I use portable storage devices to move data from one place to another. Generally, I need to remove the old stuff to make room for the new stuff. Until these device become large enough to compete with hard drives, which they are nowhere near, they will tend to be used for volatile, mobile data.
When your laptop can easily have a 64Gb flash-based storage device, a comparable hard drive will hold 1TB. People have never been satisfied with 1/16th the storage they could have, and I don't think that's going to change.
This was an unintended (at least by the authors) consequence of the DMCA. It needs to be fixed... badly. If a copyright holder uses a technological mechanism that prevents fair use, it is not circumvention to bypass that mechanism.
You cannot "circumvent" a mechanism that protects a right the creator of that mechanism does not actually have.
Single-threaded applications, for cases where performance matters, will likely become a thing of the past. It has always been cheaper to make two fast cores work together than make one core twice as fast. That's a reality that Intel and AMD have kept typical application programmers from facing for much longer than most people thought possible.
On the bright side, there are a lot of programmers out there who know how to make fast software for multiprocessor machines. The problem is, most of them don't have experience with programming things like games. One nice thing about DirectX 10 and similar high-level APIs is helps to shift the burden of taking advantage of sophisticated CPU technologies to the authors of driver code and away from the authors of application code.
I can foresee few problems with application programmers learning to take advantage of four core machines in the next decade or so. However, I can no way for such application to usefully employ more than four cores. The problem is that the types of things we need our computers to do only parallelize to a certain degree.
With a game, you can have a thread that handles network I/O, a thread that handles AI, a thread that handles physics, and a thread that handles the user interface and 3D rendering. If you try to go much past that, your threads wind up getting in each others' way more than they cooperate.
So ultimate we're going to need either each core to be faster or we're going to need radically different ways to *use* the other cores. There are a few crazy ideas out there, but none of them are anywhere close to reality... yet!
That's too indirect to count. There's no precise rule for how direct the causality has to be, but that's clearly too indirect. If mechanical problems take a plane out of service, we don't blame the mechanical problems for car accidents when people who would have taken the plane have to take cars instead.
Does anyone care to predict what affect this will have on Windows? Will people create their own complete driver sets for Windows? The hardware specifications can be used to make more than just software for Linux.
Is it just way too much effort to develop a Windows display driver? Or might we see projects to develop an open source Vista driver for ATI hardware? Might ATI even open source their Windows driver?
It's hard to see what their downside is. The upside would be that they may wind up getting better drivers without having to do all the work themselves.
I totally agree, and I boycott manufacturers who slipstream. You should be able to evaluate a product by its model number and make and be reasonably assured of getting substantially the same product if you purchase another of the same model number.
I can understand issues with parts availability and fixing design flaws. I cannot understand or defend substantive hardware changes without a change in model identifier.
IMO, reviewers should refuse to review products from manufacturers who slipstream. What is the point of reviewing a product when there is no way you can tell a consumer what product you reviewed such that they can be reasonably assured of buying and getting the product you reviewed?!
I almost never buy products that have rebates. I think they are basically a scam, for three reasons:
1) You pay sales tax on the amount rebated to you. So when you buy something with a $20 rebate in an area with 7% sales tax, you are basically having $1.40 stolen from you. A $199 rebate steals $14 from you.
2) The rebates are often not sent. You could wind up spending as much as an hour or more getting the rebate processed, and you still may not get it.
3) Too often, a product will have a higher price but a rebate at an office supply type store. The idea is that you buy the more expensive product, get reimbursed for the full price on the receipt by your employer, and then keep the rebate for yourself. The companies that promote these more-expensive products with-rebate sold next to less-expensive no-rebate products know exactly what they're doing and why they're doing it.
4) You often have to consent to your contact information being used for all kinds of marketing purposes. This often includes the manufacturer selling it.
I haven't looked at the OOXML specification, but if it's anything like the CIFS and SMB specifications, I'd have to agree with you. The CIFS specification says things like, "you either need to put X here or Y here, and it is not known how to tell which".
In any event, it looks like Microsoft deservedly lost in its attempt to standardize OOXML. And really, this (OOXML) doesn't have anything to do with browsers.
"The term ''interactive computer service'' means any information service, system, or access software provider that provides or enables computer access by multiple users to a computer server, including specifically a service or system that provides access to the Internet and such systems operated or services offered by libraries or educational institutions."
So, yes, it's a bit of stretch for that to cover a piece of software that only incidentally uses a server. Oddly, the law has similar protections for an "access software provider", which seems to fit a bit better:
"The term ''access software provider'' means a provider of software (including client or server software), or enabling tools that do any one or more of the following:
(A) filter, screen, allow, or disallow content; (B) pick, choose, analyze, or digest content; or (C) transmit, receive, display, forward, cache, search, subset, organize, reorganize, or translate content."
Seems the result should follow from (A) and (B). Malware scanners certainly filter, screen, and allow or disallow content.
That's a common myth. At one time, it might have been the truth, though it was never really tested. Now there are laws that specifically give ISPs and others the right to filter without become liable for things they filter inadvertently or things they fail to filter. This law is still relatively untested, so it's not clear exactly when it applies and who it applies to.
It was mostly intended to cover the AOL issue. AOL wanted to offer some moderation of its forums to create more child-friendly forums. But they were worried that if they tried to moderate the forums, they might become liable for anything that slipped through.
The law has recently covered identifying programs as spyware or malware. Apparently, so long as you do this in good faith, you are not liable for false positives or false negatives.
So this should cover most filtering an ISP might do. Whether it will cover *forging* packets for traffic level management, I don't know. That's quite a stretch. But the myth "you're a common carrier unless you filter" is false for many reasons.
I've had the "pleasure" of implementing other standards that were created in this way (such as CIFS). And while the standards that result are truly horrible, at least there is some documentation rather than none.
If Microsoft pushes through a truly horrible standard, that doesn't compel anyone else to implement it. It doesn't prevent anyone else from implementing a different/better standard. It does mean that anyone who has to implement it has decent documentation to follow.
Microsoft has basically documented its specification and licensed any patent rights it holds that would cover the specification. This allows, but does not force, other people to use its specification where they think it benefits them and doesn't compel anyone to use the standard if they don't like it.
Is there any real reason this is bad other than that it's Microsoft? Is it preventing anyone else from doing anything else?
I just don't think you can have it both ways. You can either have everyone using the same browser or you can have a world where you have to test and tweak your web pages for every browser.
As for Vista, I agree that it's quite a screwed up OS right now. It will, I think, get better in the next few months as the most horrible bugs will get fixed.
Still, I am deeply saddened that all of the core enhancements that I expected to see in Vista failed to materialize. I don't think that's going to change. I was expecting the core to be rewritten for modern CPUs, and I don't see any evidence that's happened. Memory management, networking, scheduling and other key services that were broken in XP seem to be just as broken, of not more so, in Vista.
Why the hell doesn't apple sell OSX as a product?! Back when their hardware was ludicrously overpriced, I could understand it. Why buy crappy overpriced Macs if you could buy commodity hardware and still run OSX and Macintosh applications on it?
But now, Apple's hardware is really nice (people are even buying it to run Windows and Linux on!) and it's reasonably priced. So there is no reason not to sell OSX as a product. Does Apple have no interest in really taking on Microsoft?
My issue is that the post I'm responding to criticizes Microsoft for two things. First, it criticizes Microsoft for ignoring or failing to implement standards. Then, it criticizes Microsoft for participating in the standards process and implementing standards.
There might be a legitimate complaint in there. Perhaps he feels there is some particular reason Microsoft should implement those particular standards. Perhaps he has a beef with the way Microsoft participated in the standards process. But he doesn't say, so his post is really no better than "Microsoft is bad".
If there's a legitimate complaint in there, I'd like to know what it is. It may well be legitimate and valid, but it doesn't help anybody if it's so vague nobody can figure out what it actually is.
Frankly, I don't see that standards-compliance in IE is inherently particularly good or bad. If Microsoft doesn't implement standards, that gives competing browsers an advantage if those standards are beneficial. If they're not, there doesn't seem to be any reason Microsoft should implement them.
If Microsoft makes a good browser, that's good for everyone who uses it. If they make a bad browser, it's good for everyone who competes with it.
If they don't take ideas from other people, they are ignoring standards. If they do, they are stealing.
You need to complain with enough detail that people can figure out what your complaining about.
I guess you're just too subtle for me. Maybe you could explain what your actual complaint is rather than just saying "Microsoft = bad" and just assuming everyone knows why. Seriously, I'd love to know what your issue is, but you have to make at least some effort to explain it.
The problem is when the initialization is not quite so simple. In any event, I wouldn't use 'first(first)' under any circumstances. I agree that prefixing the inputs with 'i' (if they start with a capital letter) is better than deleting a letter.
The problem is, though, to make a tool like this useful, all your coding conventions would have to be tweaked to accommodate the tool. Abbreviations would have to be banned.
The simplest way is to buy a $40 case/powersupply, a $50 motherboard, $36 worth of RAM (1GB DDR2-667), a $50 CPU (Celeron D, for example), and a $20 CDROM drive. For less than $200 (not including a monitor), you are good to go. Use onboard sound and video.
Use any live distribution of any OS that boots from CD that you like. If anything goes wrong, just reboot. Upgrades are simple, give them a new CD. If they ever want a full computer, just add a hard drive (and a video card if desired).
If it's just used for browsing, performance will be essentially flawless. There is no host OS to maintain and no hard drive to get corrupted or fail.
The spelling error is that "functionSigniture" is not a word? Or are you suggesting that it should recognize mixed case and check each word individually? I just looked at some random code of mine, and it contains so many things that are perfectly legal that aren't English words, aside from language keywords.
For example, I have a function that provides a pretty-printed time called 'Gmtime'. The function that produces a GUID in BER form is called 'BERGUID'. In lots of places, "clear" is abbreviated as "clr". A few "Init*" functions are matched by corresponding "Deinit*" functions. An "operator new" assistant is called "OpNew".
C++ requires that a parameter passed to a function have a different name from the same class member. So it's not unusual to have two variables that do the same thing logically but must have different names. Deleting a letter from one variable name is not uncommon, like "first" and "frst".
A static version of a class member function that's used to hook into an API that doesn't support C++ natively often has an 'S' before it's name, so you may have 'ShutDown' and 'SShutDown'.
I believe that most other languages have similar issues.
You'd either have to have one crazy set of coding standards or you'd have to expect your programmers to pick the legitimate errors out of pages and pages of nonsense.
I don't think the principles in that suit would apply in this case. The decision to arrest him was made after it was firmly established that he hadn't stolen anything. At that point, any justification for a Terry stop evaporates. The officer had better be prepared to argue that he still had something to investigate.
In any event, the officer's behavior is obvious evidence of supreme incompetence.
I don't know what axe you have to grind, but your argument is nonsense.
1) I asked if you *knew* that the copyright office retained these documents so that they would be guaranteed to be available. No answer.
2) I asked if you *knew* that SYD retains timestamps after an account is terminated. They obviously can't if they're out of business. No answer.
They utterly fail to do the one thing they're supposed to do for these two reasons. Unless I can be confident that there's going to be some way for me to get to my timestamps in a provable way (not a printout with no cryptographic validator), the service fails to provide the one thing it claims to provide.
It is very simple to provide cryptographic proof that one had a document on a particular date. The document is not required to do this. Nothing SYD does prevents them from doing this as well. The fact that they don't actually do this speaks volumes, as does the fact that their timestamps are ambiguous.
As for a court trusting that SYD didn't stamp a document ex-post, I don't know what to say. You obviously don't understand digital cryptography. This is an easily solved problem. For example, you could submit the hash to two independent free public trusted timestamp providers (say, Verisign and Certum). It's hard to imagine a conspiracy involving Verisign, Certum, and SYD. It's as plausible as someone in the copyright office retroactively sneaking a document in.
The point is, their service is deliberately badly crippled to ensure that you lose your timestamps if you don't pony up the yearly fee or pay for the permanent service. As a consequence, if they go out of business, you lose your timestamps. You may or may not be able to find your timestamp among god knows how many others in the myriad submissions at the copyright office.
If you want to continue this, you need to answer two questions:
1) Is your argument that their service is useful based primarily on the copy that goes to the copyright office rather than the copy that they retain?
2) Do you have any reason to believe it would be possible or practical to get that copy from the copyright office and find your hash among the god knows how many that SYD submits?
So your argument comes down to "there might be something I don't know". Well, sure, there might be. But it's the job of their web site to make their customers comfortable with their service, and it utterly fails to do that. Again, their whole purpose is to allow you to prove you submitted a document at a particular time and they fail to do that.
You claim that you know the basics of cryptography, but have no addressed my specific explanation for why I think you don't. You misunderstand how digital signatures are generated, thinking that they cannot be produced with only the hash of the document signed. This is manifestly false, and your whole reason for why their service is designed as it is was based on this false understanding.
The BSD permits things the GPL prohibits. You can spin that all you like, but it's quite obvious that being allowed to do something makes you more free than being prohibited from doing it.
Also, you totally ignore the fact that improving proprietary software is a huge benefit to everyone who uses that software. It's so funny that the open source community blasts the closed source community for not following standards, but then they see no problem in forcing the standards to be re-implemented. Re-implementation of standards and the inability to benefit from existing implementations causes the very lack of standards-compliance the open source community complains about.
The truth is, higher quality software benefits everyone.
It really doesn't matter. The BSD does not assure that people you distribute derived works too get all the rights you got to the original code. So whether or not it's dual-licensed, you are not required to grant a BSD-style license to derived works.
Technically, you can't remove the BSD license from the file. But it will no longer apply to the modified file. It will still only apply to the original file.
The biggest difference between the BSD and the GPL is that the GPL must apply to all derivative works of GPL'd works. The BSD license does not work that way.
Slashdot Mirror
I use portable storage devices to move data from one place to another. Generally, I need to remove the old stuff to make room for the new stuff. Until these device become large enough to compete with hard drives, which they are nowhere near, they will tend to be used for volatile, mobile data.
When your laptop can easily have a 64Gb flash-based storage device, a comparable hard drive will hold 1TB. People have never been satisfied with 1/16th the storage they could have, and I don't think that's going to change.
For practical purposes, there's no difference. Every write (to a particular cell) is preceded by an erase (that includes that cell).
This was an unintended (at least by the authors) consequence of the DMCA. It needs to be fixed ... badly. If a copyright holder uses a technological mechanism that prevents fair use, it is not circumvention to bypass that mechanism.
You cannot "circumvent" a mechanism that protects a right the creator of that mechanism does not actually have.
Single-threaded applications, for cases where performance matters, will likely become a thing of the past. It has always been cheaper to make two fast cores work together than make one core twice as fast. That's a reality that Intel and AMD have kept typical application programmers from facing for much longer than most people thought possible.
... yet!
On the bright side, there are a lot of programmers out there who know how to make fast software for multiprocessor machines. The problem is, most of them don't have experience with programming things like games. One nice thing about DirectX 10 and similar high-level APIs is helps to shift the burden of taking advantage of sophisticated CPU technologies to the authors of driver code and away from the authors of application code.
I can foresee few problems with application programmers learning to take advantage of four core machines in the next decade or so. However, I can no way for such application to usefully employ more than four cores. The problem is that the types of things we need our computers to do only parallelize to a certain degree.
With a game, you can have a thread that handles network I/O, a thread that handles AI, a thread that handles physics, and a thread that handles the user interface and 3D rendering. If you try to go much past that, your threads wind up getting in each others' way more than they cooperate.
So ultimate we're going to need either each core to be faster or we're going to need radically different ways to *use* the other cores. There are a few crazy ideas out there, but none of them are anywhere close to reality
That's too indirect to count. There's no precise rule for how direct the causality has to be, but that's clearly too indirect. If mechanical problems take a plane out of service, we don't blame the mechanical problems for car accidents when people who would have taken the plane have to take cars instead.
Does anyone care to predict what affect this will have on Windows? Will people create their own complete driver sets for Windows? The hardware specifications can be used to make more than just software for Linux.
Is it just way too much effort to develop a Windows display driver? Or might we see projects to develop an open source Vista driver for ATI hardware? Might ATI even open source their Windows driver?
It's hard to see what their downside is. The upside would be that they may wind up getting better drivers without having to do all the work themselves.
I totally agree, and I boycott manufacturers who slipstream. You should be able to evaluate a product by its model number and make and be reasonably assured of getting substantially the same product if you purchase another of the same model number.
I can understand issues with parts availability and fixing design flaws. I cannot understand or defend substantive hardware changes without a change in model identifier.
IMO, reviewers should refuse to review products from manufacturers who slipstream. What is the point of reviewing a product when there is no way you can tell a consumer what product you reviewed such that they can be reasonably assured of buying and getting the product you reviewed?!
I almost never buy products that have rebates. I think they are basically a scam, for three reasons:
1) You pay sales tax on the amount rebated to you. So when you buy something with a $20 rebate in an area with 7% sales tax, you are basically having $1.40 stolen from you. A $199 rebate steals $14 from you.
2) The rebates are often not sent. You could wind up spending as much as an hour or more getting the rebate processed, and you still may not get it.
3) Too often, a product will have a higher price but a rebate at an office supply type store. The idea is that you buy the more expensive product, get reimbursed for the full price on the receipt by your employer, and then keep the rebate for yourself. The companies that promote these more-expensive products with-rebate sold next to less-expensive no-rebate products know exactly what they're doing and why they're doing it.
4) You often have to consent to your contact information being used for all kinds of marketing purposes. This often includes the manufacturer selling it.
I haven't looked at the OOXML specification, but if it's anything like the CIFS and SMB specifications, I'd have to agree with you. The CIFS specification says things like, "you either need to put X here or Y here, and it is not known how to tell which".
In any event, it looks like Microsoft deservedly lost in its attempt to standardize OOXML. And really, this (OOXML) doesn't have anything to do with browsers.
The statute says:
"The term ''interactive computer service'' means any information service, system, or access software provider that provides or enables computer access by multiple users to a computer server, including specifically a service or system that provides access to the Internet and such systems operated or services offered by libraries or educational institutions."
So, yes, it's a bit of stretch for that to cover a piece of software that only incidentally uses a server. Oddly, the law has similar protections for an "access software provider", which seems to fit a bit better:
"The term ''access software provider'' means a provider of software (including client or server software), or enabling tools that do any one or more of the following:
(A) filter, screen, allow, or disallow content;
(B) pick, choose, analyze, or digest content; or
(C) transmit, receive, display, forward, cache, search, subset, organize, reorganize, or translate content."
Seems the result should follow from (A) and (B). Malware scanners certainly filter, screen, and allow or disallow content.
That's a common myth. At one time, it might have been the truth, though it was never really tested. Now there are laws that specifically give ISPs and others the right to filter without become liable for things they filter inadvertently or things they fail to filter. This law is still relatively untested, so it's not clear exactly when it applies and who it applies to.
It was mostly intended to cover the AOL issue. AOL wanted to offer some moderation of its forums to create more child-friendly forums. But they were worried that if they tried to moderate the forums, they might become liable for anything that slipped through.
The law has recently covered identifying programs as spyware or malware. Apparently, so long as you do this in good faith, you are not liable for false positives or false negatives.
So this should cover most filtering an ISP might do. Whether it will cover *forging* packets for traffic level management, I don't know. That's quite a stretch. But the myth "you're a common carrier unless you filter" is false for many reasons.
I've had the "pleasure" of implementing other standards that were created in this way (such as CIFS). And while the standards that result are truly horrible, at least there is some documentation rather than none.
If Microsoft pushes through a truly horrible standard, that doesn't compel anyone else to implement it. It doesn't prevent anyone else from implementing a different/better standard. It does mean that anyone who has to implement it has decent documentation to follow.
Microsoft has basically documented its specification and licensed any patent rights it holds that would cover the specification. This allows, but does not force, other people to use its specification where they think it benefits them and doesn't compel anyone to use the standard if they don't like it.
Is there any real reason this is bad other than that it's Microsoft? Is it preventing anyone else from doing anything else?
I just don't think you can have it both ways. You can either have everyone using the same browser or you can have a world where you have to test and tweak your web pages for every browser.
As for Vista, I agree that it's quite a screwed up OS right now. It will, I think, get better in the next few months as the most horrible bugs will get fixed.
Still, I am deeply saddened that all of the core enhancements that I expected to see in Vista failed to materialize. I don't think that's going to change. I was expecting the core to be rewritten for modern CPUs, and I don't see any evidence that's happened. Memory management, networking, scheduling and other key services that were broken in XP seem to be just as broken, of not more so, in Vista.
Why the hell doesn't apple sell OSX as a product?! Back when their hardware was ludicrously overpriced, I could understand it. Why buy crappy overpriced Macs if you could buy commodity hardware and still run OSX and Macintosh applications on it?
But now, Apple's hardware is really nice (people are even buying it to run Windows and Linux on!) and it's reasonably priced. So there is no reason not to sell OSX as a product. Does Apple have no interest in really taking on Microsoft?
My issue is that the post I'm responding to criticizes Microsoft for two things. First, it criticizes Microsoft for ignoring or failing to implement standards. Then, it criticizes Microsoft for participating in the standards process and implementing standards.
There might be a legitimate complaint in there. Perhaps he feels there is some particular reason Microsoft should implement those particular standards. Perhaps he has a beef with the way Microsoft participated in the standards process. But he doesn't say, so his post is really no better than "Microsoft is bad".
If there's a legitimate complaint in there, I'd like to know what it is. It may well be legitimate and valid, but it doesn't help anybody if it's so vague nobody can figure out what it actually is.
Frankly, I don't see that standards-compliance in IE is inherently particularly good or bad. If Microsoft doesn't implement standards, that gives competing browsers an advantage if those standards are beneficial. If they're not, there doesn't seem to be any reason Microsoft should implement them.
If Microsoft makes a good browser, that's good for everyone who uses it. If they make a bad browser, it's good for everyone who competes with it.
If they don't take ideas from other people, they are ignoring standards. If they do, they are stealing.
You need to complain with enough detail that people can figure out what your complaining about.
I guess you're just too subtle for me. Maybe you could explain what your actual complaint is rather than just saying "Microsoft = bad" and just assuming everyone knows why. Seriously, I'd love to know what your issue is, but you have to make at least some effort to explain it.
First you complain that Microsoft ignores standards. Then you complain that they steal standards. What's the option you'd like them to take?
If all the obvious reasons to pick Linux over Solaris go away, it will be very interesting to see which OS people choose.
The problem is when the initialization is not quite so simple. In any event, I wouldn't use 'first(first)' under any circumstances. I agree that prefixing the inputs with 'i' (if they start with a capital letter) is better than deleting a letter.
The problem is, though, to make a tool like this useful, all your coding conventions would have to be tweaked to accommodate the tool. Abbreviations would have to be banned.
I just don't see how it's worth it.
The simplest way is to buy a $40 case/powersupply, a $50 motherboard, $36 worth of RAM (1GB DDR2-667), a $50 CPU (Celeron D, for example), and a $20 CDROM drive. For less than $200 (not including a monitor), you are good to go. Use onboard sound and video.
Use any live distribution of any OS that boots from CD that you like. If anything goes wrong, just reboot. Upgrades are simple, give them a new CD. If they ever want a full computer, just add a hard drive (and a video card if desired).
If it's just used for browsing, performance will be essentially flawless. There is no host OS to maintain and no hard drive to get corrupted or fail.
The spelling error is that "functionSigniture" is not a word? Or are you suggesting that it should recognize mixed case and check each word individually? I just looked at some random code of mine, and it contains so many things that are perfectly legal that aren't English words, aside from language keywords.
For example, I have a function that provides a pretty-printed time called 'Gmtime'. The function that produces a GUID in BER form is called 'BERGUID'. In lots of places, "clear" is abbreviated as "clr". A few "Init*" functions are matched by corresponding "Deinit*" functions. An "operator new" assistant is called "OpNew".
C++ requires that a parameter passed to a function have a different name from the same class member. So it's not unusual to have two variables that do the same thing logically but must have different names. Deleting a letter from one variable name is not uncommon, like "first" and "frst".
A static version of a class member function that's used to hook into an API that doesn't support C++ natively often has an 'S' before it's name, so you may have 'ShutDown' and 'SShutDown'.
I believe that most other languages have similar issues.
You'd either have to have one crazy set of coding standards or you'd have to expect your programmers to pick the legitimate errors out of pages and pages of nonsense.
I don't think the principles in that suit would apply in this case. The decision to arrest him was made after it was firmly established that he hadn't stolen anything. At that point, any justification for a Terry stop evaporates. The officer had better be prepared to argue that he still had something to investigate.
In any event, the officer's behavior is obvious evidence of supreme incompetence.
I don't know what axe you have to grind, but your argument is nonsense.
1) I asked if you *knew* that the copyright office retained these documents so that they would be guaranteed to be available. No answer.
2) I asked if you *knew* that SYD retains timestamps after an account is terminated. They obviously can't if they're out of business. No answer.
They utterly fail to do the one thing they're supposed to do for these two reasons. Unless I can be confident that there's going to be some way for me to get to my timestamps in a provable way (not a printout with no cryptographic validator), the service fails to provide the one thing it claims to provide.
It is very simple to provide cryptographic proof that one had a document on a particular date. The document is not required to do this. Nothing SYD does prevents them from doing this as well. The fact that they don't actually do this speaks volumes, as does the fact that their timestamps are ambiguous.
As for a court trusting that SYD didn't stamp a document ex-post, I don't know what to say. You obviously don't understand digital cryptography. This is an easily solved problem. For example, you could submit the hash to two independent free public trusted timestamp providers (say, Verisign and Certum). It's hard to imagine a conspiracy involving Verisign, Certum, and SYD. It's as plausible as someone in the copyright office retroactively sneaking a document in.
The point is, their service is deliberately badly crippled to ensure that you lose your timestamps if you don't pony up the yearly fee or pay for the permanent service. As a consequence, if they go out of business, you lose your timestamps. You may or may not be able to find your timestamp among god knows how many others in the myriad submissions at the copyright office.
If you want to continue this, you need to answer two questions:
1) Is your argument that their service is useful based primarily on the copy that goes to the copyright office rather than the copy that they retain?
2) Do you have any reason to believe it would be possible or practical to get that copy from the copyright office and find your hash among the god knows how many that SYD submits?
So your argument comes down to "there might be something I don't know". Well, sure, there might be. But it's the job of their web site to make their customers comfortable with their service, and it utterly fails to do that. Again, their whole purpose is to allow you to prove you submitted a document at a particular time and they fail to do that.
You claim that you know the basics of cryptography, but have no addressed my specific explanation for why I think you don't. You misunderstand how digital signatures are generated, thinking that they cannot be produced with only the hash of the document signed. This is manifestly false, and your whole reason for why their service is designed as it is was based on this false understanding.
The BSD permits things the GPL prohibits. You can spin that all you like, but it's quite obvious that being allowed to do something makes you more free than being prohibited from doing it.
Also, you totally ignore the fact that improving proprietary software is a huge benefit to everyone who uses that software. It's so funny that the open source community blasts the closed source community for not following standards, but then they see no problem in forcing the standards to be re-implemented. Re-implementation of standards and the inability to benefit from existing implementations causes the very lack of standards-compliance the open source community complains about.
The truth is, higher quality software benefits everyone.
It really doesn't matter. The BSD does not assure that people you distribute derived works too get all the rights you got to the original code. So whether or not it's dual-licensed, you are not required to grant a BSD-style license to derived works.
Technically, you can't remove the BSD license from the file. But it will no longer apply to the modified file. It will still only apply to the original file.
The biggest difference between the BSD and the GPL is that the GPL must apply to all derivative works of GPL'd works. The BSD license does not work that way.