you have trouble understanding concepts that are either new, or outside your comfort bubble. Basically, all flame-bait and no value.
End Of Line. PLEASE MODERATE.
With all due respect, you should at least consider the possibility that your argument could be more clear. Immediately ending a discussion because someone needs clarification is rude, and would seem to indicate an unwillingness to subject your point to debate.
you believe I am wrong, even if my assertions are true;
The argument I was trying to make outlined relatively clearly:
Second, the complaints you make do not show a problem with economics, even if they are true. You allege that people are making bad business decisions, but if your points are correct, neoclassical economic theory would agree that these decisions are bad.
However, unlike you, I'm willing to clarify. You argue that we need a new form of economics. Your argument is that currently business decisions are being made that are obviously bad, that these decisions are supported by economics, and therefore economics is bad. I am agreeing that at least some bad decisions probably are being made, but I can in any case just stipulate that all your points about bad decisions are true, because I am arguing that these bad decisions are not supported by economics. If they aren't, we don't need "Quantum Economics." This is all I am trying to argue.
My first claim, that your post had no relation to its parent, was just dumb. Sorry.
I see now that you interpreted the statement in the parent about 100% loss incorrectly, and attributed it to 19th-century economics. When that post said the "unsold bandwidth is a 100% loss," that didn't literally mean that it would be recorded as a loss on the balance sheet, any more than I could claim a $100,000 business loss on my taxes if I try to sell a banana for that amount, but it goes unsold and becomes spoiled and I throw it away. Now THAT would be voodoo economics:) All I lose is the cost of producing the banana. There is nothing in economics that says otherwise. However, you seem to think that there is, because you say that multiplying the bandwidth must multiply the loss (that is, you say that economics says that). By analogy, suppose I had a device that produced 10 bananas per minute ex nihilio. If I can upgrade the machine for $1000 to produce a million bananas per minute, I would do so as a rational economic agent as long I can recoup the $1000. The value of excess bananas that might spoil is of no concern. I believe the burden is on you to show why any economics, even from the 19th century, would have me do otherwise.
Yea -- same thing happened to me...the disk fried put the firewall kept routing packets for 2 months -- granted I was running a remote syslog -- if the syslog was local, I am sure it would have halted at some point.
You're mistaken. Even running a local syslog, a disk failure will not prevent the kernel from routing packets. I know this empirically:)
First, I don't understand how this relates to the post you replied to. That post basically argued that companies should try harder to sell unutilized bandwidth (perhaps by making it cheaper at off-peak times). Failing to sell something that someone is willing to buy is practically like losing money. I can't imagine why you would have a problem with companies selling off-peak bandwith at cheaper rates.
Second, the complaints you make do not show a problem with economics, even if they are true. You allege that people are making bad business decisions, but if your points are correct, neoclassical economic theory would agree that these decisions are bad.
Very unlike industrial age planes, trains, and automobiles, the capacity of the fiber is a VARIABLE... it grows or shrinks with different devices on the end points. So how much "bandwidth inventory" is there on a fiber? You CANNOT know, because there is so much more innovation to complete in this arena. You can only pretend to know, in order to cook up a spreadsheet that Proves Your Presupposition, by putting some device or another on the ends of the fibers and pretending like that is a constant that won't change in 18 months.
If I understand your point correctly, you claim that people aren't taking into account that the bandwidth of fiber will improve, and that said improvement will not require replacing the fiber. I don't understand why you think that economic theory requires you not to take that into account. That simply isn't the case. You'd have to be an idiot not to realize that technological improvement will allow you to get increasing bandwidth out of your investment.
Let's take it another step. Since we know that the capacity of fiber is only limited by the performance of the end devices and we know that devices will continue to improve into the foreseeable future, we must concede that the capacity of the fiber we put in the ground is at least 100X more than we are able to deliver today. That's a future 100X LOSS! Therefore, improving the Internet is the stupidest investment in the world. Is this really the way you want to model the future of communications? I don't think so. Let's not pretend like we are all beholden to some immutable universal law of capacity utilization. We have CHOICES on how to account for all these things. Nineteenth century economics may have kludged us through the twentieth century world of planes, trains, and automobiles, but we're in the quantum world now; and that is going to require some Quantum Economics. Just because you can't conceive of what that looks like, yet, doesn't mean it's any less true.
I'm afraid I have trouble understanding this paragraph. Specifically, I don't understand why you think that anyone would think that they shouldn't invest in technology because it might be improved.
What we need are MBA's who are as able and willing to innovate as technologists. Instead, as a general rule, they march into the lab with textbook models from P&G in hand and a set of nineteenth century rules etched into their neurons. The economics of communications will either be completely re-written over the next few years, or we truly risk be permanent AOL doom by John Perry Barlow's
Death From Above.
John Perry Barlow's article alleges that demand for upstream bandwidth is being underestimated because media executives are obsessed with control. It's probably true that said demand is underestimated, though less so now than in 1995 when he wrote that, but we don't need "Quantum Economics" to tell us that underestimating demand leads to poor business decisions.
How much power is an EULA actually allowed to provide? If I pay to participate in something, and during my participation I acquire some item of worth, what restraints are there on the overseeing entity telling me what I can and can't do with my acquisition?
Bear in mind that your items of worth are merely records in Mythic's servers. In the absence of any agreement to the contrary, they would have every right to alter their records to show that you have no gold. Since the only thing preventing them from doing this is the agreement you made when you started paying for it, if said agreement disallows the sale of in-game items, that is absolutely enforceable.
What usually gets people riled up about EULAs is when they prohibit something that would be allowed in the absence of an agreement to the contrary, such as reverse engineering software. You can complain that the EULA is void because you never agreed to it, and that you therefore have the right to reverse engineer some piece of software. If you complain to Mythic that you never agreed to their Terms of Use/EULA/whatever, they can simply delete your account. What are you going to do, sue them for breaching the contract you argue doesn't exist?
If you're keeping stuff on your hard drive that you worry about the FBI finding, wouldn't it be smarter just to use an encrypted partition or something?
Perhaps publicly funded code needs a modified GPL type license that is free to use (even to run a business) but incurs significant royalties if the code is incorporated into commercial software products. I wonder if RMS would be OK with that?
RMS has never objected to commercial software, but rather to proprietary software. If you replace the word "commercial" with "proprietary," I can't imagine that he'd be OK with it, since I'm not aware of him ever being OK with proprietary software.
Note: I just checked and noticed that Security Focus changed the page. Basically, they used to have a row in the table for "Linux (aggregate)". With the exception of 1998, if you added the numbers of bugs for each distribution, you got a number greater than in the aggregate row. This would seem to indicate that were removing duplicates.
However, I'm repeating myself, so I'll try a concrete example with small numbers. Suppose they listed RedHat as having 10 bugs and Debian as having 5 bugs, and those were the only distributions that they listed. If they then listed Linux (aggregate) as having 15 bugs, then it would be clear that they were double counting. On the other hand, if they listed Linux (aggregate) as having 12 bugs, then they clearly might have been removing duplicates from the total, so there is no evidence from their statistics alone that they are counting duplicates. The latter situation is analogous to what was on their site.
I was never intending to argue that the numbers were useful. For example, they probably were counting bugs in any package in a distribution as a Linux bug, whereas they probably would not have counted a bug in IIS or SQL Server as a Windows OS bug. However, I was getting annoyed at everyone claiming that they were double counting when there wasn't any evidence given to back up that claim.
Now, this is all moot, because they've removed the Linux (aggregate) statistic, so there's no way for them to be double counting.
What you say is that, of course, they do not include duplicates of the same vulnerability. But then there's no such program as rsync-2.07-3.i386.rpm on Debian 2.2 . Can you see it?
Look, if you look at the big table, "Number of OS Vulnerabilities by Year," in the SecurityFocus stats, you will find (except for 1998, where they seem to have made a mistake) that the Linux aggregate number they list is less than the sum of the numbers listed for each Linux distribution. This is why your claim that they are counting duplicate bugs is dubious.
Of course, you are probably correct that if there is an rsync bug they will count it in each distribution, but you can see that they do not simply add the numbers of all the distributions together to get the aggregate, so this is not a problem.
Actually, according
to this, it sounds like the problem is that the http server doesn't obey the other rules you set, i.e. to share or not share certain file types, bandwidth limits, number of upload limits, etc.
I would speculate that the people who posted that are idiots. The guy who had Zone Alarm Pro must have clicked OK when it asked if Morpheus could act as a server, and yet he is shocked that he is able to access it as a client, "penetrating" Zone Alarm Pro.
Since HTTP is the normal method used by Morpheus for file transfers, it is doubtful that the restrictions Morpheus imposes on outgoing connections do not apply just because you use a different HTTP client. I have in fact tested myself and determined that it does not share files you don't choose to share, that it respects limits on both bandwidth and number of uploads, and that such transfers are listed under Uploads on the Morpheus server like any other outgoing transfer.
In short, this is the normal method Morpheus uses for peer-to-peer connections, not an exploit.
Let's say, on assumption (and probably being conservative) that there's a billion lines of code in all of their current products (by MS standards, up to 3 years old). Lets say they have 1000 programmers (again being conservative) working on this initative. Each programmer puts in a record 10 hours a day, 50 hours a week, and doesn't take a lunch break. That means each one has to find 5000 bugs PER HOUR. 5000 buffer overflows, incorrect types, etc. which are hard enough to find when you're only looking for one of them.
You seem to be confusing bugs and lines of code. I seriously doubt that there are a billion bugs in Microsoft products. Also, your "conservative" estimate of 1000 programmers is low, if they are in fact stopping all other work, as they claim. They certainly employ more than 1000 programmers.
Actually, you were assUmeing that I was talking only about Outlook. I was also including messages from older email programs as well (whatever that abomination shipped with '95 was called). NO, it didn't mark the messages as quoted printable, it simply appended an '=' to the end of each line that didn't contain an explicit CR/LF for no good reason.
As far as assuming goes, you referred to MS by name. I didn't know they made a mail agent other than Outlook. Your experience sounded exactly like reading a multipart MIME format message in a non-MIME-compliant mail reader, so I assumed that's what you were talking about. I had no actual knowledge of which Windows mail agents were compliant. Sorry about that.
When I said "the whole point," I was referring to the point of this thread, which started when I responded to sjames whining about getting MIME messages from Outlook users. Errors is Outlook's ability to display certain standards-compliant e-mail doesn't mean that the e-mail it sends violates the standard.
So, you're saying that Outlook uses the quoted-printable encoding without labeling it as such? Digging through my old e-mail, I quickly found a couple random e-mails I received from Outlook Express 5.5 and 4.7 that used = for soft returns, and they were properly labeled as quoted-printable.
Also, MIME was meant more to allow sending attachments rather than to replace body text with an attachment.
Actually, multipart/alternative was meant to facilitate exactly the latter.
As I recall, many netizens found the whole thing to be quite rude, especially on mailing lists.
But my whole point was that all MS did was make a mail agent that implemented an RFC. Certainly, if clueless users send MIME format e-mails to mailing lists whose policy requires plain text, that is rude, but it isn't MS's responsibility to ensure that its users are clueful.
Re:This is the result of double counting...
on
EverQuest and the UN
·
· Score: 2, Insightful
What the hell? You can't measure California's economy because it's part of the US? If I tell you the GDP of North America, are we suddenly not allowed to discuss the GDP of the US? Your point is nonsensical. Whether the EQ economy is part of a larger economy has no bearing on whether it can be measured and discussed.
Thanks to MS, email has gone from a simple text message to a mixed bag of html, lines ending with an '=' and messages with unreadable attachments which repeat the body text (and are often larger than the message itself). Since asking politely hasn't done away with the crap, perhaps this will have to do.
Um, those would be examples of messages in MIME format. MIME is defined by RFC 1521. It predates Outlook, so son't blame MS if you aren't using a MIME-conformant mail reader.
IIRC algorithms used to be unpatentable. This changed when someone proved that a mechanical equivalent to something like an NAND gate could be built and therefore combined to make any digital circuit.
True, but if the Patent Office wanted to, it could request a working model <evil grin>. Oh well.
NP hard problems will not be affected. The TSP and all the optimization problems are NP hard. NP complete means the solution can be checked in polynomial time, which is not true for optimization problems.
You're simply wrong.
First, all NP-complete problems are NP-hard by definition. A problem is NP-hard if every problem in NP is reducible to it in polynomial time. A problem is NP-complete if it is NP-hard and it is in NP.
Second, TSP is a classic NP-complete problem. Formally, the problem is: given a weighted graph and a cost k, is there a Hamiltonian path with cost at most k? To verify the solution, you just take the path as the certificate, and verification is trivial: add up the cost of the path, check that it's not more than k, and check that the path is Hamiltonian.
Also factroring primes is not NP complete as there is no polynomial time algorithm yet to check if a number is prime.
(Insert obligatory comment about factoring primes here.)
Usually, the goal isn't to find the prime factorization of the input, but to find any factors at all. This is certainly true in cryptography, for example. In this case, verification is just multiplication, which is of course polynomial.
The point is, everything that is not NP-complete, and still computable that has been found by man to date, is NP-complete (that is, exponential; O(a^n) for some a).
Huh? Everthing we know of that isn't NP-complete is NP-complete, unless it is not computable? This barely manages not to be self contradictory by implying that everything we know of that isn't NP-complete isn't computable. But this is false. Also, NP-complete isn't the same thing as exponential.
Yay! I just discovered a new algorithm! Take the number to be factored and call it n. Then, for i=2 to n/2, see if i divides n. This is much faster than exponential time.
When choosing an encoding to map an abstract decision problem to a concrete problem, it is traditional (read: all sane people do this) to represent numbers using a place value system (such as binary) rather than something stupid like unary bit strings. Therefore, the size of the input is Theta(log(k)), if the input is a single number k, so a O(k) solution is O(b^n), where n is the size of the input and b is a constant.
You've got to love it when a clueless poster follows another clueless poster and tries to show how fucking great they are.
Slashdot Mirror
you have trouble understanding concepts that are either new, or outside your comfort bubble. Basically, all flame-bait and no value.
End Of Line. PLEASE MODERATE.
With all due respect, you should at least consider the possibility that your argument could be more clear. Immediately ending a discussion because someone needs clarification is rude, and would seem to indicate an unwillingness to subject your point to debate.
you believe I am wrong, even if my assertions are true;
The argument I was trying to make outlined relatively clearly:
Second, the complaints you make do not show a problem with economics, even if they are true. You allege that people are making bad business decisions, but if your points are correct, neoclassical economic theory would agree that these decisions are bad.
However, unlike you, I'm willing to clarify. You argue that we need a new form of economics. Your argument is that currently business decisions are being made that are obviously bad, that these decisions are supported by economics, and therefore economics is bad. I am agreeing that at least some bad decisions probably are being made, but I can in any case just stipulate that all your points about bad decisions are true, because I am arguing that these bad decisions are not supported by economics. If they aren't, we don't need "Quantum Economics." This is all I am trying to argue.
My first claim, that your post had no relation to its parent, was just dumb. Sorry.
I see now that you interpreted the statement in the parent about 100% loss incorrectly, and attributed it to 19th-century economics. When that post said the "unsold bandwidth is a 100% loss," that didn't literally mean that it would be recorded as a loss on the balance sheet, any more than I could claim a $100,000 business loss on my taxes if I try to sell a banana for that amount, but it goes unsold and becomes spoiled and I throw it away. Now THAT would be voodoo economics :) All I lose is the cost of producing the banana. There is nothing in economics that says otherwise. However, you seem to think that there is, because you say that multiplying the bandwidth must multiply the loss (that is, you say that economics says that). By analogy, suppose I had a device that produced 10 bananas per minute ex nihilio. If I can upgrade the machine for $1000 to produce a million bananas per minute, I would do so as a rational economic agent as long I can recoup the $1000. The value of excess bananas that might spoil is of no concern. I believe the burden is on you to show why any economics, even from the 19th century, would have me do otherwise.
I hope this accurately represented your argument.
Yea -- same thing happened to me...the disk fried put the firewall kept routing packets for 2 months -- granted I was running a remote syslog -- if the syslog was local, I am sure it would have halted at some point.
You're mistaken. Even running a local syslog, a disk failure will not prevent the kernel from routing packets. I know this empirically :)
First, I don't understand how this relates to the post you replied to. That post basically argued that companies should try harder to sell unutilized bandwidth (perhaps by making it cheaper at off-peak times). Failing to sell something that someone is willing to buy is practically like losing money. I can't imagine why you would have a problem with companies selling off-peak bandwith at cheaper rates.
Second, the complaints you make do not show a problem with economics, even if they are true. You allege that people are making bad business decisions, but if your points are correct, neoclassical economic theory would agree that these decisions are bad.
Very unlike industrial age planes, trains, and automobiles, the capacity of the fiber is a VARIABLE ... it grows or shrinks with different devices on the end points. So how much "bandwidth inventory" is there on a fiber? You CANNOT know, because there is so much more innovation to complete in this arena. You can only pretend to know, in order to cook up a spreadsheet that Proves Your Presupposition, by putting some device or another on the ends of the fibers and pretending like that is a constant that won't change in 18 months.
If I understand your point correctly, you claim that people aren't taking into account that the bandwidth of fiber will improve, and that said improvement will not require replacing the fiber. I don't understand why you think that economic theory requires you not to take that into account. That simply isn't the case. You'd have to be an idiot not to realize that technological improvement will allow you to get increasing bandwidth out of your investment.
Let's take it another step. Since we know that the capacity of fiber is only limited by the performance of the end devices and we know that devices will continue to improve into the foreseeable future, we must concede that the capacity of the fiber we put in the ground is at least 100X more than we are able to deliver today. That's a future 100X LOSS! Therefore, improving the Internet is the stupidest investment in the world. Is this really the way you want to model the future of communications? I don't think so. Let's not pretend like we are all beholden to some immutable universal law of capacity utilization. We have CHOICES on how to account for all these things. Nineteenth century economics may have kludged us through the twentieth century world of planes, trains, and automobiles, but we're in the quantum world now; and that is going to require some Quantum Economics. Just because you can't conceive of what that looks like, yet, doesn't mean it's any less true.
I'm afraid I have trouble understanding this paragraph. Specifically, I don't understand why you think that anyone would think that they shouldn't invest in technology because it might be improved.
What we need are MBA's who are as able and willing to innovate as technologists. Instead, as a general rule, they march into the lab with textbook models from P&G in hand and a set of nineteenth century rules etched into their neurons. The economics of communications will either be completely re-written over the next few years, or we truly risk be permanent AOL doom by John Perry Barlow's Death From Above.
John Perry Barlow's article alleges that demand for upstream bandwidth is being underestimated because media executives are obsessed with control. It's probably true that said demand is underestimated, though less so now than in 1995 when he wrote that, but we don't need "Quantum Economics" to tell us that underestimating demand leads to poor business decisions.
"Well" can be an adjective.
How much power is an EULA actually allowed to provide? If I pay to participate in something, and during my participation I acquire some item of worth, what restraints are there on the overseeing entity telling me what I can and can't do with my acquisition?
Bear in mind that your items of worth are merely records in Mythic's servers. In the absence of any agreement to the contrary, they would have every right to alter their records to show that you have no gold. Since the only thing preventing them from doing this is the agreement you made when you started paying for it, if said agreement disallows the sale of in-game items, that is absolutely enforceable.
What usually gets people riled up about EULAs is when they prohibit something that would be allowed in the absence of an agreement to the contrary, such as reverse engineering software. You can complain that the EULA is void because you never agreed to it, and that you therefore have the right to reverse engineer some piece of software. If you complain to Mythic that you never agreed to their Terms of Use/EULA/whatever, they can simply delete your account. What are you going to do, sue them for breaching the contract you argue doesn't exist?
If you're keeping stuff on your hard drive that you worry about the FBI finding, wouldn't it be smarter just to use an encrypted partition or something?
Perhaps publicly funded code needs a modified GPL type license that is free to use (even to run a business) but incurs significant royalties if the code is incorporated into commercial software products. I wonder if RMS would be OK with that?
RMS has never objected to commercial software, but rather to proprietary software. If you replace the word "commercial" with "proprietary," I can't imagine that he'd be OK with it, since I'm not aware of him ever being OK with proprietary software.
Note: I just checked and noticed that Security Focus changed the page. Basically, they used to have a row in the table for "Linux (aggregate)". With the exception of 1998, if you added the numbers of bugs for each distribution, you got a number greater than in the aggregate row. This would seem to indicate that were removing duplicates.
However, I'm repeating myself, so I'll try a concrete example with small numbers. Suppose they listed RedHat as having 10 bugs and Debian as having 5 bugs, and those were the only distributions that they listed. If they then listed Linux (aggregate) as having 15 bugs, then it would be clear that they were double counting. On the other hand, if they listed Linux (aggregate) as having 12 bugs, then they clearly might have been removing duplicates from the total, so there is no evidence from their statistics alone that they are counting duplicates. The latter situation is analogous to what was on their site.
I was never intending to argue that the numbers were useful. For example, they probably were counting bugs in any package in a distribution as a Linux bug, whereas they probably would not have counted a bug in IIS or SQL Server as a Windows OS bug. However, I was getting annoyed at everyone claiming that they were double counting when there wasn't any evidence given to back up that claim.
Now, this is all moot, because they've removed the Linux (aggregate) statistic, so there's no way for them to be double counting.
What you say is that, of course, they do not include duplicates of the same vulnerability. But then there's no such program as rsync-2.07-3.i386.rpm on Debian 2.2 . Can you see it?
Look, if you look at the big table, "Number of OS Vulnerabilities by Year," in the SecurityFocus stats, you will find (except for 1998, where they seem to have made a mistake) that the Linux aggregate number they list is less than the sum of the numbers listed for each Linux distribution. This is why your claim that they are counting duplicate bugs is dubious.
Of course, you are probably correct that if there is an rsync bug they will count it in each distribution, but you can see that they do not simply add the numbers of all the distributions together to get the aggregate, so this is not a problem.
1) If a package has a security issue, usually all distros announce the security bug. Thus, the bug gets counted multiple times.
You obviously didn't check the numbers to see if your claim made sense. See this post.
Actually, according to this, it sounds like the problem is that the http server doesn't obey the other rules you set, i.e. to share or not share certain file types, bandwidth limits, number of upload limits, etc.
I would speculate that the people who posted that are idiots. The guy who had Zone Alarm Pro must have clicked OK when it asked if Morpheus could act as a server, and yet he is shocked that he is able to access it as a client, "penetrating" Zone Alarm Pro.
Since HTTP is the normal method used by Morpheus for file transfers, it is doubtful that the restrictions Morpheus imposes on outgoing connections do not apply just because you use a different HTTP client. I have in fact tested myself and determined that it does not share files you don't choose to share, that it respects limits on both bandwidth and number of uploads, and that such transfers are listed under Uploads on the Morpheus server like any other outgoing transfer.
In short, this is the normal method Morpheus uses for peer-to-peer connections, not an exploit.
Err... isn't the same as Right Mouse Button -> Find from the same -> User built into Morpheous?
Yeah, but if you do it the slightly harder way, you're a 1337 hacker.
Let's say, on assumption (and probably being conservative) that there's a billion lines of code in all of their current products (by MS standards, up to 3 years old). Lets say they have 1000 programmers (again being conservative) working on this initative. Each programmer puts in a record 10 hours a day, 50 hours a week, and doesn't take a lunch break. That means each one has to find 5000 bugs PER HOUR. 5000 buffer overflows, incorrect types, etc. which are hard enough to find when you're only looking for one of them.
You seem to be confusing bugs and lines of code. I seriously doubt that there are a billion bugs in Microsoft products. Also, your "conservative" estimate of 1000 programmers is low, if they are in fact stopping all other work, as they claim. They certainly employ more than 1000 programmers.
Actually, you were assUmeing that I was talking only about Outlook. I was also including messages from older email programs as well (whatever that abomination shipped with '95 was called). NO, it didn't mark the messages as quoted printable, it simply appended an '=' to the end of each line that didn't contain an explicit CR/LF for no good reason.
As far as assuming goes, you referred to MS by name. I didn't know they made a mail agent other than Outlook. Your experience sounded exactly like reading a multipart MIME format message in a non-MIME-compliant mail reader, so I assumed that's what you were talking about. I had no actual knowledge of which Windows mail agents were compliant. Sorry about that.
When I said "the whole point," I was referring to the point of this thread, which started when I responded to sjames whining about getting MIME messages from Outlook users. Errors is Outlook's ability to display certain standards-compliant e-mail doesn't mean that the e-mail it sends violates the standard.
That doesn't explain the soft returns '='.
So, you're saying that Outlook uses the quoted-printable encoding without labeling it as such? Digging through my old e-mail, I quickly found a couple random e-mails I received from Outlook Express 5.5 and 4.7 that used = for soft returns, and they were properly labeled as quoted-printable.
Also, MIME was meant more to allow sending attachments rather than to replace body text with an attachment.
Actually, multipart/alternative was meant to facilitate exactly the latter.
As I recall, many netizens found the whole thing to be quite rude, especially on mailing lists.
But my whole point was that all MS did was make a mail agent that implemented an RFC. Certainly, if clueless users send MIME format e-mails to mailing lists whose policy requires plain text, that is rude, but it isn't MS's responsibility to ensure that its users are clueful.
What the hell? You can't measure California's economy because it's part of the US? If I tell you the GDP of North America, are we suddenly not allowed to discuss the GDP of the US? Your point is nonsensical. Whether the EQ economy is part of a larger economy has no bearing on whether it can be measured and discussed.
Thanks to MS, email has gone from a simple text message to a mixed bag of html, lines ending with an '=' and messages with unreadable attachments which repeat the body text (and are often larger than the message itself). Since asking politely hasn't done away with the crap, perhaps this will have to do.
Um, those would be examples of messages in MIME format. MIME is defined by RFC 1521. It predates Outlook, so son't blame MS if you aren't using a MIME-conformant mail reader.
IIRC algorithms used to be unpatentable. This changed when someone proved that a mechanical equivalent to something like an NAND gate could be built and therefore combined to make any digital circuit.
True, but if the Patent Office wanted to, it could request a working model <evil grin>. Oh well.
Er, the recipient's mail server will, of course, be an SMTP server. I meant that it doesn't have to connect to one other than that.
No, the mail program can just connect directly to the mail server for the recipient's domain. It doesn't need to connect to an SMTP server.
It's building 24, damn it!
NP hard problems will not be affected. The TSP and all the optimization problems are NP hard. NP complete means the solution can be checked in polynomial time, which is not true for optimization problems.
You're simply wrong.
First, all NP-complete problems are NP-hard by definition. A problem is NP-hard if every problem in NP is reducible to it in polynomial time. A problem is NP-complete if it is NP-hard and it is in NP.
Second, TSP is a classic NP-complete problem. Formally, the problem is: given a weighted graph and a cost k, is there a Hamiltonian path with cost at most k? To verify the solution, you just take the path as the certificate, and verification is trivial: add up the cost of the path, check that it's not more than k, and check that the path is Hamiltonian.
Also factroring primes is not NP complete as there is no polynomial time algorithm yet to check if a number is prime.
(Insert obligatory comment about factoring primes here.)
Usually, the goal isn't to find the prime factorization of the input, but to find any factors at all. This is certainly true in cryptography, for example. In this case, verification is just multiplication, which is of course polynomial.
The point is, everything that is not NP-complete, and still computable that has been found by man to date, is NP-complete (that is, exponential; O(a^n) for some a).
Huh? Everthing we know of that isn't NP-complete is NP-complete, unless it is not computable? This barely manages not to be self contradictory by implying that everything we know of that isn't NP-complete isn't computable. But this is false. Also, NP-complete isn't the same thing as exponential.
Yay! I just discovered a new algorithm! Take the number to be factored and call it n. Then, for i=2 to n/2, see if i divides n. This is much faster than exponential time.
When choosing an encoding to map an abstract decision problem to a concrete problem, it is traditional (read: all sane people do this) to represent numbers using a place value system (such as binary) rather than something stupid like unary bit strings. Therefore, the size of the input is Theta(log(k)), if the input is a single number k, so a O(k) solution is O(b^n), where n is the size of the input and b is a constant.
You've got to love it when a clueless poster follows another clueless poster and tries to show how fucking great they are.
You're right, it's quite amusing.