I worried a bit that Amazon might discontinue their service someday, in a way that would break books people have already bought, but then I realized that this didn't really matter to me. Different people will weigh things differently.
Looking at my modest physical library (a couple thousand volumes or so), I note that most of them have only been read once.
Kindle books (at least the ones I've bought, and the ones on my current to-buy list) are about 20-30% off the least expensive physical edition.
If Amazon does NOT end up screwing us down the line, then I'll be in the same position as I am with my physical library. I'll have a bunch of books that I am not going to read again, and a handful that I do reread. Except I'll have saved a lot of money. Works for me!
And if Amazon DOES screw us someday? Then I use some of that 20-30% I've saved to re-purchase physical copies of the handful of books that I will want to reread. I'll still end up with a library that contains all the books I actually will want to reread. It will simply be missing the books that I only wanted to read once. But it will probably have cost me less for that library. Seems like a good gamble to me.
It is kind of interesting to compare this to music with DRM. With music, I do listen to most of my albums more than once. If my albums were to go away, I'd want to replace pretty much all of them.
Thus, for music, I am much more DRM-adverse. I have bought a few things from the iTunes store, but it has been obscure things that I could not reasonably find on CD, and there is the old "burn and rip" method to keep them working even if Apple pulls the plug. I also figured that disk space would be cheap and plentiful enough that if I did have to do "burn and rip", I could do the rip losslessly, and so this method of stripping DRM would not lead to a loss of quality. Thus, I had things covered, and could go ahead and buy a few things from iTunes. But I buy from the Amazon DRM-free music store if I can.
If they just decided they wanted to leave their town without telling anyone, you've committed a huge breach of their personal privacy for nothing.
They could leave their cell phone behind, or remove the battery if they want to take the cell phone with them. Net result: a very slight inconvenience on those who want to leave town without telling someone.
If they win against Google, I wonder how much farther they will go?
For example, if I go to Amazon and search for some specific, branded, product, Amazon includes on that product's page a list of products that people shopping for that product bought. These are often competing products. Will the trademark owners object to this?
If you are going to be as picky as Brown was there, then you run into the problem that ODF suffers from the same problem. OpenOffice often produces ODF documents that do not strictly follow the standard. The deviations of Office from transitional OOXML are no more severe than the deviations of OpenOffice from the ISO version of ODF.
If, on the other hand, you are going to cut enough slack to allow OpenOffice to count as implementing ODF, then anything that implements ECMA-376 is also at least a transitional OOXML implementation, and then besides Office and Office Mac, there is Pages and Numbers from Apple, NeoOffice on the Mac, TextEdit on the Mac, at least three open source OOXML/ODF translators, Thinkfree Office, Dataviz, and Google.
You are the one making the positive claim, so you are the one that has the burden of proof, under all normal standards of argument. Show us the binary chunks.
Just in case you're not trolling; ISO standards should be independently implementable by anyone. OOXML cannot be independently implemented. Therefor MS should not have submitted & ISO body should not have approved.
Because the standard is missing lots of information, There are large parts of the Office XML files that contain chunks of binary, which we do not understand, this should be in the specification somewhere - but it isn't
That is simply not true. If OOXML opponents had stuck to criticism that was actually true, instead of blatant lies like the binary chunk claim, they would have had a better chance of winning.
Option 1. You visit your client in person, carry out a conversation, and come back. The government asks what you talked about, and you refuse to talk, citing attorney client privilege. End of story.
Option 2. You engage in PGP email exchange with client. Government tries to subpoena the encryption key from you, but does not ask for the emails themselves. You say no. You now get involved in a long court fight over whether or not attorney client privilege covers the keys.
Excuse my ignorance, but please explain how this this different (or superior) to doxygen or any of the many systems that do just this
Those systems don't rearrange code. You give them program files containing code and documents, and they pull the documentation for humans out of that. The program files, though, go straight to the compiler. Thus, the files you write (the program files), have to be organized the way the compiler wants them to be organized, and that generally imposes an organization on the documentation that is pulled from the program files.
In literate programming systems, the files you write are NOT passed through to the compiler. The tools pull from your input files the code for the compiler, and the documentation for humans. Your documentation is no longer constrained by the limits of the compiler.
This is important because the best order for presenting code to a human is often different from the best order to present that code to a compiler. Literate programming toolsets can deal with this. Similarly, some things that have to (or should be for efficiency) inline code in the compiler input might make more sense in documentation if they were more like subroutines. Literate programming tools can deal with that, by essentially having a macro mechanism.
Know why? There are people that don't realise how bad are DRM downloads until they get royally fucked in the ass and this is what's going to happen on sept 1 2008
When Google did the same thing in August 2007 with video, it didn't seem to have all that much effect. Why will this be different?
It was the demotion of Bender that prompted Ivan Krstic' to resign last month, so the damage to OLPC by their stupid demotion of Bender is not limited to just the loss of him. I wonder if anyone else will be leaving over this?
Those are all interesting questions. And they would have just as interesting months ago, when ODF was being standardized, as EXACTLY THE SAME THING happened there. A large number of technical issues raised about ODF were "resolved" by "OASIS will fix that in a later revision of the standard".
A very large number of the criticism of OOXML is about things where OOXML was treated exactly the same as most prior standards, including ODF. For God's sake, ODF got away with pushing spreadsheet formulas off to a later version--how can you have a viable office document format when spreadsheet formulas aren't specified? (This is addressed in ODF 1.2).
I suspect it's an un-Constitutional form of inter-state commerce taxation by RI, but there you have it.
It's Constitutional, as long as it is not discriminatory against out-of-state items.
What that means, in a nutshell, is it has to be the same rate as the sales tax rate, so that the net result is that a given item is taxes the same, regardless of origin.
Vista won't run well on the increasingly popular lightweight and low end laptops like the eepc, olpc xo, and what are sure to be many imitators. People have demonstrated they're willing to use linux on these machines, and Microsoft has demonstrated they Don't Get It
I'd prefer to see people pushing Linux on high end machines. The spate of low end laptops with Linux, combined with the puzzling propensity of penguin proselytizers to promote Linux on old hardware that would otherwise be throw out, is going to give Linux an image of being what you run when you can't afford something better.
Aim for the high end, not the low end. That's the way to get people interested.
Why is that even an issue? ODF passed, it's a clear and well-defined standard that nobody has a problem with and nobody had to be bribed to support
You haven't talked to anyone who has actually tried to implement ODF from the spec, have you? It is not very well-defined. For example, do you know how it handles password hashing? It just says you should do it. No list of allowed hashes. No documented way to record what hash you've used.
Want another example? Calendars. There it at least lists the names of the allowed calendar system. But no reference to what those names mean.
The fact is that to implement ODF in a fairly complete fashion (no one has ever done a complete implementation), and have your implementation interoperate with other implementations, you have to base it off the OpenOffice source code, and that's what everyone has done (some indirectly, by basing theirs off code that is based off OpenOffice).
Compare to OOXML. It lists all the allowed hashes, and cites to the specification for each one. Same for calendars. It actually gives you enough information to implement, unlike ODF.
And the funny thing is, these are both areas OOXML was slammed on for being inadequate, even though it was vastly more well-defined than ODF in these areas, even on its very first submission. This nicely illustrates the hypocrisy of the anti-OOXML crowd. A good 90% of their objections to OOXML were either things like the hash handling, where OOXML was much better than ODF, or were flat out untruths.
If Microsoft wanted the Office 20xx standard to be an open standard, it could have joined the ODF forum, when everyone wanted it to, and pushed the DOC format into ODF
Sun shot that idea down back when it was first proposed. They said ODF would not be allowed to have any features not needed by StarOffice.
Some people object to DRM on ideological grounds, but not many
And I object to many of those objectors on ideological grounds.
* They say things like DRM is slavery, and if I were to dare to buy, say, a Kindle, I am helping promote slavery. That's an insult to all the people who have been victims of actual slavery.
* I'm an adult. I can look at the terms of, say, Amazon's DRM (to continue using the Kindle for my examples), and weigh the conveniences the device offers, vs. the risk of being screwed if Amazon pulls a Google and cancels the service with no refunds, and the annoyance of not being able to loan books to friends, and so on. The ideological DRM objectors seem to dismiss my adulthood, and want to treat me like a child who has to be protected from making the "wrong" decision and willing buying a DRMed product. They are no better than those who want to "protect" me from seeing porn on the internet, or from finding heretical books in my local library.
It's basically two things: a use tax, and a scheme to collect it.
The Supreme Court cases on State use taxes are clear: they are Constitutional.
So that just leaves their scheme to collect. The obvious problem here is the Quill case, but they seem to have found a somewhat plausible argument to distinguish from that, so I could see this go either way on that.
Slashdot Mirror
I worried a bit that Amazon might discontinue their service someday, in a way that would break books people have already bought, but then I realized that this didn't really matter to me. Different people will weigh things differently.
Looking at my modest physical library (a couple thousand volumes or so), I note that most of them have only been read once.
Kindle books (at least the ones I've bought, and the ones on my current to-buy list) are about 20-30% off the least expensive physical edition.
If Amazon does NOT end up screwing us down the line, then I'll be in the same position as I am with my physical library. I'll have a bunch of books that I am not going to read again, and a handful that I do reread. Except I'll have saved a lot of money. Works for me!
And if Amazon DOES screw us someday? Then I use some of that 20-30% I've saved to re-purchase physical copies of the handful of books that I will want to reread. I'll still end up with a library that contains all the books I actually will want to reread. It will simply be missing the books that I only wanted to read once. But it will probably have cost me less for that library. Seems like a good gamble to me.
It is kind of interesting to compare this to music with DRM. With music, I do listen to most of my albums more than once. If my albums were to go away, I'd want to replace pretty much all of them.
Thus, for music, I am much more DRM-adverse. I have bought a few things from the iTunes store, but it has been obscure things that I could not reasonably find on CD, and there is the old "burn and rip" method to keep them working even if Apple pulls the plug. I also figured that disk space would be cheap and plentiful enough that if I did have to do "burn and rip", I could do the rip losslessly, and so this method of stripping DRM would not lead to a loss of quality. Thus, I had things covered, and could go ahead and buy a few things from iTunes. But I buy from the Amazon DRM-free music store if I can.
They missed Fog Creek.
They could leave their cell phone behind, or remove the battery if they want to take the cell phone with them. Net result: a very slight inconvenience on those who want to leave town without telling someone.
If they win against Google, I wonder how much farther they will go?
For example, if I go to Amazon and search for some specific, branded, product, Amazon includes on that product's page a list of products that people shopping for that product bought. These are often competing products. Will the trademark owners object to this?
If you are going to be as picky as Brown was there, then you run into the problem that ODF suffers from the same problem. OpenOffice often produces ODF documents that do not strictly follow the standard. The deviations of Office from transitional OOXML are no more severe than the deviations of OpenOffice from the ISO version of ODF.
If, on the other hand, you are going to cut enough slack to allow OpenOffice to count as implementing ODF, then anything that implements ECMA-376 is also at least a transitional OOXML implementation, and then besides Office and Office Mac, there is Pages and Numbers from Apple, NeoOffice on the Mac, TextEdit on the Mac, at least three open source OOXML/ODF translators, Thinkfree Office, Dataviz, and Google.
You are the one making the positive claim, so you are the one that has the burden of proof, under all normal standards of argument. Show us the binary chunks.
OOXML can and has been independently implemented.
That is simply not true. If OOXML opponents had stuck to criticism that was actually true, instead of blatant lies like the binary chunk claim, they would have had a better chance of winning.
How is this any different than getting full access to a Linux computer using a live CD?
Option 1. You visit your client in person, carry out a conversation, and come back. The government asks what you talked about, and you refuse to talk, citing attorney client privilege. End of story.
Option 2. You engage in PGP email exchange with client. Government tries to subpoena the encryption key from you, but does not ask for the emails themselves. You say no. You now get involved in a long court fight over whether or not attorney client privilege covers the keys.
Option 1 seems the better way to go.
In other words, you'd never convict someone unless you personally witnessed the crime.
In pretty much every other case, there can be another reasonable explanation.
Those systems don't rearrange code. You give them program files containing code and documents, and they pull the documentation for humans out of that. The program files, though, go straight to the compiler. Thus, the files you write (the program files), have to be organized the way the compiler wants them to be organized, and that generally imposes an organization on the documentation that is pulled from the program files.
In literate programming systems, the files you write are NOT passed through to the compiler. The tools pull from your input files the code for the compiler, and the documentation for humans. Your documentation is no longer constrained by the limits of the compiler.
This is important because the best order for presenting code to a human is often different from the best order to present that code to a compiler. Literate programming toolsets can deal with this. Similarly, some things that have to (or should be for efficiency) inline code in the compiler input might make more sense in documentation if they were more like subroutines. Literate programming tools can deal with that, by essentially having a macro mechanism.
That has nothing to do with literate programming. That Wikipedia article is very poorly written.
You can find an example of the old classic ADVENTURE game, ported to C, here, in the file adventure.zip.
Run that through the appropriate tools, and you end up with code to feed the compiler, and a listing to give humans. The listing for humans is here.
The Clarke story they should be looking at is Superiority.
When Google did the same thing in August 2007 with video, it didn't seem to have all that much effect. Why will this be different?
It was the demotion of Bender that prompted Ivan Krstic' to resign last month, so the damage to OLPC by their stupid demotion of Bender is not limited to just the loss of him. I wonder if anyone else will be leaving over this?
Those are all interesting questions. And they would have just as interesting months ago, when ODF was being standardized, as EXACTLY THE SAME THING happened there. A large number of technical issues raised about ODF were "resolved" by "OASIS will fix that in a later revision of the standard".
A very large number of the criticism of OOXML is about things where OOXML was treated exactly the same as most prior standards, including ODF. For God's sake, ODF got away with pushing spreadsheet formulas off to a later version--how can you have a viable office document format when spreadsheet formulas aren't specified? (This is addressed in ODF 1.2).
It's Constitutional, as long as it is not discriminatory against out-of-state items.
What that means, in a nutshell, is it has to be the same rate as the sales tax rate, so that the net result is that a given item is taxes the same, regardless of origin.
I'd prefer to see people pushing Linux on high end machines. The spate of low end laptops with Linux, combined with the puzzling propensity of penguin proselytizers to promote Linux on old hardware that would otherwise be throw out, is going to give Linux an image of being what you run when you can't afford something better.
Aim for the high end, not the low end. That's the way to get people interested.
You haven't talked to anyone who has actually tried to implement ODF from the spec, have you? It is not very well-defined. For example, do you know how it handles password hashing? It just says you should do it. No list of allowed hashes. No documented way to record what hash you've used.
Want another example? Calendars. There it at least lists the names of the allowed calendar system. But no reference to what those names mean.
The fact is that to implement ODF in a fairly complete fashion (no one has ever done a complete implementation), and have your implementation interoperate with other implementations, you have to base it off the OpenOffice source code, and that's what everyone has done (some indirectly, by basing theirs off code that is based off OpenOffice).
Compare to OOXML. It lists all the allowed hashes, and cites to the specification for each one. Same for calendars. It actually gives you enough information to implement, unlike ODF.
And the funny thing is, these are both areas OOXML was slammed on for being inadequate, even though it was vastly more well-defined than ODF in these areas, even on its very first submission. This nicely illustrates the hypocrisy of the anti-OOXML crowd. A good 90% of their objections to OOXML were either things like the hash handling, where OOXML was much better than ODF, or were flat out untruths.
Sun shot that idea down back when it was first proposed. They said ODF would not be allowed to have any features not needed by StarOffice.
And I object to many of those objectors on ideological grounds.
* They say things like DRM is slavery, and if I were to dare to buy, say, a Kindle, I am helping promote slavery. That's an insult to all the people who have been victims of actual slavery.
* I'm an adult. I can look at the terms of, say, Amazon's DRM (to continue using the Kindle for my examples), and weigh the conveniences the device offers, vs. the risk of being screwed if Amazon pulls a Google and cancels the service with no refunds, and the annoyance of not being able to loan books to friends, and so on. The ideological DRM objectors seem to dismiss my adulthood, and want to treat me like a child who has to be protected from making the "wrong" decision and willing buying a DRMed product. They are no better than those who want to "protect" me from seeing porn on the internet, or from finding heretical books in my local library.
How so?
It's basically two things: a use tax, and a scheme to collect it.
The Supreme Court cases on State use taxes are clear: they are Constitutional.
So that just leaves their scheme to collect. The obvious problem here is the Quill case, but they seem to have found a somewhat plausible argument to distinguish from that, so I could see this go either way on that.
Where did you get the impression that it is meant for that kind of thing?
I don't see Red Hat's brief listed, but here are links to most of the amicus briefs filed in this case, along with a short summary of each.