"Now that Warner Brothers (read clueless Muggles) owns Harry
Potter body, soul, and Nimbus 2001,
Come on, any fan can tell you that Harry doesn't have a Nimbus 2001, Draco Malfoy and the rest of the Slytherins have them. Harry once had a Nimbus 2000, but now he has a Firebolt. If you're going to try to make a topical reference to the books, at least try to get the facts straight.
He tried to get it published several times before the
Hobbit was even written, and nearly gave up in frustration. If he had, we never would have had the Hobbit, nor the
LotR.
The really remarkable thing about the Silmarillion, IMO, is that it was apparently never really finished. Tolkein continued to revise and completely rewrite sections well after the Hobbit and the Lord of the Rings were published. Part of the reason that the Silmarillion is of such uneven tone and polish throughout is that some of the stories hadn't been revised since the 1930's, while others had been rewritten shortly before his death. The Silmarillion as published was really a compilation of stories in different stages of revision, and even from slightly different stages of his conception of the underlying story, edited just enough to ensure coherence, as they existed at the time of Tolkein's death.
This stuff is revealed (in inordinate detail) in the whole History of Middle Earth series, which is basically his son's editing and deciphering of his papers. If you're a hardcore Tolkein fan, or just somebody who's very interested in the creative process, it's fascinating to see how the stories developed over time. What is particularly interesting is the way in which elements that were minor and trivial in the earliest versions gradually became more important, and vice versa. It makes you wonder if such a thing could even be possible today, given the way that people tend to overwrite their old word processing files instead of leaving paper copies of their work at different stages around to be looked over by later scholars.
But a number of the inventions (just about all, in fact) mentioned in the article were not completely eliminated, just largely displaced by alternatives. I think that's the general way of things, in fact. Old technology is rarely eliminated at a stroke. Instead, it's bumped into a niche market where it's particular advantages are significat enough to keep it alive. Over the very long haul that niche may dry up completely, but it rarely happens overnight.
In theory you're probably right. Since the GPL grants specific exemptions to traditional copyright restrictions, hackers will no longer need the rights granted by the GPL to hack the code after the copyright expires. In practice, though, that's not likely to make much difference. For one thing, copyright has been continually extended to the point that there's serious reason to think that it will never actually expire. More practically, though, modifications to the code are under copyright from the time that they're written, not the time that the original code was written. That means that when the copyright expires, you only get the non-GPL right to the oldest, cruftiest parts of the code. The new stuff that's likely to make any difference to the computer on your desk rather than one in a museum will still be under copyright.
The real problem is that people want to advertize, and they're going to keep trying to think up ways of doing so to keep one step ahead of the blockers. The real danger is that the ultimate in advertizing is not far away: product placement. Of course with media conglomeration this is already at least somewhat underway, but it's only going to get worse. How long is it going to be until web sites start incorporating ads for their commercial partners into the main content of their pages? I guess that search engines are already well down that road, from Google's "sponsored link" to goto.com's outright selling of placement to many other search engines' more insidious under-the-counter acceptance of pay for better search results. How long until other web sites start similar practices?
But openness is a necessary precondition for the freedom to continue hacking. If I want to excercise my freedom to hack your code, I must first have that code; you can't hack what you can't see. A freedom with no opportunity to excercise it is no freedom at all. Therefore preserving openness is essential to preserving freedom.
5. Free Software license - license for authors who whish to guarantee that all derivative works based on their code are Free Software.
This means that you could mix my code with software under any free software license, assuming that the license of the other software would not be violated.
This is much harder than you think, and probably impossible in the long run. Why? Because some of the licenses that your code might wind up in (like the BSD license) are specifically designed to allow incorporation of their source into non-free projects. Once your code was put into a BSD licensed project, it could then be transfered into a proprietary one. To protect your intent, they'd have to relicense the part of their project that incorporated your code under a different style of license, which specifically undermines your goal of letting your code play friendly with everyone else. It's just not possible to have code that's useable by any free software project but not by a proprietary one.
Perhaps some day, people will understand that the GPL does not make code free. Freedom is not something
which can be forced, for if it is forced, it is not freedom.
You're right that the GPL doesn't make code free. Code is made free when its author(s) decide to make it free. What the GPL does is to keep code free. Once code is GPLed, it can't be taken back and put into a completely unfree box. Given the number of times that free software has been made non-free in the past, though, I think that GPL advocates have a good argument that strong protection is needed.
The GPL sets up rules from which it's impossible to escape. It does so in the name of freedom, to keep code from falling into the paws of capitalists.
Actually, the restrictions are not there to hamper capitalists. The FSF openly advocates selling software for as much money as people will pay you for it. The restrictions are there to prevent people from distributing modified code without making the modifications available. Red Hat, Mandrake, Caldera, etc. would vigorously argue with your assertion that the GPL prevents capitalists from making money from software!
This restraint is in the name of freedom. How can a restraint be in the name of freedom? By enabling a bigger freedom. How, though, does the GPL make anything more free than public domain?
By ensuring that changes remain public. Code that is put in the public domain or BSD licenced can be absorbed into non-free projects without making any changes publically available, and frequently has been. The GPL essentially says that the right to see and further modify code is more important than the right to keep modifications private. (Note, though, that the GPL does allow private modifications so long as they are private; it only prevents distribution of modified binaries without the modified sources.) IMHO, that's a reasonable assertion, especially considering the way that those who aren't interested in sharing have taken advantage of those who are in the past. This is a legitimate point of contention, but I think that experience is on the side of the GPL.
It appears that he stuck in a hyperlink with a very long, all lower case HREF to reduce the percent of the comment that was in caps to below the filter threshold. He avoided having the link actually work by giving it a zero length field to link to, so your browser doesn't display it. It appears that the lameness filter needs some work.
I guess that my personal opinion is that Google is an excellent example of why patents exist, and that a comparison with the much blasted on-click shopping patent reveals why one is deserving and the other is not. The underlying logic behind patents is to encourage an inventor who might otherwise keep his invention secret to divulge his idea instead. The patent does this by giving him what he would try to achieve by secrecy- the exlusive right to his invention- in exchange for making the idea available to all after a certain time.
The key is that this implies that to be patent worthy an idea must be not only hard enough to think of that it's worth giving someone exclusive rights to in exchange for revealing it and also not something that is inherently revealed by the process of trying to implement it. The first criterion is the basis for obviousness and prior art restrictions. If something is so easy to invent that it's likely to be reinvented regularly, or even worse so obvious that somebody else has already published it, it's clearly not worth giving somebody exclusive rights to it in exchange for divulging it. The second criterions, which is the real objection to business model and similar patents, is that if you have to reveal your idea to use it, there's no point in granting a patent because there's no need for leverage to force the inventor to divulge it.
I think that Page Rank meets both of my criteria, and one-click doesn't meet either. Page Rank is not an obvious development of standard citation analysis, since it incorporates aspects of citation that are a product of the unique nature of the web. I think also that the fact that nobody had come up with the idea yet despite the fact that people had been thinking about web searching for quite some time is further, indirect evidence that the idea was non-obvious. One-click, though, was obvious enough that other sites were incorporating many of the same ideas (particularly the use of persistent cookies to prevent shoppers from having to re-enter their data every time) into their shopping at about the same time that Amazon did. The revelation side is even more clear. Google could have kept the guts of Page Rank hidden very easily; we only know as much about how it works as we do because they've told people about it. If they had decided just to say that they had a unique, effective new way of indexing sites, the chances are that you couldn't figure out how they were doing it just by looking at their search results. One click shopping, though, reveals itself by its very existence; in order to implement it you have to give people enough of an idea about what you're doing that it would be trivial for anyone to reimplement it.
OTOH, AIUI Google is more complex than a simple citation count, since the contribution from the citing papers is weighted by how important those papers themselves are, based on their own citations... I haven't heard that traditional citation analysis does that. But I don't know that much about the details.
I assume that the reason that traditional citation analysis doesn't do this is because traditional citations only go one way. You can only cite something that was published at the time that you wrote your work, so you can't cite anything that cites you. This greatly complicates the process, since inherently the recent papers that should have the best grasp of which older papers were most influential can't have many citations yet and hence won't themselves be considered very important. Because publication on the web is dynamic, though, this is not a problem, and bidirectional citation is possible.
The other cool thing about Page Rank is that it isn't just a recursive analysis. Instead it uses a mathematical trick that is roughly equivalent to doing a recursive analysis except that it turns up more data because it can find substantially non-overlapping networks of sites that would otherwise be missed. Suppose, for instance, that you have a topic on which opinion is very polarized (abortion was the one that I've heard this first turned up in). You would tend to have strongly opinionated sites that link almost exclusively to other sites that share the same opinion. In a straight recursive analysis, one side of the issue will have fewer total links and hence less credibility, which then in turn deprives its linked-to sites of credibility, etc. The result is that one side is effectively silenced by the simple analysis. The trick used by Page Rank not only saves computer time, it also turns up the second cluster of sites and can bump them up the list.
Re:Nevertheless, it's not too different from ...
on
A Pair of Google Bits
·
· Score: 1
But Page Rank isn't an algorithm. It's the application of an existing algorithm to a novel area of interest. The beauty of Google's system is that it's applying quite standard concepts from Linear Algebra to a novel application. The proof that the concepts are novel is that despite the fact that people have been doing citation analysis for some time, apparently nobody had come up with the idea of recursively analyzing the reliability of citations by taking into account the quality of the citing source. Nor, for that matter, had they gone the additional step and proven that this is mathematically equivalent to finding the eigenvectors of a citation matrix, which is the neat mathematical trick that makes it computationally reasonable. This is, IMO, exactly the type of powerful, novel development that deserves to be given patent protection.
Some of us think that approaches to a problem that are both non-obvious and extremely powerful (which, IMO at least, is true of Page Rank) are exactly the reason that patents were invented. This is not a patent that is:
Overly broad so that it covers everything under the sun,
Extremely obvious to the point that it's clearly not patentable, or
Actually not patentable because there's prior art.
Patents exist precisely to protect inventions that don't fall into one or more of the above categories. It's not so much patents that most slashdotters are angry about, it's patent abuses, and Google's patent clearly isn't abusing the system.
One thing that I should emphasize, though, is that the GoType keys are small. Much too small for me to
touch-type on them. In fact, they were too small for me to use them properly at all, even hunt-and-peck, and I do
not have big fingers.
This was exactly my feeling when I looked at both of them, so I tend to disagree with the article when they say that the Go Type gives you better value. IMO it's totally useless as a keyboard, so you're wasting your money when buying it. If it doesn't work right, it's not a good deal no matter how cheap it is.
One thing that bothered me deeply about the article is that the author made very little distinction between hardware and software quality problems. I realize that this probably reflects the majority, non-tech view of computers (the computer is a unit that succeeds or fails as a unit, not as hardware or software) but it made the article less comprehensible. Most of the specific problems that he talked about sound as though they were hardware problems, but the experts he consulted were talking about curing software reliability. This is probably reasonable, since reliable, high quality hardware is available, and the companies that produce junk often go out of business when people stop buying their crap. Of course high quality, reliable software is available too, but most desktop PCs don't use it. I'm inclined to agree with the closing statement of the article: we won't get high quality software until companies suffer financially from putting out crappy software.
The problem is that much of the web seems to be designed by graphic artists coming from a print background rather than computer people. As such, they're deeply attached to the idea of fixing their work precisely, as paper printing both rewards and demands. They simply haven't adapted to the new format, and I get the impression that this actually gives something of an advantage to the inexperienced over the experienced. People who have never done graphics design before are more able to accept the limitations (and strengths) of the web than people transfering from another field. That's not to say that some traditional graphic artists haven't been able to make the transition effectively, but I think that it's a handicap.
The guy who came up with this isn't a lame-o he's a philantropist who uses the company dollar
So was the guy who came up with the AOL CD. He apparently decided that American coffee tables were under critical threat of being destroyed by condensation stains. His brilliant response was to provide every home in the country with an endless supply of free coasters.
Second, there is the issue of visual formatting. LaTeX and HTML both, in theory, are based on the principle of content-based markup- you specify the data in content terms, and the browser/LaTeX engine determines how best to format it for display. Anyone who has ever used either of these languages knows that this is a total lie, especially for HTML.
All that shows is that people have unreasonable expectations for HTML. It was never intended for hyper-precise description of the layout of the material. If you use it the way that it was really indended to be used, i.e. to make an acceptable layout from content descriptive markup, it is quite capable of doing a good job. It is even quite possible to make nice looking, artistically designed pages that way without using incredibly elaborate formatting. It's when people fight against the intent of the format and try to use it as a page description language that they start having to bend, fold, spindle, staple, and mutilate it. Hint: if your page expects the browser window to be a particular size, so that it won't fit if it's 10% too narrow or has an unattractive white gutter if it's 10% too wide, you're doing something wrong.
But honestly, the versions of those programs under Gnome (and also KDE, AFAIK) are actually much superior to their Windows equivalents. The graphics are substantially better (I was amazed at how lousy MS Solitare looked when I booted back into Windows) and AisleRiot has a much larger selection of solitare games to choose from. You also get such classics as Mahjong, XBill, Robots, etc. The built in games are a real improvement over Windows.
How about looking at tools to allow insecure software to be used without compromising the integrity of the system? The effort to produce a secure OS base is largely wasted if adding new programs trashes the security. It's nice to have an OS like OpenBSD as the basis for high security bits like firewalls, but it's never going to get beyond a niche market if the security evaporates the moment that desktop/workstation applications are installed. What kinds of things is OpenBSD doing to help make it easier for developers to make secure applications?
Not necessarily. It can run on other Unixlike OSes, too, which is part of the reason that major vendors like Sun and HP are backing the Gnome Foundation. You can certainly run it with Free/Open/NetBSD. FWIW, I think that Dell is actually shipping RedHat, but there certainly wouldn't be anything to stop them from shipping some other *nix with Gnome if they felt like it.
Actually, most of the changes in the RH7 file system were made to bring it closer to the FHS2.1 standard. There is, in fact, an agreed upon standard for where things are supposed to go, and AFAIK RedHat is as close to conformant as anybody. They even added a bunch of symlinks to their rc.d directory structure so that programs from other distributions would find inits where they thought they should be.
I think that a significant role of the privacy officer will be to research and decide on privacy policies that make the most sense from a business standpoint. As an example, one obvious thing to look at is whether guaranteeing customers' privacy would increase sales enough to make up for potential income from selling that information. If their eventual privacy decision is made based on real research (e.g. finding that selling email addresses to spammers hurts sales) it's going to be much more convincing to other corporate officers than vague ideas about breaching that privacy being morally wrong. The fact that the CPO is also a lawyer suggests that she may also be able to back that up with legal arguments about potential lawsuits caused by breaking a promise to keep certain information private.
I also suspect that IBM's history works in favor of privacy. IBM has always concentrated on selling to businesses, rather than to consumers, and those businesses are both more protective of their corporate information and more able to make a stink if it's not kept private than typical consumers. Selling private corporate information is likely to result in losing a profitable client and quite possibly a lawsuit to boot, so a business oriented company is going to want strong protections in place. That attitude is going to impact the whole corporate culture and carry over to their consumer branch.
Slashdot Mirror
Come on, any fan can tell you that Harry doesn't have a Nimbus 2001, Draco Malfoy and the rest of the Slytherins have them. Harry once had a Nimbus 2000, but now he has a Firebolt. If you're going to try to make a topical reference to the books, at least try to get the facts straight.
The really remarkable thing about the Silmarillion, IMO, is that it was apparently never really finished. Tolkein continued to revise and completely rewrite sections well after the Hobbit and the Lord of the Rings were published. Part of the reason that the Silmarillion is of such uneven tone and polish throughout is that some of the stories hadn't been revised since the 1930's, while others had been rewritten shortly before his death. The Silmarillion as published was really a compilation of stories in different stages of revision, and even from slightly different stages of his conception of the underlying story, edited just enough to ensure coherence, as they existed at the time of Tolkein's death.
This stuff is revealed (in inordinate detail) in the whole History of Middle Earth series, which is basically his son's editing and deciphering of his papers. If you're a hardcore Tolkein fan, or just somebody who's very interested in the creative process, it's fascinating to see how the stories developed over time. What is particularly interesting is the way in which elements that were minor and trivial in the earliest versions gradually became more important, and vice versa. It makes you wonder if such a thing could even be possible today, given the way that people tend to overwrite their old word processing files instead of leaving paper copies of their work at different stages around to be looked over by later scholars.
But a number of the inventions (just about all, in fact) mentioned in the article were not completely eliminated, just largely displaced by alternatives. I think that's the general way of things, in fact. Old technology is rarely eliminated at a stroke. Instead, it's bumped into a niche market where it's particular advantages are significat enough to keep it alive. Over the very long haul that niche may dry up completely, but it rarely happens overnight.
In theory you're probably right. Since the GPL grants specific exemptions to traditional copyright restrictions, hackers will no longer need the rights granted by the GPL to hack the code after the copyright expires. In practice, though, that's not likely to make much difference. For one thing, copyright has been continually extended to the point that there's serious reason to think that it will never actually expire. More practically, though, modifications to the code are under copyright from the time that they're written, not the time that the original code was written. That means that when the copyright expires, you only get the non-GPL right to the oldest, cruftiest parts of the code. The new stuff that's likely to make any difference to the computer on your desk rather than one in a museum will still be under copyright.
The real problem is that people want to advertize, and they're going to keep trying to think up ways of doing so to keep one step ahead of the blockers. The real danger is that the ultimate in advertizing is not far away: product placement. Of course with media conglomeration this is already at least somewhat underway, but it's only going to get worse. How long is it going to be until web sites start incorporating ads for their commercial partners into the main content of their pages? I guess that search engines are already well down that road, from Google's "sponsored link" to goto.com's outright selling of placement to many other search engines' more insidious under-the-counter acceptance of pay for better search results. How long until other web sites start similar practices?
But openness is a necessary precondition for the freedom to continue hacking. If I want to excercise my freedom to hack your code, I must first have that code; you can't hack what you can't see. A freedom with no opportunity to excercise it is no freedom at all. Therefore preserving openness is essential to preserving freedom.
This is much harder than you think, and probably impossible in the long run. Why? Because some of the licenses that your code might wind up in (like the BSD license) are specifically designed to allow incorporation of their source into non-free projects. Once your code was put into a BSD licensed project, it could then be transfered into a proprietary one. To protect your intent, they'd have to relicense the part of their project that incorporated your code under a different style of license, which specifically undermines your goal of letting your code play friendly with everyone else. It's just not possible to have code that's useable by any free software project but not by a proprietary one.
You're right that the GPL doesn't make code free. Code is made free when its author(s) decide to make it free. What the GPL does is to keep code free. Once code is GPLed, it can't be taken back and put into a completely unfree box. Given the number of times that free software has been made non-free in the past, though, I think that GPL advocates have a good argument that strong protection is needed.
Actually, the restrictions are not there to hamper capitalists. The FSF openly advocates selling software for as much money as people will pay you for it. The restrictions are there to prevent people from distributing modified code without making the modifications available. Red Hat, Mandrake, Caldera, etc. would vigorously argue with your assertion that the GPL prevents capitalists from making money from software!
By ensuring that changes remain public. Code that is put in the public domain or BSD licenced can be absorbed into non-free projects without making any changes publically available, and frequently has been. The GPL essentially says that the right to see and further modify code is more important than the right to keep modifications private. (Note, though, that the GPL does allow private modifications so long as they are private; it only prevents distribution of modified binaries without the modified sources.) IMHO, that's a reasonable assertion, especially considering the way that those who aren't interested in sharing have taken advantage of those who are in the past. This is a legitimate point of contention, but I think that experience is on the side of the GPL.
It appears that he stuck in a hyperlink with a very long, all lower case HREF to reduce the percent of the comment that was in caps to below the filter threshold. He avoided having the link actually work by giving it a zero length field to link to, so your browser doesn't display it. It appears that the lameness filter needs some work.
I guess that my personal opinion is that Google is an excellent example of why patents exist, and that a comparison with the much blasted on-click shopping patent reveals why one is deserving and the other is not. The underlying logic behind patents is to encourage an inventor who might otherwise keep his invention secret to divulge his idea instead. The patent does this by giving him what he would try to achieve by secrecy- the exlusive right to his invention- in exchange for making the idea available to all after a certain time.
The key is that this implies that to be patent worthy an idea must be not only hard enough to think of that it's worth giving someone exclusive rights to in exchange for revealing it and also not something that is inherently revealed by the process of trying to implement it. The first criterion is the basis for obviousness and prior art restrictions. If something is so easy to invent that it's likely to be reinvented regularly, or even worse so obvious that somebody else has already published it, it's clearly not worth giving somebody exclusive rights to it in exchange for divulging it. The second criterions, which is the real objection to business model and similar patents, is that if you have to reveal your idea to use it, there's no point in granting a patent because there's no need for leverage to force the inventor to divulge it.
I think that Page Rank meets both of my criteria, and one-click doesn't meet either. Page Rank is not an obvious development of standard citation analysis, since it incorporates aspects of citation that are a product of the unique nature of the web. I think also that the fact that nobody had come up with the idea yet despite the fact that people had been thinking about web searching for quite some time is further, indirect evidence that the idea was non-obvious. One-click, though, was obvious enough that other sites were incorporating many of the same ideas (particularly the use of persistent cookies to prevent shoppers from having to re-enter their data every time) into their shopping at about the same time that Amazon did. The revelation side is even more clear. Google could have kept the guts of Page Rank hidden very easily; we only know as much about how it works as we do because they've told people about it. If they had decided just to say that they had a unique, effective new way of indexing sites, the chances are that you couldn't figure out how they were doing it just by looking at their search results. One click shopping, though, reveals itself by its very existence; in order to implement it you have to give people enough of an idea about what you're doing that it would be trivial for anyone to reimplement it.
I assume that the reason that traditional citation analysis doesn't do this is because traditional citations only go one way. You can only cite something that was published at the time that you wrote your work, so you can't cite anything that cites you. This greatly complicates the process, since inherently the recent papers that should have the best grasp of which older papers were most influential can't have many citations yet and hence won't themselves be considered very important. Because publication on the web is dynamic, though, this is not a problem, and bidirectional citation is possible.
The other cool thing about Page Rank is that it isn't just a recursive analysis. Instead it uses a mathematical trick that is roughly equivalent to doing a recursive analysis except that it turns up more data because it can find substantially non-overlapping networks of sites that would otherwise be missed. Suppose, for instance, that you have a topic on which opinion is very polarized (abortion was the one that I've heard this first turned up in). You would tend to have strongly opinionated sites that link almost exclusively to other sites that share the same opinion. In a straight recursive analysis, one side of the issue will have fewer total links and hence less credibility, which then in turn deprives its linked-to sites of credibility, etc. The result is that one side is effectively silenced by the simple analysis. The trick used by Page Rank not only saves computer time, it also turns up the second cluster of sites and can bump them up the list.
But Page Rank isn't an algorithm. It's the application of an existing algorithm to a novel area of interest. The beauty of Google's system is that it's applying quite standard concepts from Linear Algebra to a novel application. The proof that the concepts are novel is that despite the fact that people have been doing citation analysis for some time, apparently nobody had come up with the idea of recursively analyzing the reliability of citations by taking into account the quality of the citing source. Nor, for that matter, had they gone the additional step and proven that this is mathematically equivalent to finding the eigenvectors of a citation matrix, which is the neat mathematical trick that makes it computationally reasonable. This is, IMO, exactly the type of powerful, novel development that deserves to be given patent protection.
Some of us think that approaches to a problem that are both non-obvious and extremely powerful (which, IMO at least, is true of Page Rank) are exactly the reason that patents were invented. This is not a patent that is:
Patents exist precisely to protect inventions that don't fall into one or more of the above categories. It's not so much patents that most slashdotters are angry about, it's patent abuses, and Google's patent clearly isn't abusing the system.
This was exactly my feeling when I looked at both of them, so I tend to disagree with the article when they say that the Go Type gives you better value. IMO it's totally useless as a keyboard, so you're wasting your money when buying it. If it doesn't work right, it's not a good deal no matter how cheap it is.
One thing that bothered me deeply about the article is that the author made very little distinction between hardware and software quality problems. I realize that this probably reflects the majority, non-tech view of computers (the computer is a unit that succeeds or fails as a unit, not as hardware or software) but it made the article less comprehensible. Most of the specific problems that he talked about sound as though they were hardware problems, but the experts he consulted were talking about curing software reliability. This is probably reasonable, since reliable, high quality hardware is available, and the companies that produce junk often go out of business when people stop buying their crap. Of course high quality, reliable software is available too, but most desktop PCs don't use it. I'm inclined to agree with the closing statement of the article: we won't get high quality software until companies suffer financially from putting out crappy software.
The problem is that much of the web seems to be designed by graphic artists coming from a print background rather than computer people. As such, they're deeply attached to the idea of fixing their work precisely, as paper printing both rewards and demands. They simply haven't adapted to the new format, and I get the impression that this actually gives something of an advantage to the inexperienced over the experienced. People who have never done graphics design before are more able to accept the limitations (and strengths) of the web than people transfering from another field. That's not to say that some traditional graphic artists haven't been able to make the transition effectively, but I think that it's a handicap.
So was the guy who came up with the AOL CD. He apparently decided that American coffee tables were under critical threat of being destroyed by condensation stains. His brilliant response was to provide every home in the country with an endless supply of free coasters.
Obviously, this means that it's time for the First Annual Obfuscated Javascript contest ;-)
All that shows is that people have unreasonable expectations for HTML. It was never intended for hyper-precise description of the layout of the material. If you use it the way that it was really indended to be used, i.e. to make an acceptable layout from content descriptive markup, it is quite capable of doing a good job. It is even quite possible to make nice looking, artistically designed pages that way without using incredibly elaborate formatting. It's when people fight against the intent of the format and try to use it as a page description language that they start having to bend, fold, spindle, staple, and mutilate it. Hint: if your page expects the browser window to be a particular size, so that it won't fit if it's 10% too narrow or has an unattractive white gutter if it's 10% too wide, you're doing something wrong.
But honestly, the versions of those programs under Gnome (and also KDE, AFAIK) are actually much superior to their Windows equivalents. The graphics are substantially better (I was amazed at how lousy MS Solitare looked when I booted back into Windows) and AisleRiot has a much larger selection of solitare games to choose from. You also get such classics as Mahjong, XBill, Robots, etc. The built in games are a real improvement over Windows.
How about looking at tools to allow insecure software to be used without compromising the integrity of the system? The effort to produce a secure OS base is largely wasted if adding new programs trashes the security. It's nice to have an OS like OpenBSD as the basis for high security bits like firewalls, but it's never going to get beyond a niche market if the security evaporates the moment that desktop/workstation applications are installed. What kinds of things is OpenBSD doing to help make it easier for developers to make secure applications?
Not necessarily. It can run on other Unixlike OSes, too, which is part of the reason that major vendors like Sun and HP are backing the Gnome Foundation. You can certainly run it with Free/Open/NetBSD. FWIW, I think that Dell is actually shipping RedHat, but there certainly wouldn't be anything to stop them from shipping some other *nix with Gnome if they felt like it.
Actually, most of the changes in the RH7 file system were made to bring it closer to the FHS2.1 standard. There is, in fact, an agreed upon standard for where things are supposed to go, and AFAIK RedHat is as close to conformant as anybody. They even added a bunch of symlinks to their rc.d directory structure so that programs from other distributions would find inits where they thought they should be.
I think that a significant role of the privacy officer will be to research and decide on privacy policies that make the most sense from a business standpoint. As an example, one obvious thing to look at is whether guaranteeing customers' privacy would increase sales enough to make up for potential income from selling that information. If their eventual privacy decision is made based on real research (e.g. finding that selling email addresses to spammers hurts sales) it's going to be much more convincing to other corporate officers than vague ideas about breaching that privacy being morally wrong. The fact that the CPO is also a lawyer suggests that she may also be able to back that up with legal arguments about potential lawsuits caused by breaking a promise to keep certain information private.
I also suspect that IBM's history works in favor of privacy. IBM has always concentrated on selling to businesses, rather than to consumers, and those businesses are both more protective of their corporate information and more able to make a stink if it's not kept private than typical consumers. Selling private corporate information is likely to result in losing a profitable client and quite possibly a lawsuit to boot, so a business oriented company is going to want strong protections in place. That attitude is going to impact the whole corporate culture and carry over to their consumer branch.