Yeah. IMO, that really needs to change. Teams with student-designed and student-built robots really spoke out last year, and I think that if more teams proclaim the fact that their robots were designed and built by students, we shouldn't really need regulation. Although, there will always be teams with company-built robots, but hopefully they'll become a minority.
Most non-rookie teams have learned that they have to start fundrasing long before the kickoff. Also, once a team has found a large corporate sponsor, that sponsor us usually kind and interested enough to sponsor them the next year as well. After two years of struggling for sponsorship, we have gotten great support from Rockwell who provides money, resources, and engineers to guide us. FIRST will always be an expensive competition because it requires many more resources (and plain old mass) than your run-of-the-mill high school clubs.
Yeah... This year's competition is a mixed blessing: No one robot can do all the tasks, but on the other hand, no one robot can do all the tasks. There was a real sense of competition before whereas now you are forced to work with everyone. I liked last year with 2 vs. 2 much better because there was both competition AND teamwork.
Yes, but the default program on the robot end of the control system needs to be changed a lot to be useful for most robots. Making it autonomous would be too much for most high schools.
Dynamiclly Linked Library. Though I have never programmed on the MacOS, I'm pretty sure you have something similar. Anyhow, I don't really see the point of your argument. If there is a problem with DLLs it is simply a bug in the program(or in some cases the dll), not in the concept of DLLs.
It seems to me (although I really don't have that much experience) that the generally accepted rule is that everything is statically linked with its libraries. This makes the Mac version of many programs several times larger than their Windows counterparts. Also, the Mac memory management system is not very conducive to dynamically-loaded libraries.
This practice of tacking line-items onto the ends of bills is abhorrant. If this bill gets passed (even if it doesn't!), I'm going to be sure to encrypt every email I send.
Why is it that everyone seems to love Java? Why am I the only one who thinks it's a crufty, hacked-together language that's only good for marketing? From a language-design standpoint, Python is a much more well-constructed language than Java, and it uses bytescodes and interpretation exactly the same way. Why on earth do you need strong typing when you're programming in an interpreted language? Makes no sense to me...
</rant>
Having taken the test (no preparation, by the way. I had no clue what the questions would be like), it seemed to me that all the test was really testing was the test-taker's ability to follow directions. Some of the classes they used for the problem were just begging a rewrite (especially that class for storing arbitrary precision numbers. Ick!). While it tested some simple knowledge, most of the rest of it was simply, "Implement this! Now, implement this!". They didn't even ask the test-taker to develop any kind of algorithm. It was all simple translations of algorithms into code. Didn't seem like much of a test to me...
It's already really good (recent CVS at least). Monty et al. do some really good work... I especially like the "concatenated ogg files play with no gaps and change the track info appropriately" feature, and once Icecast unicast and multicast (is Icecast doing multicast?) is done, people will have to be amazingly dense not to realize it's better than mp3.
Their quote was specifically worded to not actually say anything. That way, they can't be sued for libel. A lot of "we think" and "it's highly likely" instead of "we checked and lines abc through xyz in the file qwerty.c are a violation of our patents".
When I first started learing about computers, I started programming in Windows. I learned a lot of low-level Win32 API calls and I can make an entire simple application from scratch with only minimal trips to the documentation. Despite all that, when I decided to give Linux a try I was quickly hooked. Being able to compare the way Linux does things to the way Windows does things, I can tell you for certain that Linux/Unix is easier to program in than Windows. If that isn't "natural appeal" for programmers, then I don't know what is. Bugs caused by closed drivers? Maybe for some people.
Sady, I seem to have to explain what Vorbis is and why it's important every time I mention it to somebody. They usually don't seem to understand the concept of "free". I tell them, "A hardware manufacturer can make a Vorbis player for only the cost of R&D and manufacturing. No royalties." "But... how do they stay in business?" Oh well.
Well, he DID do a zone transfer of yankees.com, although it should be pretty easy to prove that he did it AFTER the hack. I'm sure zone transfers (or attempts) of large sites right BEFORE they're hacked is a very bad idea.
Not if there's a statistical 99% chance that that bit came from a watermark. A bit doesn't have to actually be a 1 or 0 on disk (if it were it'd probably get lost in compression). More likely there are slight statistical smearing of the data such that, if you know where to look, you can get a probability of there being a watermark. If it's over some threshold, the music is considered "watermarked".
It wouldn't surprise me if all the watermarks were indeed cracked, but the SDMI group (really just the RIAA) was just using the $10,000 as bait to get dumb hackers to do their work for them. Unfortunately for them, they really think that poorly-designed crypto algorithms can be just "patched" to make them work better (there was an interview a while back in which one of the watermark designers said that bad watermarks can be "patched" to make them stronger). Last time I checked, crypto systems (including watermarks) were either secure or not secure. The problem with this type of watermark is that they are supposed to be hard to detect but everyone will have code to detect them. Once that's compromized, it's simply a matter of plugging it into some sort of evolution model and run it until there's no watermark left.
Yes, but in memory, a running program has pages of ram mapped from its virtual, 32-bit (or 64-bit) addressable memory range to physical ram. This makes things like paging to disk, rearranging physical ram pages, and memory-mapped files really easy to implement. It also makes the system more robust as userspace programs can't f*** about with kernel memory or memory in other programs. On the Palm (and in MacOS, yuck!) every program can access all memory because all memory shows up as one linear contiguous chunk.
Censorware has a history of problems. Sometimes it doesn't block sites that should be blocked, sometimes it blocks sites that shouldn't. I can pretty much guarantee that any censorware product will block a page that somebody legitimately needs to get to. While censorware in libraries is theoretically a good idea, I think it would be a disaster in practice. What would me much better, in my opinion, is to have a big sign saying "All network traffic is monitored" even if it's not. That way people won't go to sites they wouldn't show to other people, without infringing on their rights.
I know this post is a little redundant, but a while back I emailed them asking when results could be expected and they replied with this:
"Date: Sun, 15 Oct 2000 14:28:48 -070
From: contact <contact@hacksdmi.org>
Subject: RE: When can we expect official results?
Submissions from the public challenge are still being analyzed. Thanks for
your interest."
Nobody can give a "cracked"/"not cracked" answer until all the results have been analyzed. (Although, I suppose they could give a "cracked" answer if their system REALLY sucks...)
Slashdot Mirror
Yeah. IMO, that really needs to change. Teams with student-designed and student-built robots really spoke out last year, and I think that if more teams proclaim the fact that their robots were designed and built by students, we shouldn't really need regulation. Although, there will always be teams with company-built robots, but hopefully they'll become a minority.
Aaron Plattner
Most non-rookie teams have learned that they have to start fundrasing long before the kickoff. Also, once a team has found a large corporate sponsor, that sponsor us usually kind and interested enough to sponsor them the next year as well. After two years of struggling for sponsorship, we have gotten great support from Rockwell who provides money, resources, and engineers to guide us. FIRST will always be an expensive competition because it requires many more resources (and plain old mass) than your run-of-the-mill high school clubs.
Aaron Plattner
Yeah... This year's competition is a mixed blessing: No one robot can do all the tasks, but on the other hand, no one robot can do all the tasks. There was a real sense of competition before whereas now you are forced to work with everyone. I liked last year with 2 vs. 2 much better because there was both competition AND teamwork.
Aaron Plattner
Yes, but the default program on the robot end of the control system needs to be changed a lot to be useful for most robots. Making it autonomous would be too much for most high schools.
Aaron Plattner
4.).DLL? what's that?
Dynamiclly Linked Library. Though I have never programmed on the MacOS, I'm pretty sure you have something similar. Anyhow, I don't really see the point of your argument. If there is a problem with DLLs it is simply a bug in the program(or in some cases the dll), not in the concept of DLLs.
It seems to me (although I really don't have that much experience) that the generally accepted rule is that everything is statically linked with its libraries. This makes the Mac version of many programs several times larger than their Windows counterparts. Also, the Mac memory management system is not very conducive to dynamically-loaded libraries.
This practice of tacking line-items onto the ends of bills is abhorrant. If this bill gets passed (even if it doesn't!), I'm going to be sure to encrypt every email I send.
Why is it that everyone seems to love Java? Why am I the only one who thinks it's a crufty, hacked-together language that's only good for marketing? From a language-design standpoint, Python is a much more well-constructed language than Java, and it uses bytescodes and interpretation exactly the same way. Why on earth do you need strong typing when you're programming in an interpreted language? Makes no sense to me... </rant>
Aaron
Having taken the test (no preparation, by the way. I had no clue what the questions would be like), it seemed to me that all the test was really testing was the test-taker's ability to follow directions. Some of the classes they used for the problem were just begging a rewrite (especially that class for storing arbitrary precision numbers. Ick!). While it tested some simple knowledge, most of the rest of it was simply, "Implement this! Now, implement this!". They didn't even ask the test-taker to develop any kind of algorithm. It was all simple translations of algorithms into code. Didn't seem like much of a test to me...
Aaron
It's already really good (recent CVS at least). Monty et al. do some really good work... I especially like the "concatenated ogg files play with no gaps and change the track info appropriately" feature, and once Icecast unicast and multicast (is Icecast doing multicast?) is done, people will have to be amazingly dense not to realize it's better than mp3.
No. That doesn't involve conversion from time domain to frequency domain or any kind of quantization (because it's lossless).
Their quote was specifically worded to not actually say anything. That way, they can't be sued for libel. A lot of "we think" and "it's highly likely" instead of "we checked and lines abc through xyz in the file qwerty.c are a violation of our patents".
Oh well.
When I first started learing about computers, I started programming in Windows. I learned a lot of low-level Win32 API calls and I can make an entire simple application from scratch with only minimal trips to the documentation. Despite all that, when I decided to give Linux a try I was quickly hooked. Being able to compare the way Linux does things to the way Windows does things, I can tell you for certain that Linux/Unix is easier to program in than Windows. If that isn't "natural appeal" for programmers, then I don't know what is. Bugs caused by closed drivers? Maybe for some people.
Aaron Plattner
Sady, I seem to have to explain what Vorbis is and why it's important every time I mention it to somebody. They usually don't seem to understand the concept of "free". I tell them, "A hardware manufacturer can make a Vorbis player for only the cost of R&D and manufacturing. No royalties." "But... how do they stay in business?" Oh well.
Well, he DID do a zone transfer of yankees.com, although it should be pretty easy to prove that he did it AFTER the hack. I'm sure zone transfers (or attempts) of large sites right BEFORE they're hacked is a very bad idea.
Aaron Plattner
Not if there's a statistical 99% chance that that bit came from a watermark. A bit doesn't have to actually be a 1 or 0 on disk (if it were it'd probably get lost in compression). More likely there are slight statistical smearing of the data such that, if you know where to look, you can get a probability of there being a watermark. If it's over some threshold, the music is considered "watermarked".
Aaron Plattner
It wouldn't surprise me if all the watermarks were indeed cracked, but the SDMI group (really just the RIAA) was just using the $10,000 as bait to get dumb hackers to do their work for them. Unfortunately for them, they really think that poorly-designed crypto algorithms can be just "patched" to make them work better (there was an interview a while back in which one of the watermark designers said that bad watermarks can be "patched" to make them stronger). Last time I checked, crypto systems (including watermarks) were either secure or not secure. The problem with this type of watermark is that they are supposed to be hard to detect but everyone will have code to detect them. Once that's compromized, it's simply a matter of plugging it into some sort of evolution model and run it until there's no watermark left.
Aaron Plattner
Yes, but in memory, a running program has pages of ram mapped from its virtual, 32-bit (or 64-bit) addressable memory range to physical ram. This makes things like paging to disk, rearranging physical ram pages, and memory-mapped files really easy to implement. It also makes the system more robust as userspace programs can't f*** about with kernel memory or memory in other programs. On the Palm (and in MacOS, yuck!) every program can access all memory because all memory shows up as one linear contiguous chunk.
Aaron Plattner
Censorware has a history of problems. Sometimes it doesn't block sites that should be blocked, sometimes it blocks sites that shouldn't. I can pretty much guarantee that any censorware product will block a page that somebody legitimately needs to get to. While censorware in libraries is theoretically a good idea, I think it would be a disaster in practice. What would me much better, in my opinion, is to have a big sign saying "All network traffic is monitored" even if it's not. That way people won't go to sites they wouldn't show to other people, without infringing on their rights.
Aaron Plattner
I know this post is a little redundant, but a while back I emailed them asking when results could be expected and they replied with this:
"Date: Sun, 15 Oct 2000 14:28:48 -070
From: contact <contact@hacksdmi.org>
Subject: RE: When can we expect official results?
Submissions from the public challenge are still being analyzed. Thanks for
your interest."
Nobody can give a "cracked"/"not cracked" answer until all the results have been analyzed. (Although, I suppose they could give a "cracked" answer if their system REALLY sucks...)
Aaron Plattner