Thank you for clarifying my point. That's exactly what I was talking about.
I'm not strictly opposed to, say, a state-wide standard test, so that they can properly allocate (or un-allocate) funding. But the national level is not the place for any of that.
What's this "community" you're referring to? The people of the United States? The people in your state? The people in your town? The people in your school district?
I agree with you, but we need a sense of scale. The United States Federal government is way too involved in the schools. The states are too involved. Most school decisions should be made at the local level, and then some of course at the state level. No federal involvement should take place. No federal money, either.
The surest way to destroy any sense of community is to make a federal issue out of everything. If decisions are local, people will get involved, meet their neighbors, and encourage cooperation on policies that make sense to those people.
I say we should have more conservative policies at the larger scales of government and more liberal policies at the smaller scales of government. Think about it: liberal ideas work great for families and communities, not so great for the Soviet Bloc. I'm way oversimplifying, I know. Of course there are a lot of people that too closely associate with one label or the other that they don't even think that way.
It doesn't have to be just one person to take the risk. It doesn't have to be a hot-headed risk-taker that puts the whole company on the line.
Many of these risks are quite small. If OpenBSD fails, put windows back, and return to status quo, losing some money on the experiment but gaining knowledge. If OpenBSD succeeds, then you have a lifetime of rewards headed your way. If the chance of OpenBSD succeeding is high enough, it's worth the risk. That's not risky business, that's calculated risk.
And it doesn't have to be just one person. There's nothing that says to take a risk you can't take precautions. Warn another department, ask for input on the changes, ask about acceptable downtimes, test it first, and have a plan to mitigate any potential failures.
Businesses can fail because they take risks that they cannot afford to lose. Businesses also fail because they refuse to take good risks for which failure can be easily mitigated.
In other words, don't do either. The simple solution is to take good risks, and not take bad risks. If you're in an efficient business environment, the necessary incentives will be in place for both.
Thank you for the informative reply! I suspected that it could not be true when I heard it. Is there some grain of truth that the rumor started from, or was it purely false?
It actually sounds very similar to PostgreSQL's VACUUM, at least conceptually. In fact, both storage engines sound fairly similar, is there a major difference?
I heard that InnoDB had problems with dead tuples filling up a table when there are lots of updates or deletes. Is that true? Can someone explain? Is 5.0 different?
I heard that InnoDB builds up dead tuples with lots of inserts/updates, sort of like PostgreSQL without VACUUM. Is that accurate? Can someone explain? Do InnoDB tables just keep getting bigger? Is it fixed in 5.0?
Not only that, here's what you give up when you give up stored procedures/functions:
-triggers -user-defined types -user-defined aggregate functions -many types of constraints -functional indexes on user-defined functions (this one seems to get forgotten a lot. What do MySQL people do when they need one of these?)
Nice post, but I think everyone took my post WAY too seriously.
I was talking about a more general practice in argument where person A uses person B's words against him. That practice, of course, leads person B to use A's words against them. That can lead to some funny, futile arguments if you're a bystander.
It seemed like that's what was happening in the post I replied to. I have no idea what the article was saying, because at the time I posted the servers weren't responding (I tried to make a stupid joke about that, also).
I need to work on turning my jokes into print, apparently.
To use the words of the Information-wants-to-be-free movement: "It's not stolen! Mr. Kuhnt is not deprived of his work! It is Copyright infringement, not piracy or theft!"
That's the funny thing about arguing based on a double standard. Maybe the Dr. Dos guys (I can't read the article because it's slash-Dos-ed) were using the words of the anti-GPL folk by calling it "stolen".
GPL guy: "By your own words, these guys stole my code!" anti-GPL guy: "By your own words, it's just a minor copyright violation."
It's sort of like one of those dumb sentences like "This sentence is false." It doesn't matter what side of the argument you're on: you're wrong.
What is this 'anti-American' tag that some Americans try to apply to anyone that argues with them? Sounds pretty paranoid.
He said it. I merely said he came across that way, i.e. that he would most likely be percieved that way by someone who only read his one post.
I put "invented" in quote marks for a few reasons. First, it was a quotation. Second, it's not really clear to me that the internet really classifies as an invention. It's more like the Trans-Continental railroad or something; it's an accomplishment. Maybe someone invented TCP/IP, or the DNS protocol, but nobody really "invented" the internet. And of course you're right, the accomplishment was world-wide.
(1) I said I may be exaggerating. Obviously the internet wouldn't be completely useless. The point is, other countries need U.S. sites more than the U.S. needs any other single country's sites.
(2) If what you said was true, there would be no debate. The rest of the world would just not use U.S. DNS servers and routers. But they do.
My original post's point was that control of DNS has to do with economic power on the internet, not who "invented" the internet. But rather than notice my point, you latched onto an explicitly declared exaggeration (your quotation of my post certainly didn't include that "exaggeration" bit, did it Mr. Enlightened-pants?) in order to classify me as an "ignorant, arrogant American".
You certainly come across as anti-American, regardless of what you call yourself.
They could also launch a nuclear strike against the U.S., but that really has nothing to do with what we're talking about.
I think it was implied from my comment that they really had no internet-related recourse. For example, it would hurt Canada much more than the U.S. if Canada cut off the U.S. from their network.
Any reasonable person would have picked that up from my comment. It looks like you, on the other hand, were just looking for an opportunity to say that the rest of the world combined is more powerful than the U.S., which isn't all that insightful.
Having just read up on it a bit it appears decoupling my nations DNS system from the US system is implemented and can be done at the flip of a switch, so I apparently don't have to care what weird stuff the US government comes up with.
Exactly my point. The U.S. DNS servers and routers are an opt-in system. If the U.S. does anything too weird, it would lose it's authority in the blink of an eye. But in the meantime, we're not going to mess with our system and just turn it over to some other government, which may be less trustworthy.
Really, insisting on the DNS servers being US only is getting a bit laughable.
It's not laughable to suggest that U.S. DNS servers be controlled by the U.S. The rest of the world is connecting to U.S. DNS servers.
When/if China becomes more of an internet power, people will connect to their DNS servers. But obviously, that hasn't happened yet.
Oh, and IRC really has nothing to do with what we're talking about. We're talking about host names and host addresses. That means DNS and IP allocations. The U.S. set up DNS servers and routers. Other countries are now connecting to them. Now some of those other people (you) are saying that we should change the way our routers and dns servers behave.
It doesn't sound noble at all to me. It sounds like jealousy of the success of the U.S. internet and it's users.
Interesting plan. May have some security details to work out.
Also, who makes the list of root servers? Does each ISP just pick the top 50-100 that they trust? Obviously, you can't have one person with 10 entries in the list, and then an entire country with only one entry in the list. The servers would have to be somewhat representational. I'm assuming that an organization like ICANN would need to have a multiply factor next to it to give higher weight.
If you extend your idea to it's logical conclusion, each person on the planet could have their own registry, which would be representational, but pretty much defeat the purpose.
Seems almost like it could work, but I think there are a lot of things you'd need to sort out. And it doesn't really sound very deterministic. If you're a new company trying to get your name out, how do you make sure you're on all the "good" root servers? If you're a new root server, how do you get yourself on all the lists? If you get yourself on only a few of the "good" root servers, than someone can domain-squat you by getting on the rest and essentially DoSing your domain. Big companies would need to buy hundreds of domains for each name just to make sure that a worldwide audience could view their site.
To me, it sounds like the beginning of a good idea. But there's a lot of mess to sort out.
The other countries can pull the plug on the U.S. anytime they want. They can make their own internet, and even combine it with every internet in the world except the U.S. internet.
However, what you're asking is for the U.S. to change its internet, rather than you changing yours. You want this, of course, because the U.S. internet has billions of users and is a wildly successful standard.
If you can do better, create it, and let it compete for users in the open marketplace.
The problem is, of course, that every nation is susceptible to that kind of problem. There's no magical group of enlightened people that are immune from conflicts. And if there is, it's certainly not the U.N.
People sometimes think of the U.N. as a representative of the world, and therefore more impartial than the U.S. But many countries that are a part of the U.N. aren't even partially democratic. When you call them the "U.N." it sounds great. When you start naming the members, it doesn't sound so good after all. And it's certainly not representative of the people of the world.
You're talking about the moral justification, not the actual reason. If the E.U. or the U.N. wanted to, they could easily fabricate some moral justification for taking over the internet. But they would then run into the actual reasons that the U.S. has control:
(1) The U.S. has a unified language (2) The U.S. is an economic powerhouse, especially on the internet
The moral justification is orthogonal to the actual reason. It happens that they point at the same country this time.
It doesn't really have anything to do with who invented it. They can "reinvent" the internet any place they want, it's not like the U.S. has some global patent.
It has everything to do with economic power. Many people in the U.S. would hardly notice if other countries started dropping off the internet, except, perhaps, for a small decrease in spam. In any other country, the internet would basically be useless without seeing U.S. sites.
I may be somewhat exaggerating, but the basic idea is that the U.S. holds all the cards (for now at least), and the other countries don't really have any recourse.
(iii) that it was used in a manner that did not compromise it's accuracy.
When measuring physical quantities, there are lots of things that could throw the machine off. For instance, maybe when the manufacturer tests a speed laser, it works great. But maybe in the rain it gives very strange results. How would you ever be able to challenge it if you didn't have access to the code?
In a murder trial, the defense can cross examine the forensics experts to their heart's content. For instance, they could ask what method was used to determine a DNA match. The witness would never say "that's a secret".
But in routine traffic fine collection, where the money is a lot better, it would be too inconvenient to have to prove someone guilty, allow them a jury trial, or bother with any of that.
Except, in reality, MySQL falls down when it comes to separating logical relations from physical storage.
The storage layer you choose should never affect the semantics and behavior. If you choose MyISAM instead of InnoDB, transactions are no longer isolated or atomic.
Slashdot Mirror
Thank you for clarifying my point. That's exactly what I was talking about.
I'm not strictly opposed to, say, a state-wide standard test, so that they can properly allocate (or un-allocate) funding. But the national level is not the place for any of that.
What's this "community" you're referring to? The people of the United States? The people in your state? The people in your town? The people in your school district?
I agree with you, but we need a sense of scale. The United States Federal government is way too involved in the schools. The states are too involved. Most school decisions should be made at the local level, and then some of course at the state level. No federal involvement should take place. No federal money, either.
The surest way to destroy any sense of community is to make a federal issue out of everything. If decisions are local, people will get involved, meet their neighbors, and encourage cooperation on policies that make sense to those people.
I say we should have more conservative policies at the larger scales of government and more liberal policies at the smaller scales of government. Think about it: liberal ideas work great for families and communities, not so great for the Soviet Bloc. I'm way oversimplifying, I know. Of course there are a lot of people that too closely associate with one label or the other that they don't even think that way.
It doesn't have to be just one person to take the risk. It doesn't have to be a hot-headed risk-taker that puts the whole company on the line.
Many of these risks are quite small. If OpenBSD fails, put windows back, and return to status quo, losing some money on the experiment but gaining knowledge. If OpenBSD succeeds, then you have a lifetime of rewards headed your way. If the chance of OpenBSD succeeding is high enough, it's worth the risk. That's not risky business, that's calculated risk.
And it doesn't have to be just one person. There's nothing that says to take a risk you can't take precautions. Warn another department, ask for input on the changes, ask about acceptable downtimes, test it first, and have a plan to mitigate any potential failures.
Businesses can fail because they take risks that they cannot afford to lose. Businesses also fail because they refuse to take good risks for which failure can be easily mitigated.
In other words, don't do either. The simple solution is to take good risks, and not take bad risks. If you're in an efficient business environment, the necessary incentives will be in place for both.
Thank you for the informative reply! I suspected that it could not be true when I heard it. Is there some grain of truth that the rumor started from, or was it purely false?
It actually sounds very similar to PostgreSQL's VACUUM, at least conceptually. In fact, both storage engines sound fairly similar, is there a major difference?
I heard that InnoDB had problems with dead tuples filling up a table when there are lots of updates or deletes. Is that true? Can someone explain? Is 5.0 different?
Can someone please inform me about InnoDB?
I heard that InnoDB builds up dead tuples with lots of inserts/updates, sort of like PostgreSQL without VACUUM. Is that accurate? Can someone explain? Do InnoDB tables just keep getting bigger? Is it fixed in 5.0?
Not only that, here's what you give up when you give up stored procedures/functions:
-triggers
-user-defined types
-user-defined aggregate functions
-many types of constraints
-functional indexes on user-defined functions (this one seems to get forgotten a lot. What do MySQL people do when they need one of these?)
Nice post, but I think everyone took my post WAY too seriously.
I was talking about a more general practice in argument where person A uses person B's words against him. That practice, of course, leads person B to use A's words against them. That can lead to some funny, futile arguments if you're a bystander.
It seemed like that's what was happening in the post I replied to. I have no idea what the article was saying, because at the time I posted the servers weren't responding (I tried to make a stupid joke about that, also).
I need to work on turning my jokes into print, apparently.
To use the words of the Information-wants-to-be-free movement: "It's not stolen! Mr. Kuhnt is not deprived of his work! It is Copyright infringement, not piracy or theft!"
That's the funny thing about arguing based on a double standard. Maybe the Dr. Dos guys (I can't read the article because it's slash-Dos-ed) were using the words of the anti-GPL folk by calling it "stolen".
GPL guy: "By your own words, these guys stole my code!"
anti-GPL guy: "By your own words, it's just a minor copyright violation."
It's sort of like one of those dumb sentences like "This sentence is false." It doesn't matter what side of the argument you're on: you're wrong.
What is this 'anti-American' tag that some Americans try to apply to anyone that argues with them? Sounds pretty paranoid.
He said it. I merely said he came across that way, i.e. that he would most likely be percieved that way by someone who only read his one post.
I put "invented" in quote marks for a few reasons. First, it was a quotation. Second, it's not really clear to me that the internet really classifies as an invention. It's more like the Trans-Continental railroad or something; it's an accomplishment. Maybe someone invented TCP/IP, or the DNS protocol, but nobody really "invented" the internet. And of course you're right, the accomplishment was world-wide.
2 points:
(1) I said I may be exaggerating. Obviously the internet wouldn't be completely useless. The point is, other countries need U.S. sites more than the U.S. needs any other single country's sites.
(2) If what you said was true, there would be no debate. The rest of the world would just not use U.S. DNS servers and routers. But they do.
My original post's point was that control of DNS has to do with economic power on the internet, not who "invented" the internet. But rather than notice my point, you latched onto an explicitly declared exaggeration (your quotation of my post certainly didn't include that "exaggeration" bit, did it Mr. Enlightened-pants?) in order to classify me as an "ignorant, arrogant American".
You certainly come across as anti-American, regardless of what you call yourself.
They could also launch a nuclear strike against the U.S., but that really has nothing to do with what we're talking about.
I think it was implied from my comment that they really had no internet-related recourse. For example, it would hurt Canada much more than the U.S. if Canada cut off the U.S. from their network.
Any reasonable person would have picked that up from my comment. It looks like you, on the other hand, were just looking for an opportunity to say that the rest of the world combined is more powerful than the U.S., which isn't all that insightful.
Having just read up on it a bit it appears decoupling my nations DNS system from the US system is implemented and can be done at the flip of a switch, so I apparently don't have to care what weird stuff the US government comes up with.
Exactly my point. The U.S. DNS servers and routers are an opt-in system. If the U.S. does anything too weird, it would lose it's authority in the blink of an eye. But in the meantime, we're not going to mess with our system and just turn it over to some other government, which may be less trustworthy.
After all, the current system works pretty well.
You assume the U.N. is representative of the people it claims to govern.
Really, insisting on the DNS servers being US only is getting a bit laughable.
It's not laughable to suggest that U.S. DNS servers be controlled by the U.S. The rest of the world is connecting to U.S. DNS servers.
When/if China becomes more of an internet power, people will connect to their DNS servers. But obviously, that hasn't happened yet.
Oh, and IRC really has nothing to do with what we're talking about. We're talking about host names and host addresses. That means DNS and IP allocations. The U.S. set up DNS servers and routers. Other countries are now connecting to them. Now some of those other people (you) are saying that we should change the way our routers and dns servers behave.
It doesn't sound noble at all to me. It sounds like jealousy of the success of the U.S. internet and it's users.
Interesting plan. May have some security details to work out.
Also, who makes the list of root servers? Does each ISP just pick the top 50-100 that they trust? Obviously, you can't have one person with 10 entries in the list, and then an entire country with only one entry in the list. The servers would have to be somewhat representational. I'm assuming that an organization like ICANN would need to have a multiply factor next to it to give higher weight.
If you extend your idea to it's logical conclusion, each person on the planet could have their own registry, which would be representational, but pretty much defeat the purpose.
Seems almost like it could work, but I think there are a lot of things you'd need to sort out. And it doesn't really sound very deterministic. If you're a new company trying to get your name out, how do you make sure you're on all the "good" root servers? If you're a new root server, how do you get yourself on all the lists? If you get yourself on only a few of the "good" root servers, than someone can domain-squat you by getting on the rest and essentially DoSing your domain. Big companies would need to buy hundreds of domains for each name just to make sure that a worldwide audience could view their site.
To me, it sounds like the beginning of a good idea. But there's a lot of mess to sort out.
The other countries can pull the plug on the U.S. anytime they want. They can make their own internet, and even combine it with every internet in the world except the U.S. internet.
However, what you're asking is for the U.S. to change its internet, rather than you changing yours. You want this, of course, because the U.S. internet has billions of users and is a wildly successful standard.
If you can do better, create it, and let it compete for users in the open marketplace.
The problem is, of course, that every nation is susceptible to that kind of problem. There's no magical group of enlightened people that are immune from conflicts. And if there is, it's certainly not the U.N.
People sometimes think of the U.N. as a representative of the world, and therefore more impartial than the U.S. But many countries that are a part of the U.N. aren't even partially democratic. When you call them the "U.N." it sounds great. When you start naming the members, it doesn't sound so good after all. And it's certainly not representative of the people of the world.
I don't think it's in dispute that Iraq had WMDs at some time. The dispute is about whether there was a credible threat at the time we invaded.
You're talking about the moral justification, not the actual reason. If the E.U. or the U.N. wanted to, they could easily fabricate some moral justification for taking over the internet. But they would then run into the actual reasons that the U.S. has control:
(1) The U.S. has a unified language
(2) The U.S. is an economic powerhouse, especially on the internet
The moral justification is orthogonal to the actual reason. It happens that they point at the same country this time.
It doesn't really have anything to do with who invented it. They can "reinvent" the internet any place they want, it's not like the U.S. has some global patent.
It has everything to do with economic power. Many people in the U.S. would hardly notice if other countries started dropping off the internet, except, perhaps, for a small decrease in spam. In any other country, the internet would basically be useless without seeing U.S. sites.
I may be somewhat exaggerating, but the basic idea is that the U.S. holds all the cards (for now at least), and the other countries don't really have any recourse.
J2EE isn't picked because of the language, it's because it's got Sun and IBM (through Websphere) behind it.
Can someone do a quick explanation of what J2EE is and where it fits in with Servelets, Tomcat, JBoss, etc?
For web projects, I usually use PHP. However, I've been interested in another platform for a while, because I sometimes get annoyed by PHP.
How was the move to Java for you? Anything that bothered you or slowed you down? How about Tomcat, did that work well for the most part?
I happen to have a Tomcat book and a Struts book right next to me, so this is an interesting thread to read.
(iii) that it was used in a manner that did not compromise it's accuracy.
When measuring physical quantities, there are lots of things that could throw the machine off. For instance, maybe when the manufacturer tests a speed laser, it works great. But maybe in the rain it gives very strange results. How would you ever be able to challenge it if you didn't have access to the code?
In a murder trial, the defense can cross examine the forensics experts to their heart's content. For instance, they could ask what method was used to determine a DNA match. The witness would never say "that's a secret".
But in routine traffic fine collection, where the money is a lot better, it would be too inconvenient to have to prove someone guilty, allow them a jury trial, or bother with any of that.
Except, in reality, MySQL falls down when it comes to separating logical relations from physical storage.
The storage layer you choose should never affect the semantics and behavior. If you choose MyISAM instead of InnoDB, transactions are no longer isolated or atomic.