horrible mistakes like legislating a monopoly. If we can just learn to embrace new technologies and learn from trends round the world, we can rapidly pull ourselves to the front out of the mire we are in at the moment.
This might not have any legal basis, but how many open-source projects (SAMBA in this case) have the finances to prove it in court when Microsoft sues them?
I work for a VERY large company (not in the USA). When I was appointed I received a document giving the regulations on doing consulting during one day of company time. This was one of the main reasons I accepted the post. It has since been retracted with no consultation with me. They did consult the staff representatives so they meet the contact requirements. Problem is that the staff representatives are either part of management or they are just ignored. And that's it... there is nothing I can do about it.
I'll probably vote with my feet and leave.
Your analogy of a bridge is perfect. Why do the vast majority of bridges work first time and continue working essentially forever? The answer is that you only have one chance. Once construction on a bridge is started, it cannot be changed. This means that the engineers have to make sure that their bridge is properly designed long before construction ever starts.
Software is different. It is almost trivial to introduce changes at any point during the design process. This means that most people never take the time to design their software - they just start writing it because they know that they can hack it later to fix problems. A proper design process would mean that many of those problems would never arise and reduce the number of horrible hacks that exist in most software. If people started building bridges without going throught a proper design process the same thing would happen - bridges would be insecure and crash.
A large part of the answer to the problem of poor software is to force software writers to become software designers. These people would delay typing the first line of code until the design of the software is complete.
Then we can move towards a future where software is as solid and safe as bridges.
I live in a country where we do not have a particularly advanced national ID number system such as the American social security system or the system proposed in the article. I recently had an interesting conversation with my dad who works in a hospital on the accounting side. He mentioned that people overseas don't understand how he can function without a well-developed national ID system. They can't believe that it is still possible to track people and get payment (part of my dad's job) without having that person's ID. There are some problems that arise, but the basic point that I am trying to make is that something like this is not necessary - and there are countries that prove it. People are just so used to the system that they can't believe that it is possible to exist without it.
So much of Microsoft's fortune has been made illegally.
Therefore punitive damages are absolutely fair.
Further, forcing Microsoft to use some of their illegally obtained fortune to encourage competition can hardly be considered unfair.
Lastly, Microsoft's definition of a fair settlement is irrelevant because they are the criminal here (proved in court and upheld on appeal).
You could also check out Normalize. It's a utility that converts .wav (and .mp3 if you have the MAD library) files so that they sound like they have the same volume.
Check it out at http://www.cs.columbia.edu/~cvaill/normalize/
In light of this do you believe that it is possible to make money from open source/free software alone or does a company need a hardware arm like Sun?
I'm an electronic engineer and a little while back I was playing around with an algorithm for designing impedance matching networks (circuits that make one impedance look like another one). The network can consist of series inductors, parallel inductors, series capacitors and parallel capacitors. I implemented each of these components with an object. So OOP can be used for some engineering programs!
On the other hand, heavy duty computations are probably not well suited to OOP because there is a very slight overhead involved with using objects. This small overhead can be multiplied a HUGE number of times for very large problem and can increase the computation time and memory requirements. That's one of the reasons FORTRAN is still so popular for heavy duty calculations: it's very simple and has no unnecessary overhead to speak of. Add to that the fact that it was designed for heavy duty mathematical work in the first place and you have a winner.
Just my opinion mind you!
The current resurgence of left-of-centre governments in Europe could be seen as backing the argument of the emergence of a new type of society up. While these governments lean to the left, they are still much closer to the centre than the left.
... Policeman sans donut puzzles everyone!
Moving to Windows was initially fantastic, but I very quickly grew to miss the old command-line interface of DOS. Then I found Linux/Unix and I haven't looked back! I must be the only person in the known universe to have started using Linux/Unix because of DOS!
Mmmmm, but this is slashdot, everything MS does is evil!
The first based on the way cellular systems work. The major cost of the system is centred in those parts of the network that are shared between the users, namely the base stations. This makes the handsets (which each user must have) cheaper and thus lowers the system cost by sharing the cost of the expensive parts between users. Great, but why is that relevant? It is relevant because the maximum range between a cell and a base station is primarily determined by the low noise (read: expensive) receiver and high power (read: expensive) transmitter at the base station. So the range of cell-to-cell communication will be MUCH less than the base station-to-cell range. So much so that it is possible that the distance between a cell that can see a base station and a cell that can't will be no more than a few metres. This means that a large number of hops will be required in most cases clogging up the system bandwidth and meaning that each cell will need to be able to store a large number of messages. These problems will be even worse in an emergency when everyone and their dog are sending messages. Also, cell-to-cell communication will require much more power from the cell than base station-to-cell communication because the reciever in a cell is not nearly as good as that in a base station.
The second problem has to do with battery life. The battery life of cell phones is rather short when the transmitter is used. As explained previously, the power needed for cell-to-cell communication is rather high, aggravating the problem. This could cause major problems when people's batteries start going flat because the person in the next office likes sending lots of messages and is out of range of a base station (quite possible in buildings).
I think this is a great suggestion, but it assumes that the cellular environment is the same as the internet, which it is not.
If you are using large groups then there is a lot of opportunity for people to do nothing. In groups of two or three there is no way you can hide the fact that you have done nothing (or at least it is very difficult).
During my undergrad education we did all our practicals in groups of two people and you got to choose your partner. Choosing your partner was also great because it meant that you could pick someone who you knew you could work with. The slackers got known very early on and people who wanted to do good work avoided them.
Lastly, working together should not be completely removed even if students have to hand in their own work. I would rather have a student be able to do the work after asking a friend, than not be able to do the work at all. Of course, each student still has to do their own work, so work that is nearly identical is still unacceptable.
There is no perfect solution to group work, but these ideas might help.