I can see two reasons why a company might not want to distribute source code with their applications.
The first is that it could potentially be a support nightmare. People who know just enough about programming and software to make small, "minor", changes, but not enough to debug the problem when they go wrong. And since they paid for the software, they feel entitled to have someone at the vendor fix the mistake for them. This will probably start happening in the Free Software / Open Source world as well, but Free software maintainers can send people away without fear of lost sales. Commercial vendors have to at least appear to be trying to keep customers happy.
The second reason that occurs to me is that having the source available would make reverse engineering an application relatively trivial. Want to see how that cool feature of a competitor's package was implemented? Buy a copy and look at the source. Then have some one write a detailed description of the algorithm that can be passed to someone that hasn't seen the source for implementation. Mind you, I think this is actually a non-issue except for a few, "innovative", algorithms, but it has to be a concern.
I think the long term benefits of opening the API (greater third party support, longer product life cycles, etc.) greatly outweigh the short term benefits of keeping it closed. However, U.S. companies tend to think on a quarterly basis...
What if the MPAA had simply ignored DeCSS and done nothing at all about it? Mr. Garbus makes a good point in that DeCSS probably wouldn't have spread nearly as quickly if not for the publicity caused by the lawsuit. It's entirely likely that it would have faded into the background until the day a linux DVD player based on the code emerged. It strikes me as trying to kill dandelions by cutting off the seed heads, not only do you not kill the weed, you spread the seeds farther than they would have propagated by themselves.
I'm sorry, but I have to disagree on the CS degree. When I'm looking to hire someone to write software, I'm much more interested in their experience than their education. What matters more to me is that they have shown the ability to learn quickly and be flexible in the real world. Maybe I've got the wrong people applying, but I've yet to see someone with a CS degree (and limited experience) who could actually work in the real world. It's been my experience that it's easier to teach someone with a degree or experience in the problem domain how to program than it is to teach a CS major the problem domain. It's gotten bad enough that I won't even consider someone with a CS degree that is fresh out of school unless they have some experience to back it up.
In the US using sprint, I pay $0.10 a minute any time, anywhere on the sprint network, to any domestic number. Of course, I have to buy 500 minutes a month to get that rate...
One really nice thing about QNX is that distributed processing comes naturally to it. If you design the system right, and split the tasks correctly, they can run on any system on the network transparently. What's really cool is that this applies to the I/O as well, not just compute tasks. You can access, say, the serial ports on one machine from any other machine in the network. The nice thing is, you don't do anything differently than you would to access the local serial port. The message passing architecture, once you get used to it, is very powerful and allows you to easily build complex systems. My last position was designing and coding the control system for a CMP tool (Chemical Mechanical Planarization -- Polishes wafers nice and flat so the next layer can be deposited). I had to control 12 different systems from 6 different manufacturers in real time, QNX didn't necessarily make it possible, but it made it a lot easier...
I seem to remember that one of the older computer magazines (might have been Byte) used Xscheme as the basis of their benchmarks. It was a readily available language, with source code, that had been ported to a wide number of platforms. Since it was interpreted, it tended to run slow enough for meaningful timings. Could not something similar be done with Java? Specifically, one of the open source implementations? Write a set of benchmarks in Java using the subset of Java that currently works on the open source VMs and then try them on different platforms. You could even use the compilation time as a benchmark (using jikes maybe). It would allow cross platform testing on at least semi-real world applications.
QNX is intended for embedded systems. Most embedded systems don't need grep or any of a host of other utilities. QNX lets you get just a realtime kernel with a fairly normal POSIX API. It's not really designed as a desktop OS. In fact, when I was developing for QNX, I usually did most of the source editing, etc. on a Linux box, then compiled it from QNX. Worked great. In any case, the idea behind QNX licensing is that you only pay for what you need.
Won't work. Ethernet is baseband, i.e. there is no carrier signal that is modulated. It would be like an old spark gap transmitter and would probably destroy radio and tv reception in the area around it. Remember, a pulse has virtually infinite harmonics (I know that is oversimplifying) and an ethernet card is generating lots of pulses.
Just a note, hydrogen burns really well, but it doesn't really detonate. It tends to disperse too quickly. People get this image of the Hindenburg going down in flames and they think, "Oh, my look at all that hydrogen burning!". But you'll note, the Hindenburg, even all aflame, actually settled to the ground fairly gently, primarily because its gas cells were still largely intact and leaking only slowly. Not to mention that when hydrogen burns, it's a colorless to blue flame. What really caused the problems on the Hindenburg was the waterproofing agent coating outer canvas skin. They've tested some surviving fragments, turns out the waterproofer was pretty much rocket fuel and highly flammable. So hydrogen, particulary in small amounts is not all that dangerous. Mind you, an open flame around a car battery (which generates hydrogen) isn't all that bright an idea, but I wouldn't be too concerned with a few milliliters of H2 on a plane in a small hardened container (like a CO2 cartridge for a pellet gun perhaps).
I love my MD player. I even bought one for my car. I can't see the point of an MP3 player when I can get 74-80 minutes of music on a nice convenient removable disk that costs $5 each. Oh and I live in the U.S.
It's important to note that there are a number of specific questions that they are seeking answers to. This includes examples of current technological access control and whether it works or not. This includes the impact of said access control on both content users and providers. A prime example of this could be Macrovision on VHS. It is relatively cheap and easy to circumvent, but almost nobody bothers to do so to copy a video tape.
The interoperability issue is important here as well, one (or more) of the questions relate to how access control can limit legitimate use of content. DVD CSS is a prime example, you purchased a DVD, you're licensed to view it, but cannot because there is no viewer for the operating system you use. LiVid circumvents the access control technology for the purpose of legitimate use (viewing a DVD). The fact that it can be used for illegitimate uses is irrelevant to the Copyright Office.
A good strategy for your comments would be to copy the questions from the web site and submit answers to them point by point. These are the questions they want to resolve, those that clearly take the time to study and write clear, concise answers for each will receive more attention than the rambling flamebaiters.
I agree. I'm an INTP (which means I can't schedule to save my life). I love the design work and prototyping, but I'd much rather leave the final production details to someone else. It was always amusing for me to go into a meeting, have someone request a new feature and be able to say, "Already thought of that... the implementation hooks are in place." Of course, I drive managers nuts because I pretty much refuse to do any but the most essential paperwork (that's what the manager is for, right?). I'm also encouraged by the 21% some college number...
I think you're referring to the short "Hardware Wars". They do all the special effects using kitchen appliances (an iron for the star cruiser, egg beaters [the old fashioned kind] as tie fighters, etc.). I thought it was pretty funny, but it's been a number of years since I've seen it. Last I saw it, it was available on a tape with "Bambi vs. Godzilla", "Closet Cases of the Nerd Kind", and "Porklips Now" (an "Apocalypse Now" spoof).
There's a whole book of this type of letter from a chap by the name of Ted Nancy. The book is titled, appropriately enough, "Letters From a Nut" (ISBN 0-380-97354-5). As far as I know, it's only available on dead trees. The funniest thing about the book is that most of the companies take the letters absolutely seriously....
Nested ternaries are great. I once wrote a function to do ROT-13 transforms on text using a for loop and a very deeply nested ternary. It crashed MSC 5.0 (couldn't handle the nesting...)
Slashdot Mirror
I can see two reasons why a company might not want to distribute source code with their applications.
The first is that it could potentially be a support nightmare. People who know just enough about programming and software to make small, "minor", changes, but not enough to debug the problem when they go wrong. And since they paid for the software, they feel entitled to have someone at the vendor fix the mistake for them. This will probably start happening in the Free Software / Open Source world as well, but Free software maintainers can send people away without fear of lost sales. Commercial vendors have to at least appear to be trying to keep customers happy.
The second reason that occurs to me is that having the source available would make reverse engineering an application relatively trivial. Want to see how that cool feature of a competitor's package was implemented? Buy a copy and look at the source. Then have some one write a detailed description of the algorithm that can be passed to someone that hasn't seen the source for implementation. Mind you, I think this is actually a non-issue except for a few, "innovative", algorithms, but it has to be a concern.
I think the long term benefits of opening the API (greater third party support, longer product life cycles, etc.) greatly outweigh the short term benefits of keeping it closed. However, U.S. companies tend to think on a quarterly basis...
jim nutt
What if the MPAA had simply ignored DeCSS and done nothing at all about it? Mr. Garbus makes a good point in that DeCSS probably wouldn't have spread nearly as quickly if not for the publicity caused by the lawsuit. It's entirely likely that it would have faded into the background until the day a linux DVD player based on the code emerged. It strikes me as trying to kill dandelions by cutting off the seed heads, not only do you not kill the weed, you spread the seeds farther than they would have propagated by themselves.
jim
I'm sorry, but I have to disagree on the CS degree. When I'm looking to hire someone to write software, I'm much more interested in their experience than their education. What matters more to me is that they have shown the ability to learn quickly and be flexible in the real world. Maybe I've got the wrong people applying, but I've yet to see someone with a CS degree (and limited experience) who could actually work in the real world. It's been my experience that it's easier to teach someone with a degree or experience in the problem domain how to program than it is to teach a CS major the problem domain. It's gotten bad enough that I won't even consider someone with a CS degree that is fresh out of school unless they have some experience to back it up.
In the US using sprint, I pay $0.10 a minute any time, anywhere on the sprint network, to any domestic number. Of course, I have to buy 500 minutes a month to get that rate...
One really nice thing about QNX is that distributed processing comes naturally to it. If you design the system right, and split the tasks correctly, they can run on any system on the network transparently. What's really cool is that this applies to the I/O as well, not just compute tasks. You can access, say, the serial ports on one machine from any other machine in the network. The nice thing is, you don't do anything differently than you would to access the local serial port. The message passing architecture, once you get used to it, is very powerful and allows you to easily build complex systems. My last position was designing and coding the control system for a CMP tool (Chemical Mechanical Planarization -- Polishes wafers nice and flat so the next layer can be deposited). I had to control 12 different systems from 6 different manufacturers in real time, QNX didn't necessarily make it possible, but it made it a lot easier...
jim nutt
I seem to remember that one of the older computer magazines (might have been Byte) used Xscheme as the basis of their benchmarks. It was a readily available language, with source code, that had been ported to a wide number of platforms. Since it was interpreted, it tended to run slow enough for meaningful timings. Could not something similar be done with Java? Specifically, one of the open source implementations? Write a set of benchmarks in Java using the subset of Java that currently works on the open source VMs and then try them on different platforms. You could even use the compilation time as a benchmark (using jikes maybe). It would allow cross platform testing on at least semi-real world applications.
jim
QNX is intended for embedded systems. Most embedded systems don't need grep or any of a host of other utilities. QNX lets you get just a realtime kernel with a fairly normal POSIX API. It's not really designed as a desktop OS. In fact, when I was developing for QNX, I usually did most of the source editing, etc. on a Linux box, then compiled it from QNX. Worked great. In any case, the idea behind QNX licensing is that you only pay for what you need.
Won't work. Ethernet is baseband, i.e. there is no carrier signal that is modulated. It would be like an old spark gap transmitter and would probably destroy radio and tv reception in the area around it. Remember, a pulse has virtually infinite harmonics (I know that is oversimplifying) and an ethernet card is generating lots of pulses.
Just a note, hydrogen burns really well, but it doesn't really detonate. It tends to disperse too quickly. People get this image of the Hindenburg going down in flames and they think, "Oh, my look at all that hydrogen burning!". But you'll note, the Hindenburg, even all aflame, actually settled to the ground fairly gently, primarily because its gas cells were still largely intact and leaking only slowly. Not to mention that when hydrogen burns, it's a colorless to blue flame. What really caused the problems on the Hindenburg was the waterproofing agent coating outer canvas skin. They've tested some surviving fragments, turns out the waterproofer was pretty much rocket fuel and highly flammable. So hydrogen, particulary in small amounts is not all that dangerous. Mind you, an open flame around a car battery (which generates hydrogen) isn't all that bright an idea, but I wouldn't be too concerned with a few milliliters of H2 on a plane in a small hardened container (like a CO2 cartridge for a pellet gun perhaps).
I love my MD player. I even bought one for my car. I can't see the point of an MP3 player when I can get 74-80 minutes of music on a nice convenient removable disk that costs $5 each. Oh and I live in the U.S.
It's important to note that there are a number of specific questions that they are seeking answers to. This includes examples of current technological access control and whether it works or not. This includes the impact of said access control on both content users and providers. A prime example of this could be Macrovision on VHS. It is relatively cheap and easy to circumvent, but almost nobody bothers to do so to copy a video tape.
The interoperability issue is important here as well, one (or more) of the questions relate to how access control can limit legitimate use of content. DVD CSS is a prime example, you purchased a DVD, you're licensed to view it, but cannot because there is no viewer for the operating system you use. LiVid circumvents the access control technology for the purpose of legitimate use (viewing a DVD). The fact that it can be used for illegitimate uses is irrelevant to the Copyright Office.
A good strategy for your comments would be to copy the questions from the web site and submit answers to them point by point. These are the questions they want to resolve, those that clearly take the time to study and write clear, concise answers for each will receive more attention than the rambling flamebaiters.
I agree. I'm an INTP (which means I can't schedule to save my life). I love the design work and prototyping, but I'd much rather leave the final production details to someone else. It was always amusing for me to go into a meeting, have someone request a new feature and be able to say, "Already thought of that... the implementation hooks are in place." Of course, I drive managers nuts because I pretty much refuse to do any but the most essential paperwork (that's what the manager is for, right?).
I'm also encouraged by the 21% some college number...
I think you're referring to the short "Hardware Wars". They do all the special effects using kitchen appliances (an iron for the star cruiser, egg beaters [the old fashioned kind] as tie fighters, etc.). I thought it was pretty funny, but it's been a number of years since I've seen it. Last I saw it, it was available on a tape with "Bambi vs. Godzilla", "Closet Cases of the Nerd Kind", and "Porklips Now" (an "Apocalypse Now" spoof).
There's a whole book of this type of letter from a chap by the name of Ted Nancy. The book is titled, appropriately enough, "Letters From a Nut" (ISBN 0-380-97354-5). As far as I know, it's only available on dead trees. The funniest thing about the book is that most of the companies take the letters absolutely seriously....
Nested ternaries are great. I once wrote a function to do ROT-13 transforms on text using a for loop and a very deeply nested ternary. It crashed MSC 5.0 (couldn't handle the nesting...)