The stupid autodetect thing came up for me too when I first clicked on it, but it gave me the option to ignore the autodetection and proceed anyway. After I did that it set some cookie telling it not to autodetect again, and now it goes straight to the video page when I click on a video link.
You might try deleting your CNN cookies and trying it again. What browser are you using and does it get to the "brushed metal" looking video page or does it never get past the autodetection thing?
You can even raise this issue against the entire Internet (disregarding the anonymity issue altogether): should we ban the Internet because people distribute child pornography through it? Are ISPs legally held responsible when their users distribute child porn without their knowledge? The answer is no, because ISPs are a common carrier and are not required to police their users unless the government (or others, like the DMCA notices) brings specific evidence of wrongdoing to them. How is Tor any different? The people who run nodes are only acting as a carrier channel for whatever passes through and do not have to police the traffic or be held responsible.
I can see governments getting pissed off if things like Tor get too powerful and popular, since it would undermine their ability to eavesdrop on people EASILY. However, governments and societies managed just fine before they could instantly and anonymously spy on their own citizens. They just used other means to find the information about who was breaking laws. Technology should not be an excuse for allowing a surveillance society just so that it's more convenient to catch "bad guys".
In my opinion, Windows is only superficially "easier" to administer than Unix systems. It appears to be easier at first, but it actually isn't because it becomes a major pain when problems crop up. What is a company that has only hired shoddy, unknowledgable Windows admins going to do when that happens? Bring in a consultant or two at astronomical cost? Sorta kills those cost savings from being able to hire unskilled admins if you have to do that a few times a year.
Governments that I've dealt with seem to be notorious for this kind of mismanagement. One federal agency that I know of just sinks millions upon millions of dollars into whatever new "solution" Microsoft puts out every couple of years, and then hires idiots who couldn't administer their way out of a paper bag to run it. I'm sure the rest do the same thing.
I think the only thing that Excel has to recommend it whatsoever as far as large-scale data analysis is exactly the thing you just mentioned: it's easy to learn how to use. It's not very powerful, but a trained monkey can be up and running on it in 5 minutes. That's probably been a good business model for MS.
In the long run, you're MUCH better off learning how to use some more powerful tool. Whatever time this costs you in the short term will be saved many times over in the end.
Oh, and Excel STILL is not the best thing to use even just for "interoperability" with other people who use it. CSV (comma separated value) files are what I use for that because they're a helluva lot more universal than Excel and they're easy to use to hook Excel or whatever up to Perl or some other real programming language.
Maybe Excel is good for financial stuff or something, I don't know. I haven't found a real use for it yet.
I agree, Excel is a joke compared to Matlab as far as scientifici functionality. There are some things that are annoying to me about Matlab, like the way its plotting interface works, but it's still way more powerful and flexible.
Almost all the real scientists I've met use Matlab or SPSS or something else more powerful than Excel if they need to do data analysis. Excel is fine if you're just doing averages and standard deviations and other basic stuff, but if you need to do more advanced numerical or statistical stuff you really start to appreciate Matlab's standard library. Maybe the scientists who use Excel use third-party addons for more mathematical functionality, or maintain their own library of functions for it?
You're a developer if you write code for a webpage. Especially if it's a professional/business site, or a page created by a CMS, etc. And in fact you could argue that the way it should be is that only developers should write actual HTML/XHTML code. People who aren't capable of writing it properly should be using some sort of content management system or WYSIWYG tool to write the code for them.
Leave the coding to people who know what they are doing and can do it correctly, and have users who can't use software that lets them publish web content without knowing how. This is how it should be and how it is going, anyway. Witness the popularity of blogging: this is possible because easy to use blog software lets people who aren't coders write their own webpages. Then if there are problems with the HTML then it's just a bug in the publishing software and not the users' problem to fix.
What's the problem? You should be fixing all the bugs like that before you release your page publically. And unlike regular HTML you'll actually know they're there in XHTML. Or is the idea that you shouldn't have to test your code beyond "looks good to me!"
There's a certain irony to the idea of banning flag-burning. The flag is supposed to represent an ideal of freedom, but you're not free to burn it? Like it or not, burning the flag is clearly political speech. It shows more respect for the flag and what it stands for to let everyone keep the freedom to burn it if they choose. If we take away the freedom to participate in any political speech we want to, then the flag becomes just a meaningless piece of cloth undeserving of protection anyway.
I think you're correct about that - or at least that's how the schools see their role these days. They're really more of a venue for people to push whatever political agenda they personally have than for any real education to take place.
This is the problem with public education in the first place. The government will seek to interfere if given control and the ability to do so. IMHO, the only reason we avoided that kind of crap for a long time in this country was because public education was largely decentralized - funded, run, and controlled by local and state government. But notice that these days the federal government is seeking to interfere more and more all the time? As soon as a central government (or at least ours) takes over total control of education, then it's over for our country. They will produce generations of students who don't know how to question authority in its many forms or be creative, and everything (economy, civil society, etc) will eventually implode. This is why the $50 billion+ budget for the Department of Education really scares me. The reason that money has been appropriated has little to do with improving education and a lot to do with gaining federal leverage over school funding - and by extension, school curriculum.
Alternatively, you could view this as a business opportunity, since you're one of the "smart ones" who realizes what's happening. Just find some sort of useless shit to sell that all the idiots being turned out by public education will just snap up, and you could become rich! Personally, I'm leaning towards trying to figure out a way to exploit the overly religious (since so many people will just buy anything if they think it comes from a "Christian company", etc).
You're completely right in this post...you do need some sort of actual code signing in order to truly verify the source of a binary or source tarball that you download.
That said, I still think the Linux situation where at least MD5's get checked is preferable to the Windows situation where a very small amount of software is signed and the vast majority is not verified in any way. Yes, SHA1 would be better, and from what I can see most distros, etc are slowly moving in that direction. Firefox does include SHA1 hashes in addition to the MD5 and code signatures, btw.
You're right that it's possible a single compromised mirror could get tainted code to a lot of people. But (at least I don't) know what sort of procedures the mozilla.org people have in place to prevent this sort of thing. I'd imagine they do at least check for file integrity periodically and/or use Tripwire or similar on mirrors. That wouldn't necessarily prevent the distribution of cracked code, but it almost certainly would ensure that the break-in would be discovered very quickly.
I'm no expert on how Windows code signing works...but, ummm...how is that different from how Microsoft does things?
Also note that most people who download Firefox will end up getting it from a mirror. So for any trojanning of the binary to go undetected they would have to compromise both the main mozilla.org server and the mirror site (or have the mozilla.org one go undetected for long enough for the changes to propogate to all mirrors). I have a feeling that this scenario isn't too likely since someone would probably notice the problem long before everything got all synced up to EVERY mirror. Does Mozilla.org even copy stuff to mirrors except when they release a new version??
I'm not saying that the current system is totally secure, and there is room for improvement, but it's certainly not inferior to the Windows code-signing. Many Linux distros automatically check signatures or at least checksums, accomplishing virtually the same thing the IE code signing does. It's not worse just because it isn't the same exact method that Microsoft uses for their own products (and I've NEVER (with a handful of exceptions) seen MS "signed" code for Windows other than Microsoft products). Every single package I install has its MD5 sum checked, which is arguably better than what Windows offers.
I don't think the issue is that Gentoo is against providing binaries out of some bias against them. The problem is simply a matter of resources. I've been reading the Gentoo mailing lists for several years now, and this has been brought up many times. The thing is, maintaining binaries that are compatible with each other and fully QA-tested is a lot of work (look at how long it takes Debian, even with a very large organization). Gentoo just doesn't have the amount of people necessary to develop and maintain binary packages for every single package in the tree while also maintaining all the from-source ones and adding major new features to the distro.
There's also an issue of disk space on the Gentoo mirrors. The Gentoo mirrors already take many gigabytes of space for a full source mirror, and adding binaries would at least double that. Given that the mirror space is mostly donated to the project, they can't really diplomatically go in and demand that their donors provide them with several gigabytes more space and the bandwidth to go with it.
You do lose some of the benefits of Gentoo when you use binary packages, such as control over what USE flags something is compiled with. That said, that's probably only an issue if some specific optional feature is really important to you. I don't use most binary packages on Gentoo, but I do use them for Firefox, Thunderbird, Eclipse, Java, and Ooffice because IMHO the compile time for those is excessive. Almost everything else my machine can handle in a matter of minutes.
Ok, I can give you specifics. I think the problem is caused by some incompatibility between SP2 and my wireless LAN card's drivers. It doesn't happen if I don't have the card in there. I need to use the card, so there isn't anything I can do to work around this problem. Unfortunately for me the manufacturer hasn't released any patches to the (buggy I'm sure) drivers. From what I've gathered online they rely on an undocumented interface in Windows that was broken by SP2. BTW, uninstalling and reinstalling SP2 didn't help. Microsoft's site actually acknowledges the problem with the blue screens and the specific DLL updated by SP2 that causes them, but they don't have any patch available yet.
You were lucky. XP has bluescreened and rebooted every 5 minutes (seriously, no joke) for me ever since I installed SP2. I simply followed Microsoft's instructions for installation. Their update fucked things up, not me.
Fortunately I only use Windows when I'm forced to (like when some SPECIFIC software I absolutely need is not available for Linux). I haven't bothered to try to fix it yet because I have enough stress from my job.
Y'know, Windows really, really, really sucks as an operating system if it's not capable of updating itself while running other software (ESPECIALLY since you can't easily exit the GUI). Does anyone know exactly why this is? Is Windows incapable of updating files while they are open, or does its scheduler suffer from major, serious bugs? What particular aspect of Windows precludes it from updating without devoting my entire computer to the (overly lengthy) process?
I know it's gotten better, but it's still absolutely laughable compared to Linux, Solaris, etc. Why exactly is the data stored on the filesystem seemingly not independent of the currently running processes, until they are restarted? It seems like Windows has some serious fundamental technical failings that should have been addressed like 10 years ago.
You know what I meant. Of course someone has to fix the dependent apps if the API changed. What I was saying is that autoconf is able to fix this "version change breakage" a lot of the time if you recompile, because it determines what you have installed on your system. In Redhat, etc you frequently have to wait for the distro to put out a new binary package that fixes it (by recompiling!). On Gentoo if a dependent package gets broken by some library upgrade you can just recompile it from the upstream source and it will usually sort itself out.
If the actual API changed someone has to fix it, yes. But even that isn't so much of a problem on Linux as on Windows because you usually have access to the source and can fix it yourself if necessary. So even if no one else has fixed it yet you can still get yourself up and running in critical situations. On Windows you're just SOL until MS/whoever issues a patch.
I think you may be missing the point of OSS. These things (breaks to backwards compatibility) aren't really as much of a problem on Linux as they would be on Windows because virtually all of the code in question is available in source form. You can always fix the problem by recompiling stuff. On Windows if an operating system API changes you have to wait for whoever made all of your software to fix and recompile it and then redownload/repurchase it. This is part of why Microsoft is unable to fix many longstanding problems with Windows and the Windows API: they are slaves to backwards compatibility. In fact the whole.NET thing seems to me to be an attempt to escape from this limitation and enjoy some of the benefits that open source now does.
Virtually all of the problems you describe are problems with binary packaging rather than with the core Linux software itself (with a very small number of exceptions such as the GCC 2.9.x -> GCC 3.x transition; and even that was fixable through recompilation). All I can say is get a better distro. Debian doesn't have so many problems with this, and Gentoo and other source-based distros certainly don't either. This is in fact why I stopped using Red Hat and switched to Gentoo in early 2002. You don't have many binary version compatibility problems on Gentoo because it doesn't use binary packages except where the software is not available in any other form. It thus manages to parallel the development of most open source software very well: it isn't a problem when developers break an API...you just use a single command to recompile everything that was broken.
Open source developers often don't worry about maintaining binary compatibility because it isn't a problem if you just recompile. Using binary packages just invites problems: whoever makes your distro has to stay on top of the constant changes in the API. So if you do use binary packages, at least do yourself a favor and use a well-tested distro like Debian.
This is a misconception. I think it might conflict with Debian's Free Software ethics, but Gentoo has quite a bit of commercial software in their repository (BitKeeper, Intel C/C++ compilers, many games off the top of my head). There is no reason why commercial developers can't just let you install their software through a repository and then have you purchase a license key through their website or retail channels. If they are worried about piracy they could even have the files you download be encrypted ala Half Life 2 and Steam. In the future I would imagine that repository-based systems might be expanded to directly handle licensing issues (Gentoo is working on this as well). No need to go through the usability hell of web-surfing just to download ANY software.
Just because software is commercial doesn't mean you can't have a standard way of installing it via a single command. Maybe you're equating the Windows Way with commercial software in general, but just because Windows has a broken model doesn't mean that Linux can't do better.
Best Buy, as far as I can tell, already has priced themselves out of the market for anyone but idiots on many items. Practically the only things they have at reasonable prices at all anymore are the things they have on sale - and they have the guts to complain that people only buy those! Most of their electronics are both obsolete with comparison to what is available online and outrageously priced...I've seen them selling the exact same component of RAM for more than double the price online.
I agree it's a perfectly reasonable model to make your money from a reputation for great service (even at higher prices). But Best Buy sure as hell isn't succeeding at that. How can they claim to provide "great" service when virtually all their stores are staffed by low-paid idiots who know nothing about the products they sell and have no problem with blatently lying to unsavvy consumers?
Screw Best Buy....I can't wait until they completely bite the dust. I think they're completely headed for oblivion if they don't reform their business model. They're not going to win on price or selection against the online places...all they can really offer is immediate gratification and good customer satisfaction/service. Since most people won't stand for completely getting ripped off on a big ticket item just for the immediate gratification, they're screwed if they don't start doing a hell of a lot better on service.
I noticed this too. I looked at the data for several schools I'm familiar with and the data was completely wrong in numerous areas on all of them. This "survey" is completely invalid if half of their data is wrong. Seriously, how hard is it to figure out the answers to these simplistic questions? I could answer almost all of these about every university just by visiting their webpage.
I think the whole concept of using polling as a way of deciding eligibility is pretty morally bankrupt. I've suspected for a while that the reason for that is to make it possible for the major parties to manipulate third parties out of the contest. Perot gave them a good scare in '92, and they've been tightening the screws on our republic ever since.
Eligibility should be decided on a more legalistic basis: if the electoral votes of the states that a candidate is officially ballot-qualified for exceeds 270, then the candidate should get a chance to debate because they have a real chance of being elected. Arguing that they do not have a chance because they are not a Republican or Democrat becomes a self-fulfilling prophecy. The only person who could possibly argue that a candidate that is ballot-qualified in 30+ states doesn't even deserve a chance to debate the other candidates is either a partisan shill or someone who has been manipulated by the partisan shills, IMHO.
Think about that: thanks to the "system" for ballot access that has been put into place, the minor party candidates have collected signatures from tens of thousands of voters all over the country. They have met the legal requirements for ballot access in many cases. But they aren't even given a chance to have a serious policy debate, because the debates are now controlled by the "major" parties. Those parties have concluded that the opinions of all those tens of thousands of people who signed petitions for Nader, Badnarik, et al don't mean shit to them as long as they can hold onto power without having to run the country well. Nice what our republic is turning into.
I think Gentoo has a policy of using bzip2 to compress all the source tarballs that they mirror themselves. Gzip is of course still used extensively for files that Gentoo legally cannot mirror on their own servers or cannot repackage from a binary format.
I think this story is relevant to the book because it concerns the actual title of the book. It's not just the author or publisher that is being attacked; it's the book itself and its horrible choice of title. Amazon certainly has every right to delete postings on their own website, but I think it's a little misleading to say that they are being impartial or fair if they only delete posts that might negatively impact book sales.
In this case, though, a lot of the negative reviews seemed to clearly violate the guidelines and the volume of new reviews coming probably just prompted them to stop spending the time reading them and do a blanket delete. I would wait a while for things to calm down and then post an intelligent critique of the book that doesn't violate their guidelines if you want people to be able to read about this side of the story. You'd also probably get a lot more credibility if you read the book or at least commented on the content of it. I'm sure you can find it at a local library if you don't want to give money to the author and publisher.
Slashdot Mirror
The stupid autodetect thing came up for me too when I first clicked on it, but it gave me the option to ignore the autodetection and proceed anyway. After I did that it set some cookie telling it not to autodetect again, and now it goes straight to the video page when I click on a video link.
You might try deleting your CNN cookies and trying it again. What browser are you using and does it get to the "brushed metal" looking video page or does it never get past the autodetection thing?
It works 100% fine for me with the mplayer plugin under Linux (w/ the WMP codecs, of course).
You can even raise this issue against the entire Internet (disregarding the anonymity issue altogether): should we ban the Internet because people distribute child pornography through it? Are ISPs legally held responsible when their users distribute child porn without their knowledge? The answer is no, because ISPs are a common carrier and are not required to police their users unless the government (or others, like the DMCA notices) brings specific evidence of wrongdoing to them. How is Tor any different? The people who run nodes are only acting as a carrier channel for whatever passes through and do not have to police the traffic or be held responsible.
I can see governments getting pissed off if things like Tor get too powerful and popular, since it would undermine their ability to eavesdrop on people EASILY. However, governments and societies managed just fine before they could instantly and anonymously spy on their own citizens. They just used other means to find the information about who was breaking laws. Technology should not be an excuse for allowing a surveillance society just so that it's more convenient to catch "bad guys".
In my opinion, Windows is only superficially "easier" to administer than Unix systems. It appears to be easier at first, but it actually isn't because it becomes a major pain when problems crop up. What is a company that has only hired shoddy, unknowledgable Windows admins going to do when that happens? Bring in a consultant or two at astronomical cost? Sorta kills those cost savings from being able to hire unskilled admins if you have to do that a few times a year.
Governments that I've dealt with seem to be notorious for this kind of mismanagement. One federal agency that I know of just sinks millions upon millions of dollars into whatever new "solution" Microsoft puts out every couple of years, and then hires idiots who couldn't administer their way out of a paper bag to run it. I'm sure the rest do the same thing.
I think the only thing that Excel has to recommend it whatsoever as far as large-scale data analysis is exactly the thing you just mentioned: it's easy to learn how to use. It's not very powerful, but a trained monkey can be up and running on it in 5 minutes. That's probably been a good business model for MS.
In the long run, you're MUCH better off learning how to use some more powerful tool. Whatever time this costs you in the short term will be saved many times over in the end.
Oh, and Excel STILL is not the best thing to use even just for "interoperability" with other people who use it. CSV (comma separated value) files are what I use for that because they're a helluva lot more universal than Excel and they're easy to use to hook Excel or whatever up to Perl or some other real programming language.
Maybe Excel is good for financial stuff or something, I don't know. I haven't found a real use for it yet.
I agree, Excel is a joke compared to Matlab as far as scientifici functionality. There are some things that are annoying to me about Matlab, like the way its plotting interface works, but it's still way more powerful and flexible.
Almost all the real scientists I've met use Matlab or SPSS or something else more powerful than Excel if they need to do data analysis. Excel is fine if you're just doing averages and standard deviations and other basic stuff, but if you need to do more advanced numerical or statistical stuff you really start to appreciate Matlab's standard library. Maybe the scientists who use Excel use third-party addons for more mathematical functionality, or maintain their own library of functions for it?
You're a developer if you write code for a webpage. Especially if it's a professional/business site, or a page created by a CMS, etc. And in fact you could argue that the way it should be is that only developers should write actual HTML/XHTML code. People who aren't capable of writing it properly should be using some sort of content management system or WYSIWYG tool to write the code for them.
Leave the coding to people who know what they are doing and can do it correctly, and have users who can't use software that lets them publish web content without knowing how. This is how it should be and how it is going, anyway. Witness the popularity of blogging: this is possible because easy to use blog software lets people who aren't coders write their own webpages. Then if there are problems with the HTML then it's just a bug in the publishing software and not the users' problem to fix.
What's the problem? You should be fixing all the bugs like that before you release your page publically. And unlike regular HTML you'll actually know they're there in XHTML. Or is the idea that you shouldn't have to test your code beyond "looks good to me!"
Story
They were implicated in some sort of shady sharing of consumer data/credit card numbers.
There's a certain irony to the idea of banning flag-burning. The flag is supposed to represent an ideal of freedom, but you're not free to burn it? Like it or not, burning the flag is clearly political speech. It shows more respect for the flag and what it stands for to let everyone keep the freedom to burn it if they choose. If we take away the freedom to participate in any political speech we want to, then the flag becomes just a meaningless piece of cloth undeserving of protection anyway.
I think you're correct about that - or at least that's how the schools see their role these days. They're really more of a venue for people to push whatever political agenda they personally have than for any real education to take place.
This is the problem with public education in the first place. The government will seek to interfere if given control and the ability to do so. IMHO, the only reason we avoided that kind of crap for a long time in this country was because public education was largely decentralized - funded, run, and controlled by local and state government. But notice that these days the federal government is seeking to interfere more and more all the time? As soon as a central government (or at least ours) takes over total control of education, then it's over for our country. They will produce generations of students who don't know how to question authority in its many forms or be creative, and everything (economy, civil society, etc) will eventually implode. This is why the $50 billion+ budget for the Department of Education really scares me. The reason that money has been appropriated has little to do with improving education and a lot to do with gaining federal leverage over school funding - and by extension, school curriculum.
Alternatively, you could view this as a business opportunity, since you're one of the "smart ones" who realizes what's happening. Just find some sort of useless shit to sell that all the idiots being turned out by public education will just snap up, and you could become rich! Personally, I'm leaning towards trying to figure out a way to exploit the overly religious (since so many people will just buy anything if they think it comes from a "Christian company", etc).
You're completely right in this post...you do need some sort of actual code signing in order to truly verify the source of a binary or source tarball that you download.
That said, I still think the Linux situation where at least MD5's get checked is preferable to the Windows situation where a very small amount of software is signed and the vast majority is not verified in any way. Yes, SHA1 would be better, and from what I can see most distros, etc are slowly moving in that direction. Firefox does include SHA1 hashes in addition to the MD5 and code signatures, btw.
You're right that it's possible a single compromised mirror could get tainted code to a lot of people. But (at least I don't) know what sort of procedures the mozilla.org people have in place to prevent this sort of thing. I'd imagine they do at least check for file integrity periodically and/or use Tripwire or similar on mirrors. That wouldn't necessarily prevent the distribution of cracked code, but it almost certainly would ensure that the break-in would be discovered very quickly.
I'm no expert on how Windows code signing works...but, ummm...how is that different from how Microsoft does things?
Also note that most people who download Firefox will end up getting it from a mirror. So for any trojanning of the binary to go undetected they would have to compromise both the main mozilla.org server and the mirror site (or have the mozilla.org one go undetected for long enough for the changes to propogate to all mirrors). I have a feeling that this scenario isn't too likely since someone would probably notice the problem long before everything got all synced up to EVERY mirror. Does Mozilla.org even copy stuff to mirrors except when they release a new version??
I'm not saying that the current system is totally secure, and there is room for improvement, but it's certainly not inferior to the Windows code-signing. Many Linux distros automatically check signatures or at least checksums, accomplishing virtually the same thing the IE code signing does. It's not worse just because it isn't the same exact method that Microsoft uses for their own products (and I've NEVER (with a handful of exceptions) seen MS "signed" code for Windows other than Microsoft products). Every single package I install has its MD5 sum checked, which is arguably better than what Windows offers.
I don't think the issue is that Gentoo is against providing binaries out of some bias against them. The problem is simply a matter of resources. I've been reading the Gentoo mailing lists for several years now, and this has been brought up many times. The thing is, maintaining binaries that are compatible with each other and fully QA-tested is a lot of work (look at how long it takes Debian, even with a very large organization). Gentoo just doesn't have the amount of people necessary to develop and maintain binary packages for every single package in the tree while also maintaining all the from-source ones and adding major new features to the distro.
There's also an issue of disk space on the Gentoo mirrors. The Gentoo mirrors already take many gigabytes of space for a full source mirror, and adding binaries would at least double that. Given that the mirror space is mostly donated to the project, they can't really diplomatically go in and demand that their donors provide them with several gigabytes more space and the bandwidth to go with it.
You do lose some of the benefits of Gentoo when you use binary packages, such as control over what USE flags something is compiled with. That said, that's probably only an issue if some specific optional feature is really important to you. I don't use most binary packages on Gentoo, but I do use them for Firefox, Thunderbird, Eclipse, Java, and Ooffice because IMHO the compile time for those is excessive. Almost everything else my machine can handle in a matter of minutes.
Ok, I can give you specifics.
I think the problem is caused by some incompatibility between SP2 and my wireless LAN card's drivers. It doesn't happen if I don't have the card in there. I need to use the card, so there isn't anything I can do to work around this problem. Unfortunately for me the manufacturer hasn't released any patches to the (buggy I'm sure) drivers. From what I've gathered online they rely on an undocumented interface in Windows that was broken by SP2.
BTW, uninstalling and reinstalling SP2 didn't help. Microsoft's site actually acknowledges the problem with the blue screens and the specific DLL updated by SP2 that causes them, but they don't have any patch available yet.
You were lucky. XP has bluescreened and rebooted every 5 minutes (seriously, no joke) for me ever since I installed SP2. I simply followed Microsoft's instructions for installation. Their update fucked things up, not me.
Fortunately I only use Windows when I'm forced to (like when some SPECIFIC software I absolutely need is not available for Linux). I haven't bothered to try to fix it yet because I have enough stress from my job.
Y'know, Windows really, really, really sucks as an operating system if it's not capable of updating itself while running other software (ESPECIALLY since you can't easily exit the GUI). Does anyone know exactly why this is? Is Windows incapable of updating files while they are open, or does its scheduler suffer from major, serious bugs? What particular aspect of Windows precludes it from updating without devoting my entire computer to the (overly lengthy) process?
I know it's gotten better, but it's still absolutely laughable compared to Linux, Solaris, etc. Why exactly is the data stored on the filesystem seemingly not independent of the currently running processes, until they are restarted? It seems like Windows has some serious fundamental technical failings that should have been addressed like 10 years ago.
You know what I meant. Of course someone has to fix the dependent apps if the API changed. What I was saying is that autoconf is able to fix this "version change breakage" a lot of the time if you recompile, because it determines what you have installed on your system. In Redhat, etc you frequently have to wait for the distro to put out a new binary package that fixes it (by recompiling!). On Gentoo if a dependent package gets broken by some library upgrade you can just recompile it from the upstream source and it will usually sort itself out.
If the actual API changed someone has to fix it, yes. But even that isn't so much of a problem on Linux as on Windows because you usually have access to the source and can fix it yourself if necessary. So even if no one else has fixed it yet you can still get yourself up and running in critical situations. On Windows you're just SOL until MS/whoever issues a patch.
I think you may be missing the point of OSS. These things (breaks to backwards compatibility) aren't really as much of a problem on Linux as they would be on Windows because virtually all of the code in question is available in source form. You can always fix the problem by recompiling stuff. On Windows if an operating system API changes you have to wait for whoever made all of your software to fix and recompile it and then redownload/repurchase it. This is part of why Microsoft is unable to fix many longstanding problems with Windows and the Windows API: they are slaves to backwards compatibility. In fact the whole .NET thing seems to me to be an attempt to escape from this limitation and enjoy some of the benefits that open source now does.
Virtually all of the problems you describe are problems with binary packaging rather than with the core Linux software itself (with a very small number of exceptions such as the GCC 2.9.x -> GCC 3.x transition; and even that was fixable through recompilation). All I can say is get a better distro. Debian doesn't have so many problems with this, and Gentoo and other source-based distros certainly don't either. This is in fact why I stopped using Red Hat and switched to Gentoo in early 2002. You don't have many binary version compatibility problems on Gentoo because it doesn't use binary packages except where the software is not available in any other form. It thus manages to parallel the development of most open source software very well: it isn't a problem when developers break an API...you just use a single command to recompile everything that was broken.
Open source developers often don't worry about maintaining binary compatibility because it isn't a problem if you just recompile. Using binary packages just invites problems: whoever makes your distro has to stay on top of the constant changes in the API. So if you do use binary packages, at least do yourself a favor and use a well-tested distro like Debian.
This is a misconception. I think it might conflict with Debian's Free Software ethics, but Gentoo has quite a bit of commercial software in their repository (BitKeeper, Intel C/C++ compilers, many games off the top of my head). There is no reason why commercial developers can't just let you install their software through a repository and then have you purchase a license key through their website or retail channels. If they are worried about piracy they could even have the files you download be encrypted ala Half Life 2 and Steam. In the future I would imagine that repository-based systems might be expanded to directly handle licensing issues (Gentoo is working on this as well). No need to go through the usability hell of web-surfing just to download ANY software.
Just because software is commercial doesn't mean you can't have a standard way of installing it via a single command. Maybe you're equating the Windows Way with commercial software in general, but just because Windows has a broken model doesn't mean that Linux can't do better.
Best Buy, as far as I can tell, already has priced themselves out of the market for anyone but idiots on many items. Practically the only things they have at reasonable prices at all anymore are the things they have on sale - and they have the guts to complain that people only buy those! Most of their electronics are both obsolete with comparison to what is available online and outrageously priced...I've seen them selling the exact same component of RAM for more than double the price online.
I agree it's a perfectly reasonable model to make your money from a reputation for great service (even at higher prices). But Best Buy sure as hell isn't succeeding at that. How can they claim to provide "great" service when virtually all their stores are staffed by low-paid idiots who know nothing about the products they sell and have no problem with blatently lying to unsavvy consumers?
Screw Best Buy....I can't wait until they completely bite the dust. I think they're completely headed for oblivion if they don't reform their business model. They're not going to win on price or selection against the online places...all they can really offer is immediate gratification and good customer satisfaction/service. Since most people won't stand for completely getting ripped off on a big ticket item just for the immediate gratification, they're screwed if they don't start doing a hell of a lot better on service.
I noticed this too. I looked at the data for several schools I'm familiar with and the data was completely wrong in numerous areas on all of them. This "survey" is completely invalid if half of their data is wrong. Seriously, how hard is it to figure out the answers to these simplistic questions? I could answer almost all of these about every university just by visiting their webpage.
I think the whole concept of using polling as a way of deciding eligibility is pretty morally bankrupt. I've suspected for a while that the reason for that is to make it possible for the major parties to manipulate third parties out of the contest. Perot gave them a good scare in '92, and they've been tightening the screws on our republic ever since.
Eligibility should be decided on a more legalistic basis: if the electoral votes of the states that a candidate is officially ballot-qualified for exceeds 270, then the candidate should get a chance to debate because they have a real chance of being elected. Arguing that they do not have a chance because they are not a Republican or Democrat becomes a self-fulfilling prophecy. The only person who could possibly argue that a candidate that is ballot-qualified in 30+ states doesn't even deserve a chance to debate the other candidates is either a partisan shill or someone who has been manipulated by the partisan shills, IMHO.
Think about that: thanks to the "system" for ballot access that has been put into place, the minor party candidates have collected signatures from tens of thousands of voters all over the country. They have met the legal requirements for ballot access in many cases. But they aren't even given a chance to have a serious policy debate, because the debates are now controlled by the "major" parties. Those parties have concluded that the opinions of all those tens of thousands of people who signed petitions for Nader, Badnarik, et al don't mean shit to them as long as they can hold onto power without having to run the country well. Nice what our republic is turning into.
I think Gentoo has a policy of using bzip2 to compress all the source tarballs that they mirror themselves. Gzip is of course still used extensively for files that Gentoo legally cannot mirror on their own servers or cannot repackage from a binary format.
I think this story is relevant to the book because it concerns the actual title of the book. It's not just the author or publisher that is being attacked; it's the book itself and its horrible choice of title. Amazon certainly has every right to delete postings on their own website, but I think it's a little misleading to say that they are being impartial or fair if they only delete posts that might negatively impact book sales.
In this case, though, a lot of the negative reviews seemed to clearly violate the guidelines and the volume of new reviews coming probably just prompted them to stop spending the time reading them and do a blanket delete. I would wait a while for things to calm down and then post an intelligent critique of the book that doesn't violate their guidelines if you want people to be able to read about this side of the story. You'd also probably get a lot more credibility if you read the book or at least commented on the content of it. I'm sure you can find it at a local library if you don't want to give money to the author and publisher.