"They've actually done it. It's time to buy an ATI card."
I've been recommending nVidia cards to everyone who asks, simply because their Linux support has been leagues ahead of ATI (now AMD, for those who haven't been paying attention). If the specs are credible enough to create a quality Free driver, then I'll switch to AMD in a heart beat.
"I admit, I think $5/mo for TV listings is a lot. TV Guide provided that service, plus interviews and articles, for less. "
$5/mo. for TV listing is just a few cents/month more than a typical TV Guide subscription (which was about $4.17/month at the time we ended our subscription). If you buy the TV Guide at the checkout counter, you pay a LOT more per month than that. TV Guide at the store is sold by the week, not the month. It's been a few years since I bought one at the store, but it was more than $1.25 for one week. For about the last five to six years, TV Guide stopped providing synopses for most shows. probably to save space for those worthless interviews and articles while keeping production costs down. To make it worse, the TV Guide has shrunk dramatically in recent years. I've subscribed to several magazines in my life, and I've seen that a shrinking page count is a sure sign of operational stress. All the magazines that went through the contraction stage eventually closed down.
I also can't have my computer automatically schedule recordings with TV Guide listings. The best I can hope for is to manually sift through the weekly listings and manually set my recording schedule. The time and effort spent doing that costs a lot more than $1.25/month.
Even at the initial price of $5/month (which is the high water mark planned for only the first 3 months, by the way), Schedules Direct's subscription comes out to $1.25 per week. That's quite a bit less than any other organized alternative. If they hit their target price of $20/year, that will come out to be about 39 cents per week. That's a trivial cost for the time and effort saved by subscribing to the service. Nothing even close will ever come from AT&T, Comcast, Cablevision, Mediacomm, or any commercial provider.
"I suppose if nothing else it taught me the importance of hardware redundancy, though it seemed a high price to pay at the time."
My primary hard drive died yesterday without any warning. My big mistake was to run fsck and let it automatically repair errors. I should have just immediately hit the power button, installed a new drive, and dd the old one. That may not have worked either, as Linux was reporting hardware buffer errors on every few sectors, but at least the filesystem for the readable inodes wouldn't have been automatically rewritten. Almost everything on that drive is a total loss (sigh).
The silver lining is that when I switched from Fedora Core 6 to Kubuntu in March of this year, I did a complete/home backup to a different hard drive just to be on the safe side (just in case I had accidentally told Kubuntu to reformat my home partition). With that one stroke of paranoia, I've "only" lost the last six months of my computing life (passwords, emails) rather than the last two years (that was the last time I had done a backup). I didn't backup my web site, so it's a total loss. It's not that big a deal, though. I should be able to get the important content back through Google.
Now I'm looking into building a dedicated RAID 5 backup server. The expense is a small price to pay for peace of mind.
Konqueror, while one of the best file managers I've ever had the pleasure to use, is comparatively slow when used as a web browser. Firefox renders noticeably faster (often VERY noticeably faster) and has a much faster Javascript engine. Konqueror as a web browser has its strengths, but raw speed is not one of them.
That is an excellent business plan, ScrewMaster. I can think of only one thing you left out:
In the late 80's and early 90's, back when the music video channels actually showed music videos, I would spend many hours of my free time recording my collection of great music videos onto video tape. The VCR would be in paused record for 30-45 minutes, and then a good video was shown. I would record the video, then hit pause and wait for another good one (this was back when I had a LOT of free time).
Not only should record companies follow your advice for audio-only tracks, but they should do the same for the entire history of music videos. It seems that the record companies are working as hard as they possibly can to lose as much money as they possibly can as fast as they possibly can. They have to be beaten over the head to near-death with the wads of money we're trying to give them before they'll actually take it. All they have to do is offer what we want, and we'll be unable to throw money at them fast enough. But yet they resist.
"Microsoft using the BSD TCP stack didn't render the code "un-free." The code that was BSD licensed was never any less available because MS used it."
So if I ask Microsoft for changes to the BSD code they used, I can get it and continue that line of development (not that I'd want to. I've seen Microsoft example code, and it makes me shudder)? If so, then the code is still available and is Free. If not, then that line of code development is dead and therefore neither Free nor free.
Stripping BSD copyright notices, even if the code is relicenseable to be uniquely GPL, is plainly wrong on its face.
To address your point I quoted above: the BSD license most certainly does allow its code derivatives to become non-Free (and non-free) at every point of exchange. I'm not going to pass judgement on whether this is good or bad, as you are the final arbiter of how you release your own code. If you want all forks of your code to remain available to everyone who wants it, then use the GPL. If you care only about your initial release being available to anyone who wants it, then use the BSD license.
"Except that it's the author and copyright holder who did it, and he's allowed to do what he likes with his copyrighted works."
The submitter wrote some of the driver, but there are a few other names in the Copyright list. There is no information in the article indicating their (dis)approval.
"Since you already know that anything in that folder is my private propery, downloading from it make "you" the thief. "
Not only that, but the downloader may also be sued under the DMCA since you can argue that your folder name is an effective means to control copyrighted material under the (retarded) law.
That wouldn't be very effective for Microsoft. It doesn't matter under which Free license the mythical MS interoperability code is released since the concepts are not restricted. All it takes is for someone to document what the code does, then the Samba team could write its own implementation.
It means if you are using someone's software which contains patented code, and you then sue that someone for infringing on any patent you may hold, your right to use the code covered by that someone's patent automatically ends.
I think the casino is just out of luck. They have to prove that the players knew the machine was faulty, which is practically impossible to do in circumstances like these. Is getting more than you put in supposed to indicate fault? What about all of those X-for-one offers merchants frequently run in order to get customers? I would simply claim that I thought it was a casino promotion. Considering the staggering amounts of money casinos routinely lift from patrons, a 10-for-1 offer for one day at such a lavishly expensive place wouldn't even seem the slightest bit out of place. I would just be happy that I happened into the casino on the day the promotion was in effect.
Was there an "out of order" sign on the machine? If there was, did any patron remove it? No on both counts? Too bad for the casino.
"Simple answer, the Copyright Royalty Board gave SoundExchange a legal monopoly..."
IANAL, but the 2007 rules set by the CRB are probably unconstitutional and in excess of the authority retained by the US government. The US Constitution gives recording owners exclusive rights to the use of their recordings, including the right to distribute their recordings without royalties. That's the entire principle behind Free Software, and it applies equally well to any other copyrighted content.
"well, some algorithms (particularly brilliant million-man-hours type ones) should be patentable."
The problem is that there is no such thing. For every million-man-hour algorithm, a thousand different legal entities contributed at some point along the way. The final contributor should not be allowed to monopolize the work. ALL significant algorithms were created with the help of many little shoulders.
Even if that weren't true, all algorithms are mathematical expressions and should not be patentable for that reason alone.
"Internet Explorer protected mode in Vista...don't have the rights to access much."
Does Internet Explorer come this way by default? If not, then it's of no use to 99+% of Vista victims...err...users since they won't change the defaults.
"The 80 character width design was set up for a reason. Don't reject it without understanding the original reason and considering whether it still applies."
The use of 80 characters for text terminals had absolutely nothing to do with readability. Hollerith punch cards were 80 characters wide, and interactive terminals later adopted the width to conform with the standard set by the punch card.
My personal programming habits evolved out of necessity. I started C programming using K&R syntax. When the insanity of that practice finally dawned on me, and I started to line up my braces, I used two spaces to indent (and derided Linus' insistence on 8-space tabs as useless and a sign of inferiority). After realizing how much longer I needed to spot subtle error with such a ludicrously narrow tab width (which I had used for years, by the way), I decided to swallow my pride and give Linus' advice on tabs a test drive.
It was one of those moments when the cartoon light bulb shined bright above my head. Subtle structural problems became much easier to see, so I realized that Linus was right all along in this regard (he's still wrong about K&R -shudder-).
When I was forced to take a COBOL class in college, the instructor introduced me to vowels in my variable names (POSIX operating systems commands could stand to discover vowels too) and descriptive names. Instead of a function called chkcd(), I learned that one could use checkcode(). It also got my started alphabetizing my variable names and function declarations in my class headers. My VB and Java classes expanded upon vowel usage to make checkcode() become checkCode(). I would much rather read "checkDeviceForFaults()" than chkdvft(). The former is self-evident to anyone who has to maintain my code, while the latter would leave me scratching my head over my own code after a few weeks away from it. In fact, many of the practices some Slashdot readers are proposing were the very practices I used which had that very effect on me and others.
I fought against camel case for several years before finally seeing their benefit. I also fought against variable type prefixes for a long time until I started seeing good uses for them. If I had the time and inclination, I could go on for many chapters about practices I abhorred in my youthful ignorance but ended up adopting because of self-evident usefulness.
There are perfectly good reasons to have multiple levels of indentation, leading to perfectly good reasons to want more than 80 columns on a screen. So yes, 80 characters are insufficient.
Holmwood, you're comparing apples and oranges here. In your comparison, the closed-source system has a paper trail while the open source system does not. Here's a more apt and fair comparison:
1) The closed source system has a touch screen and no paper trail. The company making the machines likes George Bu...err...likes a particular candidate, and accusations fly that the company has embedded a mechanism to ensure that the shrub wins. The code cannot be reviewed to prove its security and consistency, so the winner's entire presidency is clouded with an air of illegitimacy. It is also shown that the results can be tampered with in real-time during the election, but still nobody is allowed to review the code without the company's permission -- which will never be given. Malicious computer hackers discover exploitable holes, but tell no one. This leaves them with the capacity to rig elections undetected. The election process as a whole is widely discredited, and the entire political system begins to crumble.
2) The Open Source system has a touch screen and no paper trail (all proposed Open Source systems do indeed have a paper tail, so this opening statement is false, and is only for fair comparison under the same conditions). The law requiring Open Source voting machines also requires the actual installed code (authenticated under criminal penalties for non-compliance) to be posted on the Internet for transparency purposes. A potentially exploitable bug is discovered by malicious hackers. However, since the code is available to everyone, the same bug is also discovered by the good guys. The good guys provide the fix, which is certified for the next election (it's too late for the current election since it's illegal to change the code after the election begins), the affected machines are taken offline as a precaution, and the backup voting system (via paper ballots) is activated for the remainder of the election. Trust in the electoral system is maintained, the machines get new software for the next election, and life goes on.
In reality, the Open Source systems have paper trails. So when the affected machines are taken offline, the existing ballots are already in paper form and can be verified with the rest of the ballots.
So yes, Open Source is crucial for electronic voting systems. Widespread closed source voting systems would be an unmitigated disaster.
"Same with Adobe not making software for Microsoft because Microsoft sells competing software."
There's a key difference between those relations and the relationships between Microsoft and OEM's. Here's your example in the context of Microsoft/OEM relationships:
Let's pretend that Microsoft makes a plugin for Adobe Photoshop. Let's also pretend that Adobe Photoshop is largely useless (from a sales perspective) without this plugin, and Adobe cannot sell copies of Photoshop without it. The profit margin on the sale of a copy of Photoshop is so small as to be practically zero, but Microsoft pays Adobe for each copy of Photoshop that is sold with the plugin. This payment is enough to keep Adobe operating profitably.
Suddenly, Microsoft is selling its own version of Photoshop with the plugin. Adobe is also no longer allowed to sell Photoshop with the Microsoft plugin (Microsoft pulls the license out from under Adobe), so Microsoft becomes the only company providing Photoshop with the plugin. People stop buying Photoshop from Adobe, because the Microsoft version is cheaper, and Adobe goes bankrupt.
This is the kind of relationship that Microsoft has with OEM's, so let's go back to the real world: If Microsoft begins selling full PC's in India, and the campaign is successful, it becomes quite attractive to try the same tactic in Dell's markets. Don't believe for a second that Microsoft isn't doing this as a trial balloon for traditional markets. If Microsoft stops paying OEMs to include Windows, there's going to be a lot of pain. If Microsoft made every OEM go cold turkey at the same time, it wouldn't hurt as much since everyone would scale down by the same amount. But I suspect that Microsoft would attack OEM's piecemeal in a divide and conquer strategy.
I'm sure Dell sees the possibilities also, which may be why the company decided to try creating a new market populated by penguins and devoid of Microsoft.
This certainly isn't a general solution, but my spam problem is gone. I run my own email server off of my DSL (I have a business class DSL contract with a static IP address -- which is only $10/month more than a dynamic IP address contract which forbids running servers). I have a different email address provided to each company I do business with, and a small pool of email addresses I give to new people who express a desire to email me (with a permanent alias created once their intent is verified), all routed via my system's email alias file to my real email address which is never disclosed to anyone.
If I start getting spam on one of the aliases, I know who's responsible for providing the address. I can simply disable the alias and create a new one. While it seems like extra work, it's relatively simple once the routine is established.
Loki Software proved the lack of market for Linux games 5 years ago when they shut down in 2002. Not quite. Loki Software proved that a small startup gaming company for a small gaming market cannot survive when its president is draining the company coffers for personal extravagances. If you factor back in all of the large amounts of money that Scott was taking out of Loki's operations and throwing away for personal use, Loki was more than self-sustaining.
"...there are presumably fewer security checks that the person requesting the certificate is who he says he is."
When I setup my employer's online payment system, I went to VeriSign's site (which I opposed, but I was overruled) and signed up for a certificate. I got a call from VeriSign that went something like this:
VeriSign: Are you the person who requested a certificate?
Me: Yes.
VeriSign: Send us the money and we'll email your certificate to you.
The certificate arrived in unencrypted email! I kid you not.
That was hardly inspiring security. Considering that anyone who was eaves dropping on our unencrypted email traffic could have intercepted our certificate at the source, identity verification via CA's is next to worthless. Certificate Authorities are just con artists preying on people's fears.
The only thing a certificate is actually good for is to ensure that traffic between Point A and Point B is being encrypted sufficiently well to be meaningless to anyone who intercepts it. For that, a VeriSign certificate is no better than a self-signed certificate. I would prefer a self-signed certificate, because then at least it isn't susceptible to interception before implementation.
Slashdot Mirror
"They've actually done it. It's time to buy an ATI card."
I've been recommending nVidia cards to everyone who asks, simply because their Linux support has been leagues ahead of ATI (now AMD, for those who haven't been paying attention). If the specs are credible enough to create a quality Free driver, then I'll switch to AMD in a heart beat.
"I admit, I think $5/mo for TV listings is a lot. TV Guide provided that service, plus interviews and articles, for less. "
$5/mo. for TV listing is just a few cents/month more than a typical TV Guide subscription (which was about $4.17/month at the time we ended our subscription). If you buy the TV Guide at the checkout counter, you pay a LOT more per month than that. TV Guide at the store is sold by the week, not the month. It's been a few years since I bought one at the store, but it was more than $1.25 for one week. For about the last five to six years, TV Guide stopped providing synopses for most shows. probably to save space for those worthless interviews and articles while keeping production costs down. To make it worse, the TV Guide has shrunk dramatically in recent years. I've subscribed to several magazines in my life, and I've seen that a shrinking page count is a sure sign of operational stress. All the magazines that went through the contraction stage eventually closed down.
I also can't have my computer automatically schedule recordings with TV Guide listings. The best I can hope for is to manually sift through the weekly listings and manually set my recording schedule. The time and effort spent doing that costs a lot more than $1.25/month.
Even at the initial price of $5/month (which is the high water mark planned for only the first 3 months, by the way), Schedules Direct's subscription comes out to $1.25 per week. That's quite a bit less than any other organized alternative. If they hit their target price of $20/year, that will come out to be about 39 cents per week. That's a trivial cost for the time and effort saved by subscribing to the service. Nothing even close will ever come from AT&T, Comcast, Cablevision, Mediacomm, or any commercial provider.
"I suppose if nothing else it taught me the importance of hardware redundancy, though it seemed a high price to pay at the time."
/home backup to a different hard drive just to be on the safe side (just in case I had accidentally told Kubuntu to reformat my home partition). With that one stroke of paranoia, I've "only" lost the last six months of my computing life (passwords, emails) rather than the last two years (that was the last time I had done a backup). I didn't backup my web site, so it's a total loss. It's not that big a deal, though. I should be able to get the important content back through Google.
My primary hard drive died yesterday without any warning. My big mistake was to run fsck and let it automatically repair errors. I should have just immediately hit the power button, installed a new drive, and dd the old one. That may not have worked either, as Linux was reporting hardware buffer errors on every few sectors, but at least the filesystem for the readable inodes wouldn't have been automatically rewritten. Almost everything on that drive is a total loss (sigh).
The silver lining is that when I switched from Fedora Core 6 to Kubuntu in March of this year, I did a complete
Now I'm looking into building a dedicated RAID 5 backup server. The expense is a small price to pay for peace of mind.
"I can't believe they left out Konqueror!"
Konqueror, while one of the best file managers I've ever had the pleasure to use, is comparatively slow when used as a web browser. Firefox renders noticeably faster (often VERY noticeably faster) and has a much faster Javascript engine. Konqueror as a web browser has its strengths, but raw speed is not one of them.
That is an excellent business plan, ScrewMaster. I can think of only one thing you left out:
In the late 80's and early 90's, back when the music video channels actually showed music videos, I would spend many hours of my free time recording my collection of great music videos onto video tape. The VCR would be in paused record for 30-45 minutes, and then a good video was shown. I would record the video, then hit pause and wait for another good one (this was back when I had a LOT of free time).
Not only should record companies follow your advice for audio-only tracks, but they should do the same for the entire history of music videos. It seems that the record companies are working as hard as they possibly can to lose as much money as they possibly can as fast as they possibly can. They have to be beaten over the head to near-death with the wads of money we're trying to give them before they'll actually take it. All they have to do is offer what we want, and we'll be unable to throw money at them fast enough. But yet they resist.
"Microsoft using the BSD TCP stack didn't render the code "un-free." The code that was BSD licensed was never any less available because MS used it."
So if I ask Microsoft for changes to the BSD code they used, I can get it and continue that line of development (not that I'd want to. I've seen Microsoft example code, and it makes me shudder)? If so, then the code is still available and is Free. If not, then that line of code development is dead and therefore neither Free nor free.
Stripping BSD copyright notices, even if the code is relicenseable to be uniquely GPL, is plainly wrong on its face.
To address your point I quoted above: the BSD license most certainly does allow its code derivatives to become non-Free (and non-free) at every point of exchange. I'm not going to pass judgement on whether this is good or bad, as you are the final arbiter of how you release your own code. If you want all forks of your code to remain available to everyone who wants it, then use the GPL. If you care only about your initial release being available to anyone who wants it, then use the BSD license.
"Except that it's the author and copyright holder who did it, and he's allowed to do what he likes with his copyrighted works."
The submitter wrote some of the driver, but there are a few other names in the Copyright list. There is no information in the article indicating their (dis)approval.
"Since you already know that anything in that folder is my private propery, downloading from it make "you" the thief. "
Not only that, but the downloader may also be sued under the DMCA since you can argue that your folder name is an effective means to control copyrighted material under the (retarded) law.
[slipper slope, non-sequitur snipped]
"...How about gay sex?"
Are both chicks hot?
The former CEO of the now defunct "The SCO Group", Darl McBride, has recently invested heavily in Soap On A Rope stocks.
That wouldn't be very effective for Microsoft. It doesn't matter under which Free license the mythical MS interoperability code is released since the concepts are not restricted. All it takes is for someone to document what the code does, then the Samba team could write its own implementation.
"I don't understand this - can someone explain?"
It means if you are using someone's software which contains patented code, and you then sue that someone for infringing on any patent you may hold, your right to use the code covered by that someone's patent automatically ends.
I think the casino is just out of luck. They have to prove that the players knew the machine was faulty, which is practically impossible to do in circumstances like these. Is getting more than you put in supposed to indicate fault? What about all of those X-for-one offers merchants frequently run in order to get customers? I would simply claim that I thought it was a casino promotion. Considering the staggering amounts of money casinos routinely lift from patrons, a 10-for-1 offer for one day at such a lavishly expensive place wouldn't even seem the slightest bit out of place. I would just be happy that I happened into the casino on the day the promotion was in effect.
Was there an "out of order" sign on the machine? If there was, did any patron remove it? No on both counts? Too bad for the casino.
"Simple answer, the Copyright Royalty Board gave SoundExchange a legal monopoly..."
IANAL, but the 2007 rules set by the CRB are probably unconstitutional and in excess of the authority retained by the US government. The US Constitution gives recording owners exclusive rights to the use of their recordings, including the right to distribute their recordings without royalties. That's the entire principle behind Free Software, and it applies equally well to any other copyrighted content.
"well, some algorithms (particularly brilliant million-man-hours type ones) should be patentable."
The problem is that there is no such thing. For every million-man-hour algorithm, a thousand different legal entities contributed at some point along the way. The final contributor should not be allowed to monopolize the work. ALL significant algorithms were created with the help of many little shoulders.
Even if that weren't true, all algorithms are mathematical expressions and should not be patentable for that reason alone.
"Internet Explorer protected mode in Vista...don't have the rights to access much."
Does Internet Explorer come this way by default? If not, then it's of no use to 99+% of Vista victims...err...users since they won't change the defaults.
"Why not a programming mark up language akin to way that HTML is a presentation language?"
Because programming languages are not suitable to be read like a book?
"The 80 character width design was set up for a reason. Don't reject it without understanding the original reason and considering whether it still applies."
The use of 80 characters for text terminals had absolutely nothing to do with readability. Hollerith punch cards were 80 characters wide, and interactive terminals later adopted the width to conform with the standard set by the punch card.
My personal programming habits evolved out of necessity. I started C programming using K&R syntax. When the insanity of that practice finally dawned on me, and I started to line up my braces, I used two spaces to indent (and derided Linus' insistence on 8-space tabs as useless and a sign of inferiority). After realizing how much longer I needed to spot subtle error with such a ludicrously narrow tab width (which I had used for years, by the way), I decided to swallow my pride and give Linus' advice on tabs a test drive.
It was one of those moments when the cartoon light bulb shined bright above my head. Subtle structural problems became much easier to see, so I realized that Linus was right all along in this regard (he's still wrong about K&R -shudder-).
When I was forced to take a COBOL class in college, the instructor introduced me to vowels in my variable names (POSIX operating systems commands could stand to discover vowels too) and descriptive names. Instead of a function called chkcd(), I learned that one could use checkcode(). It also got my started alphabetizing my variable names and function declarations in my class headers. My VB and Java classes expanded upon vowel usage to make checkcode() become checkCode(). I would much rather read "checkDeviceForFaults()" than chkdvft(). The former is self-evident to anyone who has to maintain my code, while the latter would leave me scratching my head over my own code after a few weeks away from it. In fact, many of the practices some Slashdot readers are proposing were the very practices I used which had that very effect on me and others.
I fought against camel case for several years before finally seeing their benefit. I also fought against variable type prefixes for a long time until I started seeing good uses for them. If I had the time and inclination, I could go on for many chapters about practices I abhorred in my youthful ignorance but ended up adopting because of self-evident usefulness.
There are perfectly good reasons to have multiple levels of indentation, leading to perfectly good reasons to want more than 80 columns on a screen. So yes, 80 characters are insufficient.
"I'm sure that it cost him a fortune to defend a suit against Google."
He was not defending a suit against Google, as he was the plaintiff. Google was defending against a suit.
Holmwood, you're comparing apples and oranges here. In your comparison, the closed-source system has a paper trail while the open source system does not. Here's a more apt and fair comparison:
1) The closed source system has a touch screen and no paper trail. The company making the machines likes George Bu...err...likes a particular candidate, and accusations fly that the company has embedded a mechanism to ensure that the shrub wins. The code cannot be reviewed to prove its security and consistency, so the winner's entire presidency is clouded with an air of illegitimacy. It is also shown that the results can be tampered with in real-time during the election, but still nobody is allowed to review the code without the company's permission -- which will never be given. Malicious computer hackers discover exploitable holes, but tell no one. This leaves them with the capacity to rig elections undetected. The election process as a whole is widely discredited, and the entire political system begins to crumble.
2) The Open Source system has a touch screen and no paper trail (all proposed Open Source systems do indeed have a paper tail, so this opening statement is false, and is only for fair comparison under the same conditions). The law requiring Open Source voting machines also requires the actual installed code (authenticated under criminal penalties for non-compliance) to be posted on the Internet for transparency purposes. A potentially exploitable bug is discovered by malicious hackers. However, since the code is available to everyone, the same bug is also discovered by the good guys. The good guys provide the fix, which is certified for the next election (it's too late for the current election since it's illegal to change the code after the election begins), the affected machines are taken offline as a precaution, and the backup voting system (via paper ballots) is activated for the remainder of the election. Trust in the electoral system is maintained, the machines get new software for the next election, and life goes on.
In reality, the Open Source systems have paper trails. So when the affected machines are taken offline, the existing ballots are already in paper form and can be verified with the rest of the ballots.
So yes, Open Source is crucial for electronic voting systems. Widespread closed source voting systems would be an unmitigated disaster.
"Same with Adobe not making software for Microsoft because Microsoft sells competing software."
There's a key difference between those relations and the relationships between Microsoft and OEM's. Here's your example in the context of Microsoft/OEM relationships:
Let's pretend that Microsoft makes a plugin for Adobe Photoshop. Let's also pretend that Adobe Photoshop is largely useless (from a sales perspective) without this plugin, and Adobe cannot sell copies of Photoshop without it. The profit margin on the sale of a copy of Photoshop is so small as to be practically zero, but Microsoft pays Adobe for each copy of Photoshop that is sold with the plugin. This payment is enough to keep Adobe operating profitably.
Suddenly, Microsoft is selling its own version of Photoshop with the plugin. Adobe is also no longer allowed to sell Photoshop with the Microsoft plugin (Microsoft pulls the license out from under Adobe), so Microsoft becomes the only company providing Photoshop with the plugin. People stop buying Photoshop from Adobe, because the Microsoft version is cheaper, and Adobe goes bankrupt.
This is the kind of relationship that Microsoft has with OEM's, so let's go back to the real world: If Microsoft begins selling full PC's in India, and the campaign is successful, it becomes quite attractive to try the same tactic in Dell's markets. Don't believe for a second that Microsoft isn't doing this as a trial balloon for traditional markets. If Microsoft stops paying OEMs to include Windows, there's going to be a lot of pain. If Microsoft made every OEM go cold turkey at the same time, it wouldn't hurt as much since everyone would scale down by the same amount. But I suspect that Microsoft would attack OEM's piecemeal in a divide and conquer strategy.
I'm sure Dell sees the possibilities also, which may be why the company decided to try creating a new market populated by penguins and devoid of Microsoft.
This certainly isn't a general solution, but my spam problem is gone. I run my own email server off of my DSL (I have a business class DSL contract with a static IP address -- which is only $10/month more than a dynamic IP address contract which forbids running servers). I have a different email address provided to each company I do business with, and a small pool of email addresses I give to new people who express a desire to email me (with a permanent alias created once their intent is verified), all routed via my system's email alias file to my real email address which is never disclosed to anyone.
If I start getting spam on one of the aliases, I know who's responsible for providing the address. I can simply disable the alias and create a new one. While it seems like extra work, it's relatively simple once the routine is established.
When I setup my employer's online payment system, I went to VeriSign's site (which I opposed, but I was overruled) and signed up for a certificate. I got a call from VeriSign that went something like this:
VeriSign: Are you the person who requested a certificate?
Me: Yes.
VeriSign: Send us the money and we'll email your certificate to you.
The certificate arrived in unencrypted email! I kid you not.
That was hardly inspiring security. Considering that anyone who was eaves dropping on our unencrypted email traffic could have intercepted our certificate at the source, identity verification via CA's is next to worthless. Certificate Authorities are just con artists preying on people's fears.
The only thing a certificate is actually good for is to ensure that traffic between Point A and Point B is being encrypted sufficiently well to be meaningless to anyone who intercepts it. For that, a VeriSign certificate is no better than a self-signed certificate. I would prefer a self-signed certificate, because then at least it isn't susceptible to interception before implementation.
I would define it as a global Interlinked network of computers, but I'm no rocket scientist....