Of course, the author of that article went on to write CIPE, which is one of the problem protocols under discussion.
I use freeswan IPsec for securing wifi. The biggest problem with IPsec is that it suffers from "committee bloat" and is very complicated to use. Freeswan partially mitigates this complexity by implementing only a narrow subset of the RFCs (in fact, it is not even RFC compliant, because they deliberately removed some required features that might compromise security).
The good thing about IPsec, and freeswan in particular, is that they were openly developed with actual expert input and nobody has yet cast any doubt on the security of either.
Good ol standard SSL wrapping telnet is an unbeatable match.
The only available free-software SSL telnet implementations all use openssl, or its predecessor SSLeay (please correct me if I'm wrong; I would love to learn about other options). This SSL library has had numerous security updates in the past. I would hardly call this record unbeatable.
I use telnet over freeswan IPsec, and I like this combination very much, but no matter what you do, you have to be on your toes.
The build and dependency system is all inside the rpm program and associated libraries. Here are the source rpms for rpm. If you are worried about chicken and egg installation issues, an rpm tarball is available here.
How about their build and dependency... database?
The actual (complete) package database for redhat 9 is available in this little known gem of a package which is included in redhat but not installed by default (and IMO should be). The spec file for rebuilding the package database can be obtained from the corresponding source rpm, provided you have a copy of all of the redhat rpms for a particular version.
In general, almost everything in redhat includes source code. If you have to ask, the source is probably available. There are a few rare instances where redhat does not provide the source code, but these are pretty obscure and you have to know redhat fairly well to run into these programs -- so well that you wouldn't need to be asking in the first place!
I am talking only about the ISO format, of course. I think restricting attention to the ISO format is a very reasonable thing to do when evaluating manufacturers' claims of blank media data capacity, because these claims are based upon the assumption that you will be using the ISO format.
My point was that even in this routine context (namely, normal usage of the blank media employing the standard ISO formats), the advertised data capacities of blank optical disc media are:
Sometimes given in binary megabytes, sometimes in decimal megabytes
Never exactly accurate no matter which measurements you use, binary or decimal
My numbers are exact and they are correct. (Hint: don't argue with a math Ph.D about numbers. Contrary to popular wisdom, we can count.)
The factor of 2048/2352 in my calculations accounts exactly for the error correction in the ISO9660 mode 1 data CD format.
Without mode 1 data CD error correction, the capacity (for an 80 minute blank CD) is 80 minutes * 60 seconds/minute * 44100 audio samples / second * 2 bytes / sample * 2 audio channels, or 846720000 bytes. This is the raw capacity of the underlying audio storage. Note that the audio storage layer itself has some error correction, which I am disregarding since it is underneath even this storage layer and does not impact these numbers.
The ISO data CD standard mandates that this available space be divided into blocks of 2352 bytes of which 2048 bytes are to be used for data and 304 bytes are to be used for error correction (additional error correction, over and above the error correction already present in the audio storage layer, which I previously disregarded).
Multiplying 846720000 bytes by 2048/2352 yields 737280000 bytes of data capacity. Exactly.
Summary:
An 80 minute blank audio CD holds exactly 846720000 bytes of raw unheadered CD audio data.
The same CD used for data has exactly 737280000 bytes of data capacity when burned in the normal mode 1 data CD format.
4. For CD-R, DVD-R/w, the industry defines 1024 MB = 1 GB
No! CD-R uses binary prefixes and DVD-R uses decimal prefixes. Actually, in reality, both CD-R and DVD-R capacity labels are inaccurate under either the binary or the decimal interpretation, but you have to really be splitting hairs to notice.
The exact expected capacity of normal sized CD-Rs (not counting overburning, yadda yadda) is as follows:
For 74 minute CD-Rs, the capacity is 74*60*44100*2*2*2048/2352 = 681984000 bytes, or 650.390625 binary MiB (exactly, no roundoff error).
For 80 minute CD-Rs, the capacity is 80*60*44100*2*2*2048/2352 = 737280000 bytes, or 703.125 binary MiB (again, this figure is exact, not rounded off).
For DVD+/-R[W] media, the exact capacity is 4697620480 bytes, or just shy of 4.7 decimal GB. The capacity of a DVD-R is certainly nowhere near 4.7 binary GB.
Blaming the MS monopoly or other monopolies for that matter (in Sweden there is a sugar monopoly) on the state, is so dumb my eyes hurt.
There is nothing dumb at all about claiming the state causes the MS monopoly, because it's true. If you take away state-enforced copyright laws then the Microsoft monopoly would disappear in a heartbeat.
I am not suggesting to repeal copyright laws. I am merely pointing out that they enable the MS monopoly, and that they pretty clearly deviate from laissez faire capitalism.
For restaurants, at least, there are actually some very good reasons why you should tip the waiters when considered from a purely selfish point of view.
Eating out in the US is on average enormously cheaper than in any other first world country. Yeah, sure, there are some expensive restaurants, but in general the menu prices are far lower than in other countries with comparable standard of living, and the prices remain competitive even with the tip factored in.
The whole reason menu prices are so low in this country is because the tipping system works.
Now why does this help, you might ask? After all, don't you always "have" to pay the tip? Well, that's the thing: you don't always have to. For example, most restaurants accept take out orders, and you're not generally obliged to tip for take out orders. I'm not exactly suggesting that everyone rely on take out from now on, but it sure is nice to have the option.
It's like price discrimination, except that the customer gets to choose the price. We should be propping up this system, because it's very advantageous from our point of view to be able to choose our own price.
A private company introduces software that basically introduces built-in encryption for word documents, spreadsheets, and email.
You have it quite wrong. DRM is not encryption. It is amazing to me that people so often confuse the two.
Encryption is the art of securing a communication that both parties want secret. An example of encryption is the Pentagon-Kremlin hotline.
DRM is the art of securing a communication that only the sender wants secret. The whole point of DRM is that you are trying to keep the communication from leaking even in the face of an adversarial recipient.
The distinction is a really big deal! It's the whole reason why DRM is so difficult (and, to some, so objectionable).
Disclosure: I work for Microsoft, in the cryptography/anti-piracy/DRM group.
Make Lindows run as something other than root, ludicrous to run as root with Linux.
Let's please put this myth to rest. This is the third time this month that I have posted to correct this misimpression.
Lindows used to run everything as root, but current versions of Lindows don't run everything as root anymore. You have the option to add regular users during installation, and the installation encourages you to do so.
Just like in redhat.
Just like in debian.
I'm not advocating Lindows by any means (I don't even like their product), but I do think it is important to get the facts correct.
Why does 'intelligent' sound more sophisticated than 'smart'? Because it comes directly from french rather than Old English?
This phenomenon is not limited to English. Many other languages have the property that foreign imported words are more acceptable in polite company than native words.
For example, in Japanese, there are three major categories of words:
Native Japanese words, inherited from antiquity
Chinese words, imported roughly 1000 years ago
English words, imported since the 20th century and continuing to this day
In almost all cases the more recently imported words are more sophisticated than the older words. For example, the polite way to say restroom in Japanese is either "toire" (derived from the English word toilet) or "otearai" (imported from chinese, literally meaning "hand-wash"). There exist native Japanese words for restroom, but they connote dirtiness and one would never use them in polite company.
The three-level categorization of Japanese allows for more interesting observations than English's two level Latin/Germanic split. Note here that the most recent English import "toilet" can be used directly in polite speech, while the older Chinese import requires a euphemism and the original native words cannot be used at all. Compare this to native English, where "toilet" is one of the crudest possible ways to refer to a restroom. Familiarity breeds contempt, in any language.
But I feel in this day and age with all the acts of terrorisim people will give up a little bit of privacy to feel safer.
The problem with facial recognition is not privacy. The problem is not immature technology. The problem is mathematics. Facial recognition can never possibly work as a tool against terrorism, at least in the United States.
Let me explain the mathematics. Suppose for the sake of argument that 1 in 50000 people are terrorists (this ratio is absurdly high -- it implies 6000 terrorists in the US). Also suppose your facial recognition system is correct 99% of the time (again, absurdly high). You might think this system would be pretty good at catching terrorists. Well, you're wrong. This system would catch 500 innocent people for every terrorist it caught. The reason is that the mistake ratio (1 in 100) is much higher than the terrorist ratio (1 in 50000).
Lowering the numbers to more realistic values only makes the performance of the system worse. The conclusion, dictated by the mathematics, is inescapable: facial recognition technology would have to be absolutely perfect in order to be of use against terrorism.
That is a heck of an accusation. Where is your documentation?
When it comes to Japanese products, it is certainly not a big secret how to tell bootlegs apart from the legitimate products. This is especially the case for Japanese CDs which have a number of sure fire criteria any one of which suffices to make the distinction. Most of the information below is taken from the Pirate Anime FAQ.
Any CD that says "Made in Japan" or something to that effect on the back cover or the paper spine is certainly legit. Even for Japanese CDs, the line "Made in Japan" is almost always printed in English.
Anything by the Taiwanese labels Son May or Ever Anime is bootleg, because these companies have no licensed Japanese items in their portfolios.
If it says neither "Made in Japan" nor "Son May" or "Ever Anime", then there is a small chance that it is actually legitimate. However in my experience this has never actually been relevant to Tokyo Kid because every single one of their in-store CDs fits into one of the above two categories.
I moved out of the Boston area last month, but I can say with certainty that as of last month their in-store CD inventory consisted 99% of Son May and Ever Anime labels and only 1% of "Made in Japan" material.
I have never purchased anything from Tokyo Kid via mail order, so it is possible that they sell the bootleg CDs only in-store and not via mail order.
Back in 1996, I gave up games completely in order to switch to Linux.
Everything you say about games is correct, and none of it matters. Windows will always be the best gaming platform. There is nothing the Linux community can possibly to do change that fact. The power of numbers is just too much to overcome.
If someone values gaming too much to switch to linux, it's really not my problem.
From your tone it almost sounds as if you think Linux has to win over gamers in order to survive. Nothing could be further from the truth. Linux does not need a large userbase in order to thrive. All it needs is a small group of dedicated developers and the assurance that it will not be outlawed. Anything more than that is nice but not necessary.
Linux is not useful for gaming. Linux is not meant for gaming. I don't use Linux for gaming. Gaming is not the only thing in the world that computers are used for.
The mindset that a computer platform has to win market share or die is an artifact of the commercial software paradigm that has no relevance to open source software like Linux. With Linux, the users are the developers, and while new users are certainly welcome, there will always be certain markets like the gaming market where Linux serves no purpose and plays no role.
Tokyo Kid, unfortunately, sells bootleg taiwanese CDs almost exclusively. Everything else in the store is legit (except for a few of the cheap posters), but their CDs are about 99% bootleg.
I never buy bootleg CDs because when it comes down to that I'd rather just download the music illegally than buy illegal CDs.
Anything that restricts access to any data is DRM.
This is very very false, and comments like this one illustrate how ignorant most people are of the DRM concept.
There are at least two kinds of access restriction: traditional encryption, and DRM. The difference between the two is as follows. Traditional encryption restricts access to data that both the sender and the recipient want to keep secret.
DRM, on the other hand, restricts access to data that only the sender but not the recipient wants to keep secret.
People like you, who can't tell the two situations apart, are the reason why DRM is a failure today. You cannot use traditional encryption in a DRM setting, because the premises underlying the two models are very different.
There are ways to do DRM, but encryption is not one of them. Companies will not succeed at DRM until they realize this.
There are two major flaws with all of these statistics; one of them is hinted at in the cdrinfo article, but the other is not.
The first is that if you take a weighted average by player market share instead of just a straight numerical average then the numbers lean much more heavily in favor of DVD-R. In fact under such a weighted average I would expect even DVD-RW to beat out DVD+R.
The second flaw is that averaging across disc brands is completely irrelevant for most users' actual usage patterns. Most users do not choose a random disc brand for each burned disc that they make. On the contrary, most users (those that distribute discs, anyway) pick a single brand and stick with it for some period of time.
The question therefore should be "Which disc brand has the highest compatibility?" instead of "Which format has the highest compatibility averaged across all disc brands". And here DVD-R wins out as well: a top DVD-R disc brand (such as the Pioneer 4x media) has almost universal compatibility, which no single brand of DVD+R disc can match.
Please, someone show us where these drives are tested under Linux!!!
I do all my DVD burning in Linux (in fact I have never burned a disc in Windows), and I cannot see how the system OS matters at all for compatibility testing.
The mmc optical writing command set has been standardized since forever. A drive running in a Linux system sees the exact same data coming down the wire as the same drive running in Windows. From personal experience I can attest that my own observations match up exactly with the results from cdrinfo: DVD-R works most widely, followed by DVD-RW and DVD+R, with DVD+RW last.
I mean, how, exactly, is giving me a clickable list of browser pages at the top of the screen any better than giving me a clickable list of browser pages at the bottom of the screen
Because multiple levels of tabbing are actually useful in most situations.
Most people do not just run the web browser and nothing else. Most people run multiple programs like word processors, spreadsheets, IM clients, IRC, remote login windows, and other stuff at the same time as they run the web browser.
When you are running lots of programs at once, a row of browser tabs at the top of the screen is vastly more efficient than twenty browser tabs mixed in the taskbar interspersed with other unrelated applications.
Even if you only use the web browser and nothing else, there are still at least two major advantages of application level tabbing.
The first is that you can group related sites into one window. For instance, when I am simultaneously browsing through slashdot and nytimes and TV listings and weather reports, I can open one window for each group of pages and use tabs in each window to get multiple pages within each group. This two-level organization is impossible with a single taskbar.
The second reason is sheer numbers: at this moment I have open on my desktop right now 8 slashdot pages, 7 nytimes pages, four TV channel listings, and three weather pages (satellite photo, radar, forecast), for a total of 22 pages. This number of pages is very easy to manage with four windows and a bunch of tabs, but very difficult to manage with a single OS-level taskbar.
after searching and contacting vendors, i finally found a slot-1, 800mhz coppermine(100 fsb).
Even if you were dead set on upgrading only the CPU, you still could have gotten a better deal at powerleap. They sell 1.2ghz/100mhz slot-1 for $129.
You do have to check your motherboard for compatibility on their list, but most motherboards will work fine with it.
And no, I'm not getting paid to say this--in fact I actually dislike how powerleap uses patents to lock out competitors from the slot-1 upgrade market. But as far as their product goes I've been using the 1.2ghz upgrade for over a year now and it has worked absolutely flawlessly.
Slashdot Mirror
Of course, the author of that article went on to write CIPE, which is one of the problem protocols under discussion.
I use freeswan IPsec for securing wifi. The biggest problem with IPsec is that it suffers from "committee bloat" and is very complicated to use. Freeswan partially mitigates this complexity by implementing only a narrow subset of the RFCs (in fact, it is not even RFC compliant, because they deliberately removed some required features that might compromise security).
The good thing about IPsec, and freeswan in particular, is that they were openly developed with actual expert input and nobody has yet cast any doubt on the security of either.
One.
Sure, it compares favorably with openssh, but it does not even come close to my standard of not having "any weekness [sic]".
The only available free-software SSL telnet implementations all use openssl, or its predecessor SSLeay (please correct me if I'm wrong; I would love to learn about other options). This SSL library has had numerous security updates in the past. I would hardly call this record unbeatable.
I use telnet over freeswan IPsec, and I like this combination very much, but no matter what you do, you have to be on your toes.
Yes: anaconda source rpms
How about their build and dependency system ... ?
The build and dependency system is all inside the rpm program and associated libraries. Here are the source rpms for rpm. If you are worried about chicken and egg installation issues, an rpm tarball is available here.
How about their build and dependency ... database?
The actual (complete) package database for redhat 9 is available in this little known gem of a package which is included in redhat but not installed by default (and IMO should be). The spec file for rebuilding the package database can be obtained from the corresponding source rpm, provided you have a copy of all of the redhat rpms for a particular version.
In general, almost everything in redhat includes source code. If you have to ask, the source is probably available. There are a few rare instances where redhat does not provide the source code, but these are pretty obscure and you have to know redhat fairly well to run into these programs -- so well that you wouldn't need to be asking in the first place!
My point was that even in this routine context (namely, normal usage of the blank media employing the standard ISO formats), the advertised data capacities of blank optical disc media are:
The factor of 2048/2352 in my calculations accounts exactly for the error correction in the ISO9660 mode 1 data CD format.
Without mode 1 data CD error correction, the capacity (for an 80 minute blank CD) is 80 minutes * 60 seconds/minute * 44100 audio samples / second * 2 bytes / sample * 2 audio channels, or 846720000 bytes. This is the raw capacity of the underlying audio storage. Note that the audio storage layer itself has some error correction, which I am disregarding since it is underneath even this storage layer and does not impact these numbers.
The ISO data CD standard mandates that this available space be divided into blocks of 2352 bytes of which 2048 bytes are to be used for data and 304 bytes are to be used for error correction (additional error correction, over and above the error correction already present in the audio storage layer, which I previously disregarded).
Multiplying 846720000 bytes by 2048/2352 yields 737280000 bytes of data capacity. Exactly.
Summary:
No! CD-R uses binary prefixes and DVD-R uses decimal prefixes. Actually, in reality, both CD-R and DVD-R capacity labels are inaccurate under either the binary or the decimal interpretation, but you have to really be splitting hairs to notice.
The exact expected capacity of normal sized CD-Rs (not counting overburning, yadda yadda) is as follows:
- For 74 minute CD-Rs, the capacity is 74*60*44100*2*2*2048/2352 = 681984000 bytes, or 650.390625 binary MiB (exactly, no roundoff error).
- For 80 minute CD-Rs, the capacity is 80*60*44100*2*2*2048/2352 = 737280000 bytes, or 703.125 binary MiB (again, this figure is exact, not rounded off).
For DVD+/-R[W] media, the exact capacity is 4697620480 bytes, or just shy of 4.7 decimal GB. The capacity of a DVD-R is certainly nowhere near 4.7 binary GB.There is nothing dumb at all about claiming the state causes the MS monopoly, because it's true. If you take away state-enforced copyright laws then the Microsoft monopoly would disappear in a heartbeat.
I am not suggesting to repeal copyright laws. I am merely pointing out that they enable the MS monopoly, and that they pretty clearly deviate from laissez faire capitalism.
Eating out in the US is on average enormously cheaper than in any other first world country. Yeah, sure, there are some expensive restaurants, but in general the menu prices are far lower than in other countries with comparable standard of living, and the prices remain competitive even with the tip factored in.
The whole reason menu prices are so low in this country is because the tipping system works.
Now why does this help, you might ask? After all, don't you always "have" to pay the tip? Well, that's the thing: you don't always have to. For example, most restaurants accept take out orders, and you're not generally obliged to tip for take out orders. I'm not exactly suggesting that everyone rely on take out from now on, but it sure is nice to have the option.
It's like price discrimination, except that the customer gets to choose the price. We should be propping up this system, because it's very advantageous from our point of view to be able to choose our own price.
You have it quite wrong. DRM is not encryption. It is amazing to me that people so often confuse the two.
Encryption is the art of securing a communication that both parties want secret. An example of encryption is the Pentagon-Kremlin hotline.
DRM is the art of securing a communication that only the sender wants secret. The whole point of DRM is that you are trying to keep the communication from leaking even in the face of an adversarial recipient.
The distinction is a really big deal! It's the whole reason why DRM is so difficult (and, to some, so objectionable).
Disclosure: I work for Microsoft, in the cryptography/anti-piracy/DRM group.
Tell me where in Windows XP I can find:
- C compiler
- Email server
- Office suite
- SQL database server
- C++ IDE
These are all included in redhat, but not in Windows XP.Your math is a bit off here...
You have: 100GB/day
You want: megabits/s
* 9.2592593
It would take about six T1 lines (1.5 megabits per second each) to handle this level of traffic.
Let's please put this myth to rest. This is the third time this month that I have posted to correct this misimpression.
Lindows used to run everything as root, but current versions of Lindows don't run everything as root anymore. You have the option to add regular users during installation, and the installation encourages you to do so.
Just like in redhat.
Just like in debian.
I'm not advocating Lindows by any means (I don't even like their product), but I do think it is important to get the facts correct.
Well said, but let's put this little myth to rest: Lindows doesn't run everything as root by default anymore.
This phenomenon is not limited to English. Many other languages have the property that foreign imported words are more acceptable in polite company than native words.
For example, in Japanese, there are three major categories of words:
- Native Japanese words, inherited from antiquity
- Chinese words, imported roughly 1000 years ago
- English words, imported since the 20th century and continuing to this day
In almost all cases the more recently imported words are more sophisticated than the older words. For example, the polite way to say restroom in Japanese is either "toire" (derived from the English word toilet) or "otearai" (imported from chinese, literally meaning "hand-wash"). There exist native Japanese words for restroom, but they connote dirtiness and one would never use them in polite company.The three-level categorization of Japanese allows for more interesting observations than English's two level Latin/Germanic split. Note here that the most recent English import "toilet" can be used directly in polite speech, while the older Chinese import requires a euphemism and the original native words cannot be used at all. Compare this to native English, where "toilet" is one of the crudest possible ways to refer to a restroom. Familiarity breeds contempt, in any language.
The problem with facial recognition is not privacy. The problem is not immature technology. The problem is mathematics. Facial recognition can never possibly work as a tool against terrorism, at least in the United States.
Let me explain the mathematics. Suppose for the sake of argument that 1 in 50000 people are terrorists (this ratio is absurdly high -- it implies 6000 terrorists in the US). Also suppose your facial recognition system is correct 99% of the time (again, absurdly high). You might think this system would be pretty good at catching terrorists. Well, you're wrong. This system would catch 500 innocent people for every terrorist it caught. The reason is that the mistake ratio (1 in 100) is much higher than the terrorist ratio (1 in 50000).
Lowering the numbers to more realistic values only makes the performance of the system worse. The conclusion, dictated by the mathematics, is inescapable: facial recognition technology would have to be absolutely perfect in order to be of use against terrorism.
Recent versions of Lindows no longer run everything as root.
When it comes to Japanese products, it is certainly not a big secret how to tell bootlegs apart from the legitimate products. This is especially the case for Japanese CDs which have a number of sure fire criteria any one of which suffices to make the distinction. Most of the information below is taken from the Pirate Anime FAQ.
Any CD that says "Made in Japan" or something to that effect on the back cover or the paper spine is certainly legit. Even for Japanese CDs, the line "Made in Japan" is almost always printed in English.
Anything by the Taiwanese labels Son May or Ever Anime is bootleg, because these companies have no licensed Japanese items in their portfolios.
If it says neither "Made in Japan" nor "Son May" or "Ever Anime", then there is a small chance that it is actually legitimate. However in my experience this has never actually been relevant to Tokyo Kid because every single one of their in-store CDs fits into one of the above two categories.
I moved out of the Boston area last month, but I can say with certainty that as of last month their in-store CD inventory consisted 99% of Son May and Ever Anime labels and only 1% of "Made in Japan" material.
I have never purchased anything from Tokyo Kid via mail order, so it is possible that they sell the bootleg CDs only in-store and not via mail order.
Everything you say about games is correct, and none of it matters. Windows will always be the best gaming platform. There is nothing the Linux community can possibly to do change that fact. The power of numbers is just too much to overcome.
If someone values gaming too much to switch to linux, it's really not my problem.
From your tone it almost sounds as if you think Linux has to win over gamers in order to survive. Nothing could be further from the truth. Linux does not need a large userbase in order to thrive. All it needs is a small group of dedicated developers and the assurance that it will not be outlawed. Anything more than that is nice but not necessary.
Linux is not useful for gaming. Linux is not meant for gaming. I don't use Linux for gaming. Gaming is not the only thing in the world that computers are used for.
The mindset that a computer platform has to win market share or die is an artifact of the commercial software paradigm that has no relevance to open source software like Linux. With Linux, the users are the developers, and while new users are certainly welcome, there will always be certain markets like the gaming market where Linux serves no purpose and plays no role.
I never buy bootleg CDs because when it comes down to that I'd rather just download the music illegally than buy illegal CDs.
This is very very false, and comments like this one illustrate how ignorant most people are of the DRM concept.
There are at least two kinds of access restriction: traditional encryption, and DRM. The difference between the two is as follows. Traditional encryption restricts access to data that both the sender and the recipient want to keep secret. DRM, on the other hand, restricts access to data that only the sender but not the recipient wants to keep secret.
People like you, who can't tell the two situations apart, are the reason why DRM is a failure today. You cannot use traditional encryption in a DRM setting, because the premises underlying the two models are very different.
There are ways to do DRM, but encryption is not one of them. Companies will not succeed at DRM until they realize this.
Not all access restriction is DRM.
The first is that if you take a weighted average by player market share instead of just a straight numerical average then the numbers lean much more heavily in favor of DVD-R. In fact under such a weighted average I would expect even DVD-RW to beat out DVD+R.
The second flaw is that averaging across disc brands is completely irrelevant for most users' actual usage patterns. Most users do not choose a random disc brand for each burned disc that they make. On the contrary, most users (those that distribute discs, anyway) pick a single brand and stick with it for some period of time.
The question therefore should be "Which disc brand has the highest compatibility?" instead of "Which format has the highest compatibility averaged across all disc brands". And here DVD-R wins out as well: a top DVD-R disc brand (such as the Pioneer 4x media) has almost universal compatibility, which no single brand of DVD+R disc can match.
I do all my DVD burning in Linux (in fact I have never burned a disc in Windows), and I cannot see how the system OS matters at all for compatibility testing.
The mmc optical writing command set has been standardized since forever. A drive running in a Linux system sees the exact same data coming down the wire as the same drive running in Windows. From personal experience I can attest that my own observations match up exactly with the results from cdrinfo: DVD-R works most widely, followed by DVD-RW and DVD+R, with DVD+RW last.
Because multiple levels of tabbing are actually useful in most situations.
Most people do not just run the web browser and nothing else. Most people run multiple programs like word processors, spreadsheets, IM clients, IRC, remote login windows, and other stuff at the same time as they run the web browser.
When you are running lots of programs at once, a row of browser tabs at the top of the screen is vastly more efficient than twenty browser tabs mixed in the taskbar interspersed with other unrelated applications.
Even if you only use the web browser and nothing else, there are still at least two major advantages of application level tabbing.
The first is that you can group related sites into one window. For instance, when I am simultaneously browsing through slashdot and nytimes and TV listings and weather reports, I can open one window for each group of pages and use tabs in each window to get multiple pages within each group. This two-level organization is impossible with a single taskbar.
The second reason is sheer numbers: at this moment I have open on my desktop right now 8 slashdot pages, 7 nytimes pages, four TV channel listings, and three weather pages (satellite photo, radar, forecast), for a total of 22 pages. This number of pages is very easy to manage with four windows and a bunch of tabs, but very difficult to manage with a single OS-level taskbar.
Even if you were dead set on upgrading only the CPU, you still could have gotten a better deal at powerleap. They sell 1.2ghz/100mhz slot-1 for $129.
You do have to check your motherboard for compatibility on their list, but most motherboards will work fine with it.
And no, I'm not getting paid to say this--in fact I actually dislike how powerleap uses patents to lock out competitors from the slot-1 upgrade market. But as far as their product goes I've been using the 1.2ghz upgrade for over a year now and it has worked absolutely flawlessly.