No you weren't. Would this type of answer work in a court room?
Okay, seriously: For me, these answers did not help at all. Did you expect the questions would be easier? I know that some of these questions dealt with complex legal matters. You completely skirted some key issues such as the legality of making MP3 files from CDs. Holding an opinion on that is almost a pre-requisite for the entire discussion, and people wanted to know what you believed the law says on that. One of the questions dealt with getting a replacement for a damaged CD. You acted like you have never heard the RIAA/MPAA claim that customers aren't buying a CD with a copy of the music, but that they are are "licensing" the content of the disk? Are you not aware of that? Or am I completely misinformed? That's what I was looking for in this discussion.
I do have to grant you that lots of this audience are pirates with no respect for copyright law or for other people's property. They really are the "they are just bits of data so I can do anything - they want to be freeeee!" crowd, so I can understand some level of frustration. But I think educating them (and me) would go a long way toward resolving this issue.
Okay, I am getting jumped on for my exaggerated statement about IANAL. I'm not really saying that lawyers are dumb - I'm really saying that this particular lawyer didnt' take the questions seriously and didn't add anything that we didn't already know. He really came-off as arrogant and not very knowledgable. Maybe my expectations were too high. Both replies to my post used one of my favorite phrases: know what you don't know. You are both correct.
Regarding some of your points: 1. You are right. That is exactly why I was hoping a lawyer would provide some answers. Or at least legally-informed opinions.
2. There is something wrong with the law when a series of experts in a field get together and conclude something, and the patent office or relevant legal authority concludes something else. I'm not saying that Slashdot provides a complete picture of any legal topic, but there are some serious clashes between law and technology that result in some real non-sequitors.
3. Nobody asked for the answer to be a few sentences. If this group wanted an interview, they should have been prepared to type more than a few sentences. I guess the expectations were off.
4. I am pretending that I heard sarcasm here. Are you saying that a group of technologists does not understand technology more than a group of older men? Most of the judges decisions rest on applying legal knowledge on top of expert witness testimony. In theory, the judges should be able to be luddites and still make accurate decisions. But it is difficult to present a clear picture of a technology issue in a courtroom, especially when a line of expensive lawyers are paying to find an expert who will say what they want.
If the government has the right to stop DRM, then they also have the right to impose it. It isn't worth it. Next thing you know, Trusted Computing will be a legal mandate.
I'm so glad I started reading "Atlas Shrugged" since the book is about this very topic. In the book, a nice benevolent guy runs a steel company that just wants to make better steel, and make a profit at the same time. But because his company didn't hire a "Washington man" the company starts becoming subject to stupid regulations and roadblocks since his competition spends more time lobbying than innovating.
I don't know how the book ends. So don't spoil it for me, but it seems like this story is true to real life. If Google doesn't hire a "Washington Man" then, they will get beaten. It doesn't matter if they improve the world, create jobs, do good, and make a profit at the same time. If they topple somebody else's stagnating empire, then they must be stopped. And in today's oligarchy this is the only way to do it. This whole thing sounds just like the RIAA.
I'm not sure if I would rather Google die honorably, or live with a poisoned heart. But I can guarantee you the stock holders have a definite opinion.
And yes, I realize that this post gets an automatic "-1, Ann Rand":-)
Mod this guy to infinity. Normally this is a trollish comment, but in this case it is dead on. This was one of the worst interviews Slashdot has had. These guys don't seem to know anything about the subject they claim to deal in. From now on, if someone says IANAL at the top of their post, that tells me the probably know MORE about the subject, not LESS. Sheesh.
I am glad to see this on Slashdot since regular expressions is an area that geeks could really use help in.
For example, instead of saying the common geek expression "Greetings Program!" try a more regular expression such as "Hello Sir" or the more casual "Wassup?" IRL, Tron references are not considered cool. Another common faux pas is using the expression "Hey n00b, what's your function?" instead of something more regular like "Hey dog, what's your problem?" If someone tries to threaten you, think about their technical skills before saying "Close your port before I pwn j00!" Life is not an FPS. "Shut up before I kick your ass" works very well.
I would love Microsoft to make this statement about their.NET related patents. C# is a very good language, and the.NET framework is excellent. I would love to use Mono to make cross-platform C# applications, but the fear of Microsoft striking down and killing various implementations of the runtime via patents is too dangerous.
Havoc (hav'-uhk) - noun: great destruction or devastation; ruinous damage.
I don't like the voting machines, but it doesn't help to have sensationalist articles against them. This is akin to someone forgetting to bring the power cords.
Even better: They can do that, then re-release the 30GB HD-DVD version in a few years. It will have with better quality, be HD remastered (tm), and some new special features.
Everyone is asking about how to defend against the suits. I want to know the other side of things:
What do you think that the RIAA should do to prevent piracy? Do you agree or disagree with the lawsuits as they are doing them now? Do you suggest a better way? How about your opinion on the current state of copyright law?
This is a classic case of a software patent: it describes a system that does something that isn't especially novel. It's a database lookup with a spelling checker and a few other tools. It's a neat idea. It's great. It shouldn't be patentable.
dbill: The point is that two years of increasing hurricanes does not prove global warming. Follow me for a moment: 1. Global warming causes more hurricanes 2. We see more hurricanes 3. Therefore, there is global warming.
Most people look at the above and thing this is correct. But it is a logical fallacy. Just because global warming can be the cause of more hurricanes does not mean that it is definitely the cause of increased hurricanes. Try this one out:
1a. Global warming causes more hurricanes 1b. An 8-12 year natural cycle causes increased hurricanes 2. We see more hurricanes 3. Therefore, there is global warming.
Now the error becomes more obvious. Seeing more hurricanes doesn't prove anything. It might be the natural cycle. It might be global warming. It might be both. It might be neither. It might be something entirely different. More information is required to draw a conclusion.
The problem with global warming, in general, is that most of the proofs for it look something like the above. So it takes a whole lot of evidence and science that is beyond most people know about to really prove it. Then, proving that mankind is involved is even more difficult, because it is proving one specific cause out of many many factors. And we can't isolate variables and make experiments since we only have one earth. So scientifically, we will never actually know if we are the cause of global warming, even if we fix the problem. That is why there is controversy and confusion and infighting on this subject.
I always wondered if an individual could setup their own DGPS and use it for navigating home robots. For example, my Roomba would be more efficient if it had a layout of my home and knew where it was. But GPS is too broad. But could I place a DGPS receiver in my house and make a GPS accurate to 10 cm (As I've heard people claim is possible)? If I understand the principle correctly, DGPS is just about knowing your position precisely and sending out correction information.
I use Cavalier Telephone and they started this practice a few weeks ago. First I thought it was a bug in Firefox (my google search bar not working). Then I spent 2 hours checking for a virus or spyware. After that, I tried another DNS server and realized the problem. I never expected a small DSL provider to do this kinda crud.
this is almost certainly the worst solution possible
The article doesn't say it is the worst solution possible. It states that it is a bad solution, points out some limitations and complications with it, and proposes no alternatives. There's nothing in there that says it is the worst solution, because it isn't. For any solution you can propose, I can one that is it worse.:-)
The problem is that a process isn't always better. The reality is that on most OSs, using a process is way more expensive. Which is really how this discussion started.
IMHO, a thread is the best solution. The complications that you list are unrealistic, out-of-date, and the conclusion about threading only applies to one OS: Linux. Using a thread is the common solution is because it works on most OSs, and is taught in most textbooks, and is the "optimal" solution taught to CS students. You can't rewrite the theory and change the text books due to an unusual implementation in one OS. Linux handles threads poorly and processes well. It is the opposite of how Windows, OS X, and BSD work.
And, again, you fail threading 101, strncpy() doesn't guarantee the order of assignment.
I never said it was. I'm not asserting that you can or should write to the same string from two processes at once using strncpy(). Really, this situation is only a problem in old C code that uses static buffers. It just shouldn't happen. In a modern application, the data is stored in TLS or in an object so the whole point is moot. The author created an example where there is no strncpy() problem, then introduced the problem, then fixed it and blamed strncpy. There is nothing wrong with strncpy(). It performs perfectly well. It doesn't uses static buffers. It is re-entrant and thread-safe. However, that doesn't mean you can't create odd situations where it won't work.
You are obviously a very smart programmer. You're knowledge of specific calls and thread issues is great. (I'm browsing your site - I plan to read some of your articles). Just don't get caught-up in the assumption that what applies to Linux applies everywhere. And don't tell people they are ignorant and "fail" according to your opinion just because they disagree with you.
I also must apologize. I probably would not have had such a scathing review of the article if I had paid attention to the fact that the person linking to it was also the author.:-) Let me make some concrete suggestions out of this. The difference in opinions here is quite minor and it boils down to a small point about threads.
1) If the article is trying to say that processes are better than threads then: - Qualify the statement to an OS/architecture/whatever (or don't, and say why you think it applies universally) - State the benefits of threads specifically - Change the title to something about processes-vs-threads - Give an example using processes. I've never used processes to get around blocking calls, so maybe it is easier than I think. For many people there is a fear of fork() because it has it's own set of complexities. Maybe there is a way that doesn't involve fork (that is portable???)
2) If the article is merely meant as a precaution, like "hey, did you think of this problem? or this one?" - State that threads are sometimes good (or rarely, or whatever), but have hidden problems - Remove the term "slow" from the title, because the article doesn't state that threads are slow. - Remove the term "buggy" from the title, because it isn't threads that are buggy, it is the code.
Maybe something like "Threads: hidden complexities" or something like that.
Traffic between me and my neighbour which traverses only the POP should cost less then traffic between me and another point in my ISP
Fine, but that has nothing to do with the type of traffic. It has to do with where the traffic goes. If you want to charge a different rate for traffic that travels within the ISP than traffic that goes outside the ISP, that is fair.
he problem will go away immediately if ISPs turn off flat pricing and users start to pay for bandwidth used.
That is the ideal, perfect solution to the problem. Get what you pay for. But don't expect most Slashdotters to agree with you because they are the ones who will be using greater than average bandwidth.
Even better - if they start charging a differential/tiered pricing depending on the type of traffic.
Whoahhhh, that's too far. It isn't fair to meter based on the type of traffic. Bittorrent traffic does not cost more than HTML over HTTP, or VOIP. This is where it goes from being economics to being politics, and everybody loses in this situation. Oh, except for the lawyers and politicians.
Of course people left
on
Duke in Trouble?
·
· Score: 2, Insightful
Duke Nukem has been in the making for 10 years! With an average lifespan in this undustry of 3 -5 years, how the heck could a 10 year project be completed with the same set of employees?
"It's time to write code and chew gum... and I'm all outta gum."
I must admit that you are right: I've done some more research, and I've learned something: gethostbyname is not thread safe on some old operating system called "Linux." Fortunately, most of my programming has been on modern operating systems such as Windows, where you can assume any system call is thread safe, except in rare cases where you are doing something odd. I've never ported to BSD, at which point I would have had to learn to use gethostbyname_r, which is the reentrant version.
I guess it is no big deal, since gethostbyname has been deprecated since 1999 so it really doesn't matter. At least now I know why some applications do their own DNS lookups. That always seemed strange to me. It's quite silly when trivial kernel calls aren't thread-safe, but they can be re-implemented in a thread-safe manner in user code. Sheesh.
strncpy() is thread-safe: You can call strncpy() in a reentrant fashion from as many threads as you wish. Now that is completely different from trying to strncpy() over the same memory area in multiple threads. Fortunately, the example code in the links doesn't do that. The author just decided to prove a point and make it lock the strings unnecessarily.
I guess what the article boils down to is that fork()ing is better than threading because the author's favorite operating system doesn't support thread-safe calls, and instead makes processes cheap. Cheap processes is a good thing. But trying to tell people that a programming paradigm is wrong because the kernel coders didn't make functions re-entrant is foolish.
All the reviews I've seen are very positive, but none of the features I want seem to be in the new kit.
1) Still can't program in a real language: I want to make a double-dimensioned array so I can map out an NxN grid for a game. It doesn't look like I can do that in this set. When I was 10 years old, I was using BASIC to code, and it had this ability.
2) I want a growable set of inputs and outputs. If I want a 4th motor, their solution is to buy another brain. Instead, I should be able to plug-in a daisy-chain connector and assign one as motor 0 and the other as motor 1. The mod community made something like this for the old Mindstorms. Without that, this is just as limited as the old set. You can't make anythin significant without more motors and sensors. They don't all need to be running at once.
I still have hopes that a more detailed review will reveal that these things are actually possible.
I'm reading these articles, and the only occurrance of the word "Bush" is in a deragatory comment. This looks like it is something the EPA is doing, but the articles are too inflammatory to explain why they are doing it. I've never heard of these libraries, so I don't know who has access to them or what is in them or what the real impact is. Was the problem that these were not used by anyone? Or is the information redundant and out-of-date? Is it already in other libraries? The article doesn't even bother to answer the who/what/where/why/when so it is impossible to make a judgement.
This reminds me of a chain-email I got a few months ago about how George Bush Senior was working with some company to try and strip-mine some place in Brazil. I was enraged - then I read the article and realized it isn't a strip-mine, it doesn't involve George Bush, and it isn't an American company, and that the local supported the effort. I don't know if this is a competitor trying to start fake grassroots efforts by using anti-Bush sentiment, or if it is political enemies, or if there is something real happening here.
These anti-thread arguments are ridiculous and obsolete. Let me take this article for example:
...not block, and threads are the worst fix for that problem...
The author tells us that threading is a bad-solution to the problem of doing asynchronous work. He then pretends that some system function is not-reentrant and not thread-safe, writes some intentionally bad naive threading code, then tells us the problems with it. Thus, threading is bad! The assumptions that the code runs under are >10 years out-of-date (a non-reentrant thread-unsafe version of strncpy!?!?). And lastly, the author proposes no alternate solution! He just tells us that the only solution that exists is very bad.
I suppose it is all really irrelevant anyhow, because none of this does anything to prove or disprove the case of threads-vs-processes, nor clarify if OSX is really better or worse at using them than Linux.
Slashdot Mirror
That's half of the answer people wanted. The other half is "do you think they are right?"
No you weren't. Would this type of answer work in a court room?
Okay, seriously: For me, these answers did not help at all. Did you expect the questions would be easier? I know that some of these questions dealt with complex legal matters. You completely skirted some key issues such as the legality of making MP3 files from CDs. Holding an opinion on that is almost a pre-requisite for the entire discussion, and people wanted to know what you believed the law says on that. One of the questions dealt with getting a replacement for a damaged CD. You acted like you have never heard the RIAA/MPAA claim that customers aren't buying a CD with a copy of the music, but that they are are "licensing" the content of the disk? Are you not aware of that? Or am I completely misinformed? That's what I was looking for in this discussion.
I do have to grant you that lots of this audience are pirates with no respect for copyright law or for other people's property. They really are the "they are just bits of data so I can do anything - they want to be freeeee!" crowd, so I can understand some level of frustration. But I think educating them (and me) would go a long way toward resolving this issue.
Okay, I am getting jumped on for my exaggerated statement about IANAL. I'm not really saying that lawyers are dumb - I'm really saying that this particular lawyer didnt' take the questions seriously and didn't add anything that we didn't already know. He really came-off as arrogant and not very knowledgable. Maybe my expectations were too high. Both replies to my post used one of my favorite phrases: know what you don't know. You are both correct.
Regarding some of your points:
1. You are right. That is exactly why I was hoping a lawyer would provide some answers. Or at least legally-informed opinions.
2. There is something wrong with the law when a series of experts in a field get together and conclude something, and the patent office or relevant legal authority concludes something else. I'm not saying that Slashdot provides a complete picture of any legal topic, but there are some serious clashes between law and technology that result in some real non-sequitors.
3. Nobody asked for the answer to be a few sentences. If this group wanted an interview, they should have been prepared to type more than a few sentences. I guess the expectations were off.
4. I am pretending that I heard sarcasm here. Are you saying that a group of technologists does not understand technology more than a group of older men?
Most of the judges decisions rest on applying legal knowledge on top of expert witness testimony. In theory, the judges should be able to be luddites and still make accurate decisions. But it is difficult to present a clear picture of a technology issue in a courtroom, especially when a line of expensive lawyers are paying to find an expert who will say what they want.
If the government has the right to stop DRM, then they also have the right to impose it. It isn't worth it. Next thing you know, Trusted Computing will be a legal mandate.
I'm so glad I started reading "Atlas Shrugged" since the book is about this very topic. In the book, a nice benevolent guy runs a steel company that just wants to make better steel, and make a profit at the same time. But because his company didn't hire a "Washington man" the company starts becoming subject to stupid regulations and roadblocks since his competition spends more time lobbying than innovating.
:-)
I don't know how the book ends. So don't spoil it for me, but it seems like this story is true to real life. If Google doesn't hire a "Washington Man" then, they will get beaten. It doesn't matter if they improve the world, create jobs, do good, and make a profit at the same time. If they topple somebody else's stagnating empire, then they must be stopped. And in today's oligarchy this is the only way to do it. This whole thing sounds just like the RIAA.
I'm not sure if I would rather Google die honorably, or live with a poisoned heart. But I can guarantee you the stock holders have a definite opinion.
And yes, I realize that this post gets an automatic "-1, Ann Rand"
Mod this guy to infinity. Normally this is a trollish comment, but in this case it is dead on. This was one of the worst interviews Slashdot has had. These guys don't seem to know anything about the subject they claim to deal in. From now on, if someone says IANAL at the top of their post, that tells me the probably know MORE about the subject, not LESS. Sheesh.
I am glad to see this on Slashdot since regular expressions is an area that geeks could really use help in.
For example, instead of saying the common geek expression "Greetings Program!" try a more regular expression such as "Hello Sir" or the more casual "Wassup?" IRL, Tron references are not considered cool. Another common faux pas is using the expression "Hey n00b, what's your function?" instead of something more regular like "Hey dog, what's your problem?" If someone tries to threaten you, think about their technical skills before saying "Close your port before I pwn j00!" Life is not an FPS. "Shut up before I kick your ass" works very well.
This is why we Americans invade small countries: to get test subjects! Why waste them?
I would love Microsoft to make this statement about their .NET related patents. C# is a very good language, and the .NET framework is excellent. I would love to use Mono to make cross-platform C# applications, but the fear of Microsoft striking down and killing various implementations of the runtime via patents is too dangerous.
Havoc (hav'-uhk) - noun: great destruction or devastation; ruinous damage.
I don't like the voting machines, but it doesn't help to have sensationalist articles against them. This is akin to someone forgetting to bring the power cords.
Even better: They can do that, then re-release the 30GB HD-DVD version in a few years. It will have with better quality, be HD remastered (tm), and some new special features.
Everyone is asking about how to defend against the suits. I want to know the other side of things:
What do you think that the RIAA should do to prevent piracy? Do you agree or disagree with the lawsuits as they are doing them now? Do you suggest a better way? How about your opinion on the current state of copyright law?
I'm reading the patent application. It sounds a lot like the Spanish Verb Conjugation engine.
This is a classic case of a software patent: it describes a system that does something that isn't especially novel. It's a database lookup with a spelling checker and a few other tools. It's a neat idea. It's great. It shouldn't be patentable.
Shut up both of you.
dbill: The point is that two years of increasing hurricanes does not prove global warming. Follow me for a moment:
1. Global warming causes more hurricanes
2. We see more hurricanes
3. Therefore, there is global warming.
Most people look at the above and thing this is correct. But it is a logical fallacy. Just because global warming can be the cause of more hurricanes does not mean that it is definitely the cause of increased hurricanes. Try this one out:
1a. Global warming causes more hurricanes
1b. An 8-12 year natural cycle causes increased hurricanes
2. We see more hurricanes
3. Therefore, there is global warming.
Now the error becomes more obvious. Seeing more hurricanes doesn't prove anything. It might be the natural cycle. It might be global warming. It might be both. It might be neither. It might be something entirely different. More information is required to draw a conclusion.
The problem with global warming, in general, is that most of the proofs for it look something like the above. So it takes a whole lot of evidence and science that is beyond most people know about to really prove it. Then, proving that mankind is involved is even more difficult, because it is proving one specific cause out of many many factors. And we can't isolate variables and make experiments since we only have one earth. So scientifically, we will never actually know if we are the cause of global warming, even if we fix the problem. That is why there is controversy and confusion and infighting on this subject.
I always wondered if an individual could setup their own DGPS and use it for navigating home robots. For example, my Roomba would be more efficient if it had a layout of my home and knew where it was. But GPS is too broad. But could I place a DGPS receiver in my house and make a GPS accurate to 10 cm (As I've heard people claim is possible)? If I understand the principle correctly, DGPS is just about knowing your position precisely and sending out correction information.
How common are DGPS receivers?
What was the original reason for the secret ballot? Was it to prevent coercion?
I use Cavalier Telephone and they started this practice a few weeks ago. First I thought it was a bug in Firefox (my google search bar not working). Then I spent 2 hours checking for a virus or spyware. After that, I tried another DNS server and realized the problem. I never expected a small DSL provider to do this kinda crud.
The problem is that a process isn't always better. The reality is that on most OSs, using a process is way more expensive. Which is really how this discussion started.
IMHO, a thread is the best solution. The complications that you list are unrealistic, out-of-date, and the conclusion about threading only applies to one OS: Linux. Using a thread is the common solution is because it works on most OSs, and is taught in most textbooks, and is the "optimal" solution taught to CS students. You can't rewrite the theory and change the text books due to an unusual implementation in one OS. Linux handles threads poorly and processes well. It is the opposite of how Windows, OS X, and BSD work.
I never said it was. I'm not asserting that you can or should write to the same string from two processes at once using strncpy(). Really, this situation is only a problem in old C code that uses static buffers. It just shouldn't happen. In a modern application, the data is stored in TLS or in an object so the whole point is moot. The author created an example where there is no strncpy() problem, then introduced the problem, then fixed it and blamed strncpy. There is nothing wrong with strncpy(). It performs perfectly well. It doesn't uses static buffers. It is re-entrant and thread-safe. However, that doesn't mean you can't create odd situations where it won't work.
You are obviously a very smart programmer. You're knowledge of specific calls and thread issues is great. (I'm browsing your site - I plan to read some of your articles). Just don't get caught-up in the assumption that what applies to Linux applies everywhere. And don't tell people they are ignorant and "fail" according to your opinion just because they disagree with you.
I also must apologize. I probably would not have had such a scathing review of the article if I had paid attention to the fact that the person linking to it was also the author.
1) If the article is trying to say that processes are better than threads then:
- Qualify the statement to an OS/architecture/whatever (or don't, and say why you think it applies universally)
- State the benefits of threads specifically
- Change the title to something about processes-vs-threads
- Give an example using processes. I've never used processes to get around blocking calls, so maybe it is easier than I think. For many people there is a fear of fork() because it has it's own set of complexities. Maybe there is a way that doesn't involve fork (that is portable???)
2) If the article is merely meant as a precaution, like "hey, did you think of this problem? or this one?"
- State that threads are sometimes good (or rarely, or whatever), but have hidden problems
- Remove the term "slow" from the title, because the article doesn't state that threads are slow.
- Remove the term "buggy" from the title, because it isn't threads that are buggy, it is the code.
Maybe something like "Threads: hidden complexities" or something like that.
Fine, but that has nothing to do with the type of traffic. It has to do with where the traffic goes. If you want to charge a different rate for traffic that travels within the ISP than traffic that goes outside the ISP, that is fair.
Whoahhhh, that's too far. It isn't fair to meter based on the type of traffic. Bittorrent traffic does not cost more than HTML over HTTP, or VOIP. This is where it goes from being economics to being politics, and everybody loses in this situation. Oh, except for the lawyers and politicians.
Duke Nukem has been in the making for 10 years! With an average lifespan in this undustry of 3 -5 years, how the heck could a 10 year project be completed with the same set of employees?
"It's time to write code and chew gum... and I'm all outta gum."
I must admit that you are right: I've done some more research, and I've learned something: gethostbyname is not thread safe on some old operating system called "Linux." Fortunately, most of my programming has been on modern operating systems such as Windows, where you can assume any system call is thread safe, except in rare cases where you are doing something odd. I've never ported to BSD, at which point I would have had to learn to use gethostbyname_r, which is the reentrant version.
I guess it is no big deal, since gethostbyname has been deprecated since 1999 so it really doesn't matter. At least now I know why some applications do their own DNS lookups. That always seemed strange to me. It's quite silly when trivial kernel calls aren't thread-safe, but they can be re-implemented in a thread-safe manner in user code. Sheesh.
strncpy() is thread-safe: You can call strncpy() in a reentrant fashion from as many threads as you wish. Now that is completely different from trying to strncpy() over the same memory area in multiple threads. Fortunately, the example code in the links doesn't do that. The author just decided to prove a point and make it lock the strings unnecessarily.
I guess what the article boils down to is that fork()ing is better than threading because the author's favorite operating system doesn't support thread-safe calls, and instead makes processes cheap. Cheap processes is a good thing. But trying to tell people that a programming paradigm is wrong because the kernel coders didn't make functions re-entrant is foolish.
All the reviews I've seen are very positive, but none of the features I want seem to be in the new kit.
1) Still can't program in a real language: I want to make a double-dimensioned array so I can map out an NxN grid for a game. It doesn't look like I can do that in this set. When I was 10 years old, I was using BASIC to code, and it had this ability.
2) I want a growable set of inputs and outputs. If I want a 4th motor, their solution is to buy another brain. Instead, I should be able to plug-in a daisy-chain connector and assign one as motor 0 and the other as motor 1. The mod community made something like this for the old Mindstorms. Without that, this is just as limited as the old set. You can't make anythin significant without more motors and sensors. They don't all need to be running at once.
I still have hopes that a more detailed review will reveal that these things are actually possible.
I'm reading these articles, and the only occurrance of the word "Bush" is in a deragatory comment. This looks like it is something the EPA is doing, but the articles are too inflammatory to explain why they are doing it. I've never heard of these libraries, so I don't know who has access to them or what is in them or what the real impact is. Was the problem that these were not used by anyone? Or is the information redundant and out-of-date? Is it already in other libraries? The article doesn't even bother to answer the who/what/where/why/when so it is impossible to make a judgement.
This reminds me of a chain-email I got a few months ago about how George Bush Senior was working with some company to try and strip-mine some place in Brazil. I was enraged - then I read the article and realized it isn't a strip-mine, it doesn't involve George Bush, and it isn't an American company, and that the local supported the effort. I don't know if this is a competitor trying to start fake grassroots efforts by using anti-Bush sentiment, or if it is political enemies, or if there is something real happening here.