There are three '*BSD' distributions, aside from the venerable BSDi:
FreeBSD, the codebase described here, is the middle-of-the-road BSD distribution. Its users make offerings at the Temple of Berkeley.
OpenBSD is the "ultrasecure" version. It has an active, ongoing audit for security flaws and so forth. Even Linux advocates in the know will recommend OpenBSD when the main objective is security.
NetBSD is the "widely ported" version; its platform count exceeds Linux's at last check (although a friend of mine working at a Maryland company curses out NetBSD/Sparc on a daily basis).
Another important thing to think about: A format doesn't suck because it has less support than other formats. If that were true then everyone should go back to using ASCII for documents because, ahem, EVERY platform has support for rendering ASCII perfectly.
PNG is a superior image format in many ways; instead of shouting nonsense like "PNG SUCKS", how about encouraging more people to use modern browsers that are PNG-aware. You'll also be looking at a size savings, since PNGs are more compact, on average, than GIFs.
My personal website has been using PNGs for awhile. I've got a link to the GNU's "no gifs" page, and a link to a page listing browser compatibility with PNG images.
To date, one person has had problems viewing my images, and I've deliberately talked to a large cross-sample of browser-using friends, including folks from AOL. PNG is more widespread than you might think, probably because many of the techies producing browser products are also at least somewhat familiar with the Unisys patent business, and either oppose it on philosophical grounds or are fearful of the legal problems that could result. If I were writing a browser, I'd build in a legal backdoor. PNG seems to have filled that niche nicely.
And oh yeah. Most of my PNGs are 20%-60% smaller than the same-size, same-quality GIFs they replaced.
Re:Will Mozilla ever gain outside contributors?
on
Mozilla M9 Released
·
· Score: 1
The models are up to the owner of each module to supply/document. This is the reason mozilla has modules and module owners - the module owner can present a view of his/her particular bailiwick in whatever manner seems clearest. Brute-force reading of the code would be counterproductive, as it goes against the workflow that the Mozilla team has tried so hard to establish.
The interview is a no-win situation. On the one hand, it would have been great to hear the view of a couple NT developers on how MS corporate culture opposes their stated goals of portability, etc. (as was hinted at). But such an interview wouldn't be permitted by the culture, which means it would have been a third-party interviewer, and the interview would have been posted somewhere other than www.microsoft.com. But it's unlikely that such a third-party interview would have been permitted............
One of the stronger arguments in favor of this in the particular case of Red Hat is that, barring a big buyout by someone with "real money", a solid block (if not a majority) of the shareholders ARE PEOPLE IN THE COMMUNITY. In other words, Red Hat is telling the Linux community at large, especially those who have been instrumental in development, "We're going to do whatever it takes to keep you happy". Well, that's good. It just means people are making a buck off what they wanted to happen in the first place.
Microsoft's attitude towards standards is well documented, and a stellar example is the whole CSS patent thing. They're willing to "work with" the W3C and company on CSS as a standard. Open the patent up and actually treat it as an open standard? "Oh please" mumble Microsoft execs.
Meanwhile the IETF (or whichever committee it is exactly that's developing the open instant-message protocol) is left out in the cold. If anything, the IETF is who we should be worried about embracing this, not the open-source community. Once they've folded, Microsoft's messaging hegemony will be complete and they can accuse AOL of not following the open standard, and have some real ammo to support the assertion.
Most ISPs DO keep extensive records of their mail-server activity. Sendmail has historically been the #1 daemon that crackers will set their sights on (for a combination of reasons); add to that the numerous times a customer will call in and say "someone sent me a large file and I can't get my mail" (or some other mail-related problem), and you get a climate where ISPs are ALREADY happy to keep logfiles of their chosen MTA's activities.
As for taxing the sender, even without hacks to sendmail, it's trivial to monitor port 25 -- which affects "real" SMTP sessions and spoofed SMTP (sorry, telnetting in will not win you anything except the Booby Prize for not having written a script in perl to do it for you). While individual Linux (etc.) users may be able to run sendmail on an alternate port, they WILL need the cooperation of an existing MTA (listening on the well-known port), or they will need to stop corresponding with any of their friends who don't also have the modified daemon.
That said, it'd be horrifically difficult to offer solid evidence of a mail message being sent without some form of authorization. SMTP is essentially a one-way protocol; there's nothing to stop it from being spoofed.
Enough of the technical: I won't even start in on the social problems this could cause.
Ever since Sun got a coherent vision statement ("the network is the computer" may have been interesting, but did any PHBs really get it?), they and Microsoft and everyone else in this industry have all been racing towards the same set of targets; at least now some of them are admitting it.
As a Christian, I fully support teaching evolution in schools. I don't personally believe that "macro-evolution" (the life-from-lifelessness, amoeba-to-man) is a viable explanation for human origins, but I'm also quite clued in on the fact that as yet, there is ABSOLUTELY NO substantive proof for any single theory of the origin of life.
Having said that, I restate: I support the teaching of evolution in schools, provided that it is taught as what it really is: a theory, no more, no less. I support the teaching of creationism with the same caveats.
The talk.origins FAQ (posted earlier) is an excellent take on this long-debated topic, and those who feel the larger ramifications of the Kansas decision merit further rehashing are encouraged to take their opinions to the talk.origins newsgroup.
The problem is that while individuals may not choose to use software with such (comparatively) draconian licensing terms, every day of our lives we are affected by people who do use that software -- the word processor at the office, the application development environment, etc. etc.
I don't think any technical solution would NEED to be implemented for remote deactivation. That opens up a can of security worms that I'm sure most big software houses are sick to death of by now. In today's lawsuit-charged climate, if a ten ton boulder rolled down the hill and a lawyer issued it a cease-and-desist letter, that boulder would stop -- legal pressures are quite sufficient to implement the remote-deactivation clause.
I sincerely doubt that the level of remote-shutdown "license recrimination" that many people are predicting will come to pass. It's much more likely that the no-reverse-engineering clause and the slippery-slope principle will get a lot of people. After all, how much time before the Samba team (housed in Australia) will find themselves on the business end of a lawsuit from the Australian government after something like UCITA passes down under?
Many of the advantages that made X so powerful in "the day" may no longer apply. X has an extensive security mechanism that exists more or less to provide authentication/authorization in an insecure, networked environment (such as a bunch of X terminals connected to one app server). While this environment does still exist (primarily in university settings), "most" users of X only have a single display to work with and won't need the security considerations X offers.
X tends to suffer from what I call "Win32APItis", a huge assortment of functions of dubious utility (it would be "GTKitis" if it weren't documented). Other projects, such as Berlin, seem to have learned this lesson -- it's possible to make an entire windowing system with the simplicity of a "regular" toolkit. Any step in the direction of a less stressful API is therefore a step towards unification of look-and-feel, since developers won't be compelled to write entirely new toolkits to scratch the itch of hating Xt/Xlib:)
While some of the various extensions that X is now sporting (patches for TrueType support, GLX, and printing) are useful accessories, ultimately they reveal X's age. A lot of the features a modern GUI designer would like simply aren't that easy (or sometimes possible) to implement.
All this isn't to say that we just need to chuck X in the garbage can and start supporting Berlin/GGI/whatever; I would prefer to see a thin (hah) Xlib wrapper on top of such projects, in parallel to ports of GDK or Qt or whatever people like. X has some very strong advantages, which should not be overlooked.
Many of the advantages that made X so powerful in "the day" may no longer apply. X has an extensive security mechanism that exists more or less to provide authentication/authorization in an insecure, networked environment (such as a bunch of X terminals connected to one app server). While this environment does still exist (primarily in university settings), "most" users of X only have a single display to work with and won't need the security considerations X offers.
X tends to suffer from what I call "Win32APItis", a huge assortment of functions of dubious utility (it would be "GTKitis" if it weren't documented). Other projects, such as Berlin, seem to have learned this lesson -- it's possible to make an entire windowing system with the simplicity of a "regular" toolkit. Any step in the direction of a less stressful API is therefore a step towards unification of look-and-feel, since developers won't be compelled to write entirely new toolkits to scratch the itch of hating Xt/Xlib:)
While some of the various extensions that X is now sporting (patches for TrueType support, GLX, and printing) are useful accessories, ultimately they reveal X's age. A lot of the features a modern GUI designer would like simply aren't that easy (or sometimes possible) to implement.
All this isn't to say that we just need to chuck X in the garbage can and start supporting Berlin/GGI/whatever; I would prefer to see a thin (hah) Xlib wrapper on top of such projects, in parallel to ports of GDK or Qt or whatever people like. X has some very strong advantages, which should not be overlooked.
Blah, think I might have posted a blank message by mistake. Anyway:
There are sufficient reasons to feel somewhat paranoid about this. "Why would governments collect this information and sell it to corporations, why not just raise taxes?" Because increased taxation looks bad, because lowering taxes has been a traditional campaign promise, and because it's harder to track monies coming from the sale of personal data than it is to track monies from taxation.
But why else would the government want to collect this sort of information? A barrel of political reasons stick out. The slippery slope from "tracking criminals" to "tracking undesirables" has existed since the advent of computer technology. Another possibility is the same theory that drives insurance rates -- if the Jews, or the people making $25,000 to $45,000 a year, or the fathers of 2.5 or more kids are more likely to commit crimes, statistically, then monitoring those groups more closely would lead to prevention. Right?
While it's easy to say "They are out to get us", I hope people will settle for "This makes it easier for them to get us" -- prepare for the worst, rather than expect it.
If you want to talk about shaping the course of history, I would start talking about Kaiser Wilhelm and any of the World Wars presidents. Or how about Richard Nixon? He threw probably the biggest monkey wrench into American politics since the Revolution, and Kissinger had a similar (if more positive) effect internationally. I am not enough of a historian to name many more people outside the U.S. but I'm confident there are many for which this argument could be made.
Then again, I would say that overall, the 20th century has converged on the concept of "mine mine mine". Today we have lawsuits flying around daily asserting "rights" that are past trivial and into ludicrous, companies who are constantly struggling to hide "intellectual property" behind their own iron curtain, and so forth.
Then along comes Torvalds, who says "ours ours ours", and who really has caused some dramatic changes in the way people think. Of course RMS has been saying this for awhile, and even comes up with the GNU project, and the tools to do it. Torvalds is the person who brought it out into the open, and for that I'll vote for him. I can't say he deserves the #1 slot completely but I expect him to make the list:)
There are better application server packages out there, at whatever level you need (from HTML::Mason to ColdFusion to EJB), many of them free or more affordable. I'm ok with them backing off on this, if the reasons are valid (lack of demand, I can believe).
Is there a HOWTO or other "comprehensive" document on how to protect your personal privacy, both online and in the real world? If not, I could probably start one. Submissions welcome:)
Well, I could see worse alternatives: "Slashdot Acquired by Politically Correct Union - Anonymous Cowards renamed to 'Differently Identified Courage-Challenged Individuals'"
Good stuff, here's wishing Slashdot and Andover a happy and successful future:)
However, it's been my experience that people have perfect MEMORIES, just faulty RECALLS (for a number of reasons). No PDAesque device will ever be able to match for relevance. On the other hand, I remember everything that's important to me.
I'd rather see an ad for a device that subliminally influences people to use their recall to best effect.
Calvin Coolidge Movie Review: "the spirit is willing but the facts are weak"
I regard this movie as insight into the mental states of several of the parties involved. I don't regard it as historically accurate, but that's ok - I enjoyed "Hackers" (yeah yeah, flame on) for the same reason but I would giggle if anyone told me they'd seen accurate technical details there.
The one thing that came across so well was the overwhelming sense of "What was going through this guy's HEAD?"
Bob Metcalfe invented ethernet (an old technology) Linux is based on UNIX principles (an old technology)
Ethernet is pretty much an open standard -- you can find the specs for it and build your own network. Linux is pretty much an open standard -- you can find the specs for it and build your own software.
Ethernet products are making money. Linux is making money.
Ethernet is used the world over. You say "ARCnet" and most IT people will start giggling. You say "Token Ring" in a crowded hall of IT people and you might see a couple hands.
This is not exactly correct -- open source projects have deadlines, just not EXTERNAL deadlines. The developers have laid out their roadmap (which you can see on mozilla.org) and have set certain goals for themselves.
As to whether ditching proprietary DOM is a good move, I'm all in favor of it -- sites that are "optimized for Blah Woof Browser" need to be fixed, and sites that use some existing "toolkit" for doing DHTML etc. will be fixed when the toolkit becomes compatible with the published standard (the best one I'm aware of is the one at www.dansteinman.com). After all, you are going to have broken compatibility with SOME browsers somewhere; may as well break it in a good cause.
Re:When will they learn?
on
Linux 2.2.10
·
· Score: 1
And how would you propose that people test a given kernel against every possible DoS, error, and bug before it's released? If it were possible to synthesize all possible attacks in a test environment, we'd be using a flawless version of Linux 1.0 (as 0.99.x was tested using our mythical Uber-Workbench).
Red Hat has traditionally been a late starter -- everybody who yells "5.2 is rock solid" may or may not be keeping in mind that 5.0 and 5.1 preceded it, and were not exactly paragons of functional software:)
I'm holding off until 6.1 hits the shelves. But so far, Red Hat has only rarely disappointed me.
Slashdot Mirror
There are three '*BSD' distributions, aside from the venerable BSDi:
FreeBSD, the codebase described here, is the middle-of-the-road BSD distribution. Its users make offerings at the Temple of Berkeley.
OpenBSD is the "ultrasecure" version. It has an active, ongoing audit for security flaws and so forth. Even Linux advocates in the know will recommend OpenBSD when the main objective is security.
NetBSD is the "widely ported" version; its platform count exceeds Linux's at last check (although a friend of mine working at a Maryland company curses out NetBSD/Sparc on a daily basis).
Another important thing to think about:
A format doesn't suck because it has less support than other formats. If that were true then everyone should go back to using ASCII for documents because, ahem, EVERY platform has support for rendering ASCII perfectly.
PNG is a superior image format in many ways; instead of shouting nonsense like "PNG SUCKS", how about encouraging more people to use modern browsers that are PNG-aware. You'll also be looking at a size savings, since PNGs are more compact, on average, than GIFs.
My personal website has been using PNGs for awhile. I've got a link to the GNU's "no gifs" page, and a link to a page listing browser compatibility with PNG images.
To date, one person has had problems viewing my images, and I've deliberately talked to a large cross-sample of browser-using friends, including folks from AOL. PNG is more widespread than you might think, probably because many of the techies producing browser products are also at least somewhat familiar with the Unisys patent business, and either oppose it on philosophical grounds or are fearful of the legal problems that could result. If I were writing a browser, I'd build in a legal backdoor. PNG seems to have filled that niche nicely.
And oh yeah. Most of my PNGs are 20%-60% smaller than the same-size, same-quality GIFs they replaced.
The models are up to the owner of each module to supply/document. This is the reason mozilla has modules and module owners - the module owner can present a view of his/her particular bailiwick in whatever manner seems clearest. Brute-force reading of the code would be counterproductive, as it goes against the workflow that the Mozilla team has tried so hard to establish.
/. is an implementation of a competing technology with JUST SLIGHTLY FEWER SERVERS and bandwidth than microsoft.com
Doesn't matter how good Linux/Apache/whatever is if there's not enough bandwidth and CPU to serve everyone who's visiting a site.
The interview is a no-win situation. On the one hand, it would have been great to hear the view of a couple NT developers on how MS corporate culture opposes their stated goals of portability, etc. (as was hinted at). But such an interview wouldn't be permitted by the culture, which means it would have been a third-party interviewer, and the interview would have been posted somewhere other than www.microsoft.com. But it's unlikely that such a third-party interview would have been permitted............
So it's a shame.
One of the stronger arguments in favor of this in the particular case of Red Hat is that, barring a big buyout by someone with "real money", a solid block (if not a majority) of the shareholders ARE PEOPLE IN THE COMMUNITY. In other words, Red Hat is telling the Linux community at large, especially those who have been instrumental in development, "We're going to do whatever it takes to keep you happy". Well, that's good. It just means people are making a buck off what they wanted to happen in the first place.
Microsoft's attitude towards standards is well documented, and a stellar example is the whole CSS patent thing. They're willing to "work with" the W3C and company on CSS as a standard. Open the patent up and actually treat it as an open standard? "Oh please" mumble Microsoft execs.
Meanwhile the IETF (or whichever committee it is exactly that's developing the open instant-message protocol) is left out in the cold. If anything, the IETF is who we should be worried about embracing this, not the open-source community. Once they've folded, Microsoft's messaging hegemony will be complete and they can accuse AOL of not following the open standard, and have some real ammo to support the assertion.
Most ISPs DO keep extensive records of their mail-server activity. Sendmail has historically been the #1 daemon that crackers will set their sights on (for a combination of reasons); add to that the numerous times a customer will call in and say "someone sent me a large file and I can't get my mail" (or some other mail-related problem), and you get a climate where ISPs are ALREADY happy to keep logfiles of their chosen MTA's activities.
As for taxing the sender, even without hacks to sendmail, it's trivial to monitor port 25 -- which affects "real" SMTP sessions and spoofed SMTP (sorry, telnetting in will not win you anything except the Booby Prize for not having written a script in perl to do it for you). While individual Linux (etc.) users may be able to run sendmail on an alternate port, they WILL need the cooperation of an existing MTA (listening on the well-known port), or they will need to stop corresponding with any of their friends who don't also have the modified daemon.
That said, it'd be horrifically difficult to offer solid evidence of a mail message being sent without some form of authorization. SMTP is essentially a one-way protocol; there's nothing to stop it from being spoofed.
Enough of the technical: I won't even start in on the social problems this could cause.
Ever since Sun got a coherent vision statement ("the network is the computer" may have been interesting, but did any PHBs really get it?), they and Microsoft and everyone else in this industry have all been racing towards the same set of targets; at least now some of them are admitting it.
As a Christian, I fully support teaching evolution in schools. I don't personally believe that "macro-evolution" (the life-from-lifelessness, amoeba-to-man) is a viable explanation for human origins, but I'm also quite clued in on the fact that as yet, there is ABSOLUTELY NO substantive proof for any single theory of the origin of life.
Having said that, I restate: I support the teaching of evolution in schools, provided that it is taught as what it really is: a theory, no more, no less. I support the teaching of creationism with the same caveats.
The talk.origins FAQ (posted earlier) is an excellent take on this long-debated topic, and those who feel the larger ramifications of the Kansas decision merit further rehashing are encouraged to take their opinions to the talk.origins newsgroup.
The problem is that while individuals may not choose to use software with such (comparatively) draconian licensing terms, every day of our lives we are affected by people who do use that software -- the word processor at the office, the application development environment, etc. etc.
I don't think any technical solution would NEED to be implemented for remote deactivation. That opens up a can of security worms that I'm sure most big software houses are sick to death of by now. In today's lawsuit-charged climate, if a ten ton boulder rolled down the hill and a lawyer issued it a cease-and-desist letter, that boulder would stop -- legal pressures are quite sufficient to implement the remote-deactivation clause.
I sincerely doubt that the level of remote-shutdown "license recrimination" that many people are predicting will come to pass. It's much more likely that the no-reverse-engineering clause and the slippery-slope principle will get a lot of people. After all, how much time before the Samba team (housed in Australia) will find themselves on the business end of a lawsuit from the Australian government after something like UCITA passes down under?
Many of the advantages that made X so powerful in "the day" may no longer apply. X has an extensive security mechanism that exists more or less to provide authentication/authorization in an insecure, networked environment (such as a bunch of X terminals connected to one app server). While this environment does still exist (primarily in university settings), "most" users of X only have a single display to work with and won't need the security considerations X offers.
:)
X tends to suffer from what I call "Win32APItis", a huge assortment of functions of dubious utility (it would be "GTKitis" if it weren't documented). Other projects, such as Berlin, seem to have learned this lesson -- it's possible to make an entire windowing system with the simplicity of a "regular" toolkit. Any step in the direction of a less stressful API is therefore a step towards unification of look-and-feel, since developers won't be compelled to write entirely new toolkits to scratch the itch of hating Xt/Xlib
While some of the various extensions that X is now sporting (patches for TrueType support, GLX, and printing) are useful accessories, ultimately they reveal X's age. A lot of the features a modern GUI designer would like simply aren't that easy (or sometimes possible) to implement.
All this isn't to say that we just need to chuck X in the garbage can and start supporting Berlin/GGI/whatever; I would prefer to see a thin (hah) Xlib wrapper on top of such projects, in parallel to ports of GDK or Qt or whatever people like. X has some very strong advantages, which should not be overlooked.
Many of the advantages that made X so powerful in "the day" may no longer apply. X has an extensive security mechanism that exists more or less to provide authentication/authorization in an insecure, networked environment (such as a bunch of X terminals connected to one app server). While this environment does still exist (primarily in university settings), "most" users of X only have a single display to work with and won't need the security considerations X offers.
:)
X tends to suffer from what I call "Win32APItis", a huge assortment of functions of dubious utility (it would be "GTKitis" if it weren't documented). Other projects, such as Berlin, seem to have learned this lesson -- it's possible to make an entire windowing system with the simplicity of a "regular" toolkit. Any step in the direction of a less stressful API is therefore a step towards unification of look-and-feel, since developers won't be compelled to write entirely new toolkits to scratch the itch of hating Xt/Xlib
While some of the various extensions that X is now sporting (patches for TrueType support, GLX, and printing) are useful accessories, ultimately they reveal X's age. A lot of the features a modern GUI designer would like simply aren't that easy (or sometimes possible) to implement.
All this isn't to say that we just need to chuck X in the garbage can and start supporting Berlin/GGI/whatever; I would prefer to see a thin (hah) Xlib wrapper on top of such projects, in parallel to ports of GDK or Qt or whatever people like. X has some very strong advantages, which should not be overlooked.
Blah, think I might have posted a blank message by mistake. Anyway:
There are sufficient reasons to feel somewhat paranoid about this. "Why would governments collect this information and sell it to corporations, why not just raise taxes?" Because increased taxation looks bad, because lowering taxes has been a traditional campaign promise, and because it's harder to track monies coming from the sale of personal data than it is to track monies from taxation.
But why else would the government want to collect this sort of information? A barrel of political reasons stick out. The slippery slope from "tracking criminals" to "tracking undesirables" has existed since the advent of computer technology. Another possibility is the same theory that drives insurance rates -- if the Jews, or the people making $25,000 to $45,000 a year, or the fathers of 2.5 or more kids are more likely to commit crimes, statistically, then monitoring those groups more closely would lead to prevention. Right?
While it's easy to say "They are out to get us", I hope people will settle for "This makes it easier for them to get us" -- prepare for the worst, rather than expect it.
If you want to talk about shaping the course of history, I would start talking about Kaiser Wilhelm and any of the World Wars presidents. Or how about Richard Nixon? He threw probably the biggest monkey wrench into American politics since the Revolution, and Kissinger had a similar (if more positive) effect internationally. I am not enough of a historian to name many more people outside the U.S. but I'm confident there are many for which this argument could be made.
:)
Then again, I would say that overall, the 20th century has converged on the concept of "mine mine mine". Today we have lawsuits flying around daily asserting "rights" that are past trivial and into ludicrous, companies who are constantly struggling to hide "intellectual property" behind their own iron curtain, and so forth.
Then along comes Torvalds, who says "ours ours ours", and who really has caused some dramatic changes in the way people think. Of course RMS has been saying this for awhile, and even comes up with the GNU project, and the tools to do it. Torvalds is the person who brought it out into the open, and for that I'll vote for him. I can't say he deserves the #1 slot completely but I expect him to make the list
There are better application server packages out there, at whatever level you need (from HTML::Mason to ColdFusion to EJB), many of them free or more affordable. I'm ok with them backing off on this, if the reasons are valid (lack of demand, I can believe).
Is there a HOWTO or other "comprehensive" document on how to protect your personal privacy, both online and in the real world? If not, I could probably start one. Submissions welcome :)
Well, I could see worse alternatives:
:)
"Slashdot Acquired by Politically Correct Union - Anonymous Cowards renamed to 'Differently Identified Courage-Challenged Individuals'"
Good stuff, here's wishing Slashdot and Andover a happy and successful future
*throws rice to the happy couple*
There's one problem with this thing.
Not every fact I encounter is relevant.
However, it's been my experience that people have perfect MEMORIES, just faulty RECALLS (for a number of reasons). No PDAesque device will ever be able to match for relevance. On the other hand, I remember everything that's important to me.
I'd rather see an ad for a device that subliminally influences people to use their recall to best effect.
Calvin Coolidge Movie Review: "the spirit is willing but the facts are weak"
I regard this movie as insight into the mental states of several of the parties involved. I don't regard it as historically accurate, but that's ok - I enjoyed "Hackers" (yeah yeah, flame on) for the same reason but I would giggle if anyone told me they'd seen accurate technical details there.
The one thing that came across so well was the overwhelming sense of "What was going through this guy's HEAD?"
Bob Metcalfe invented ethernet (an old technology)
Linux is based on UNIX principles (an old technology)
Ethernet is pretty much an open standard -- you can find the specs for it and build your own network.
Linux is pretty much an open standard -- you can find the specs for it and build your own software.
Ethernet products are making money.
Linux is making money.
Ethernet is used the world over. You say "ARCnet" and most IT people will start giggling. You say "Token Ring" in a crowded hall of IT people and you might see a couple hands.
Of course Linux will fail.
This is not exactly correct -- open source projects have deadlines, just not EXTERNAL deadlines. The developers have laid out their roadmap (which you can see on mozilla.org) and have set certain goals for themselves.
As to whether ditching proprietary DOM is a good move, I'm all in favor of it -- sites that are "optimized for Blah Woof Browser" need to be fixed, and sites that use some existing "toolkit" for doing DHTML etc. will be fixed when the toolkit becomes compatible with the published standard (the best one I'm aware of is the one at www.dansteinman.com). After all, you are going to have broken compatibility with SOME browsers somewhere; may as well break it in a good cause.
And how would you propose that people test a given kernel against every possible DoS, error, and bug before it's released? If it were possible to synthesize all possible attacks in a test environment, we'd be using a flawless version of Linux 1.0 (as 0.99.x was tested using our mythical Uber-Workbench).
Red Hat has traditionally been a late starter -- everybody who yells "5.2 is rock solid" may or may not be keeping in mind that 5.0 and 5.1 preceded it, and were not exactly paragons of functional software :)
I'm holding off until 6.1 hits the shelves. But so far, Red Hat has only rarely disappointed me.