Well, it is the journalist's duty to learn about science. Aren't journalists embarassed when they report a hoax or scam as being the truth? Don't they consider it their job to get to the bottom of such things? If they don't know anything about science, how can they evaluate whether or a not some unscrupulous scientist is trying to use them as a tool?
It is caused by the Flying Spaghetti Monster as punishment for the fact that too few of us wear pirate regalia in his honor. The evidence on this is very clear.
He states his hypothesis in the beginning (like any good paper), then shows a number of examples that support his hypothesis in the article. But if you bother to read to the very end, he gives some counter-examples that significantly detract from his hypothesis, and he puts forward a new one that he thinks does a better job of explaining the data.
The conspiracy is a very tempting explanation. I feel much the same way he does about the humanities after observing them from the outside. But I have to agree with him that the real reason is likely much more prosaic.
*chuckle* Yeah, that logic doesn't really hold. Though it's a big reason I don't use BSD. In truth, if I needed a full-featured database for something serious, I'd consider Postgress. Though I wouldn't consider BSD because I have philosophical issues with the very core of it.
Right now, MySQL seems to be really easy to set up and use for a random purpose, though I think SQLlite is even better. So if the database and its features weren't actually that important, I'd probably ignore Postgress because there's a lot less out there on it, and some idiot in Slashdot decided to post a marketing blurb as a comment.
You don't actually have to have every single filesystem out there support it. You just need the APIs in the kernel so that if a filesystem does support it, userspace programs can make calls that use it. Non-ACID supporting filesystems will fall back to some behavior that isn't actually ACID and is easy to implement. There would be sysconfig calls you could use on a filesystem to discover exactly how good its ACID support was.
And the code that really _does_ implement ACID doesn't have to live in the kernel either. The kernel can get to it through things like FUSE.
The addition of an API for good support of transactions would make distributed filesystems much easier to write.
I realize that SQL isn't trivial. I think of SQL as a really advanced and sophisticated version of the 'find' utility. And I realize that foreign key->primary key relationships don't have an obvious mapping into the hierarchical database that is a filesystem. And that's OK. I would just be happy if the database structure mapped into the filesystem in some reasonable way, and you could manipulate the database through the Unix system call API. It can return errors if you try to do something that violates a relational integrity constrain or whatever. It's the separation of namespaces that I think is really evil.
Filesystem vs. database shouldn't have to be a design choice. And the bizarre two-headed world we have now where there's a whole ton of data that has no visibility to tools that aren't specially written to access it is just ridiculous.
You stop coming up with whole new APIs and practically new OSes just so you can have your little relational database world, and you work with the kernel to get the minimum basic support needed into the filesystem for ACID. The Unix/POSIX filesystem API is old and anemic and needs a revamp.
But, that's the point of a filesystem, so you can stop caring about disks and blocks and stuff and just care about the data you have stored and what name you called it. The division between 'database' and 'filesystem' is artificial, and it's bad for everybody to keep them separate.
Ahh, but you see, no MySQL partisan came into this forum and posted that garbage. I agree, it is even worse than the Postgress blurb. But the fact such literature exists is not a negative reflection on the database. The fact that someone who's an advocate of it chooses to post it in this forum is.
*chuckle* I don't really like either of them. I think relational databases in general are evil. The only thing good about them is SQL their support of transactions.
And don't tell me that's all they are. If that's all they are, then why can't I 'ls' a table, cat a row, or insert something by creating a file?
They are evil divisions of the global filesystem namespace into pieces that need special, unique methods to deal with and administrate them. It's a ridiculous amount of overhead just to get those features.
*rolls eyes* It is not 'information'. It is awful corporatese that tells me almost nothing. If you had bothered to understand it yourself and posted stuff about your personal experiences or understanding of how Postgress SQL is better than MySQL, I'd be all ears. But this garbage deserves a -1 spam rating.
This is a cut & paste of a piece of advertisement and marketing fluff. I don't care how good PostresSQL is, if it has people like you behind it I want no part of it.
First, the document passing. Second, basically working for google before he left Microsoft.
As for non-compete clauses, I think they're stupid and wrong. The only time I think they're at all justified (and the only time I know of that courts have held them up) is in the case of something like a news anchor or radio personality where the person themselves is arguably a part of a company's trademark. I do not think that applies to Fu-Lee here.
As for Google turning evil... Well, perhaps they will. I do not trust any large organization (corporate or government) to have ends or use means that are compatible with individual human freedom. But for now, Google's seem to (more or less). All that could change tomorrow though.
I note, with interest, that your site chooses to use IIS. I will not be trusting any of my personal data to it.
Well, Microsoft essentially is an organized crime syndicate. They were convicted in a court of law after all, and I haven't noticed any behavior and/or attitude changes since.
Well, I have no opinion about whether or not the findings are correct or not. I imagine that study will be nit-picked to death, and there may even be some horrible major flaw in it. But...
You're average female mathematician is most likely still much brighter than your average male restaraunt manager. Just because the humps on the bell curve for some statistic as measured over the entire population happen to lie in different places for men and women still doesn't say much for what an individual you happen to meet is capable of.
I think this is a bit overly harsh. I think it's reasonable for them to want to keep secrets until they're ready to be out with them, and I don't recall any outright denials or other falsehoods, just insisting that it was a rumor. That's skirting the truth, but it's not lying.
I do think the secret keeping thing has some limits, but I don't think they came anywhere close to them with this bit about IM services.
I would've refused to use it outside school regardless. I would've left it in my locker and refused to take it home. Any computer that's running software that I don't have control over might be doing anything at all.
I don't care what they're monitoring. The computer is an active agent of the school administration, and they have no power or authority outside school grounds, and they aren't going to use some stupid laptop they're requiring me to carry to extend it there.
Hi there, please take this camera home and make sure it remains pointed at you at all times. You will not be allowed to see or edit what it has taken pictures of. If you do not point the camera at yourself, we will give you a failing grade. Any attempt to tamper with the camera or attempt to edit what it sends back will be met with felony charges.
I think the only reasonable answer to this situation is to state the school purposes can't extend beyond the school grounds.
Sure, iBook's don't have cameras built into them, but they do have microphones. Who knows what the monitoring software monitors? Maybe the iBooks were special ones with GPS devices in them.
Really, being required to take someone else's active device around isn't at all reasonable.
IMHO, they should've been able to monitor the admin from the get-go. People in positions of power should always be under much more scrutiny than the scrutiny they're allowed to put others under.
Not that I think it was OK for them to break into the computers. I think they should simply have refused to take them home.
Are the passwords to prevent co-workers from impersonating eachother, or are they to prevent outsiders from breaking in? If it's the latter, then the last one's are doing a really good job, and even the second level ones are probably significantly adding to your security.
Whenever you think about a security system, you should have the threat model in mind instead of just deciding on some simple set of rules you blanketly apply to every situation.
Personally, I feel the school has no right whatsoever to tell them that they must use a laptop that has monitoring software on it, and in other ways isn't really theirs. If the computers were in the school, that's one thing, but they were laptops.
I think it's to most places benefit to put stuff under the GPL unless there's a specific reason not to. One worry in GPLing your code is a competitor commercializing your code in a way that you can't work with. The GPL is one of the better licenses for making sure that can't happen.
Electricity is 'free' in various places like coffee shops. You pay for the wireless with your coffee.
Having to pay 50 different service providers $20/mo so you have wireless access wherever it is that they have a hotspot is pretty ridiculous.
Going through some silly sign-on screen for wireless whenever I open up my laptop isn't reasonable. I should be able to just crack it open wherever I need data. It would be incredibly useful, for example, to be able to get some of the nice realtime traffic maps while on the freeway.
Those are basically the reasons. Wifi is just structured all wrong for a subscription system to work. Maybe if there were several different competing systems of providers that all blanketed a city with hotspots it would work. Then you could just pay one fee to one of the providers and have it automatically sign you in via IPSEC or something.
But the current system where I have to pay 5 different people a monthly fee so I can have coverage in the few places I sit down every month isn't reasonable.
I was noticing the error as well. I was disappointed that the article didn't say how much work (watts*time) could be extracted from a particular volume of urine. Though I would imagine that it might very well vary a lot.
If my keyboard isn't responding to input (not even Alt-Control-F1 to switch to a different virtual terminal) my mouse isn't moving, my screen isn't updating and I have no network connected, I fail to see how they're any different. Oh, you might have SETI@Home running on a work unit you downloaded when you did have a network connection, and it can continue going til it finishes even though you can't do anything else. Whee!
Yes, that happens to me very rarely on X, but it does happen. I used to even have whole system crashes or spontaneous resets even from buggy display drivers. But that hasn't been an issue now for 2-3 years.
Anyway, this is all moot. To my gf, who just wants to use the computer, she can't see much of a difference between having the X-server go funny so she has to shut it down and log back in, and having to reboot the computer. She seems more able than most to cause weird things like that to happen.
Slashdot Mirror
Well, it is the journalist's duty to learn about science. Aren't journalists embarassed when they report a hoax or scam as being the truth? Don't they consider it their job to get to the bottom of such things? If they don't know anything about science, how can they evaluate whether or a not some unscrupulous scientist is trying to use them as a tool?
It is caused by the Flying Spaghetti Monster as punishment for the fact that too few of us wear pirate regalia in his honor. The evidence on this is very clear.
He states his hypothesis in the beginning (like any good paper), then shows a number of examples that support his hypothesis in the article. But if you bother to read to the very end, he gives some counter-examples that significantly detract from his hypothesis, and he puts forward a new one that he thinks does a better job of explaining the data.
The conspiracy is a very tempting explanation. I feel much the same way he does about the humanities after observing them from the outside. But I have to agree with him that the real reason is likely much more prosaic.
*chuckle* Yeah, that logic doesn't really hold. Though it's a big reason I don't use BSD. In truth, if I needed a full-featured database for something serious, I'd consider Postgress. Though I wouldn't consider BSD because I have philosophical issues with the very core of it.
Right now, MySQL seems to be really easy to set up and use for a random purpose, though I think SQLlite is even better. So if the database and its features weren't actually that important, I'd probably ignore Postgress because there's a lot less out there on it, and some idiot in Slashdot decided to post a marketing blurb as a comment.
You don't actually have to have every single filesystem out there support it. You just need the APIs in the kernel so that if a filesystem does support it, userspace programs can make calls that use it. Non-ACID supporting filesystems will fall back to some behavior that isn't actually ACID and is easy to implement. There would be sysconfig calls you could use on a filesystem to discover exactly how good its ACID support was.
And the code that really _does_ implement ACID doesn't have to live in the kernel either. The kernel can get to it through things like FUSE.
The addition of an API for good support of transactions would make distributed filesystems much easier to write.
I realize that SQL isn't trivial. I think of SQL as a really advanced and sophisticated version of the 'find' utility. And I realize that foreign key->primary key relationships don't have an obvious mapping into the hierarchical database that is a filesystem. And that's OK. I would just be happy if the database structure mapped into the filesystem in some reasonable way, and you could manipulate the database through the Unix system call API. It can return errors if you try to do something that violates a relational integrity constrain or whatever. It's the separation of namespaces that I think is really evil.
Filesystem vs. database shouldn't have to be a design choice. And the bizarre two-headed world we have now where there's a whole ton of data that has no visibility to tools that aren't specially written to access it is just ridiculous.
You stop coming up with whole new APIs and practically new OSes just so you can have your little relational database world, and you work with the kernel to get the minimum basic support needed into the filesystem for ACID. The Unix/POSIX filesystem API is old and anemic and needs a revamp.
But, that's the point of a filesystem, so you can stop caring about disks and blocks and stuff and just care about the data you have stored and what name you called it. The division between 'database' and 'filesystem' is artificial, and it's bad for everybody to keep them separate.
Ahh, but you see, no MySQL partisan came into this forum and posted that garbage. I agree, it is even worse than the Postgress blurb. But the fact such literature exists is not a negative reflection on the database. The fact that someone who's an advocate of it chooses to post it in this forum is.
*chuckle* I don't really like either of them. I think relational databases in general are evil. The only thing good about them is SQL their support of transactions.
And don't tell me that's all they are. If that's all they are, then why can't I 'ls' a table, cat a row, or insert something by creating a file?
They are evil divisions of the global filesystem namespace into pieces that need special, unique methods to deal with and administrate them. It's a ridiculous amount of overhead just to get those features.
*rolls eyes* It is not 'information'. It is awful corporatese that tells me almost nothing. If you had bothered to understand it yourself and posted stuff about your personal experiences or understanding of how Postgress SQL is better than MySQL, I'd be all ears. But this garbage deserves a -1 spam rating.
This is a cut & paste of a piece of advertisement and marketing fluff. I don't care how good PostresSQL is, if it has people like you behind it I want no part of it.
Well, you have two interesting points...
First, the document passing. Second, basically working for google before he left Microsoft.
As for non-compete clauses, I think they're stupid and wrong. The only time I think they're at all justified (and the only time I know of that courts have held them up) is in the case of something like a news anchor or radio personality where the person themselves is arguably a part of a company's trademark. I do not think that applies to Fu-Lee here.
As for Google turning evil... Well, perhaps they will. I do not trust any large organization (corporate or government) to have ends or use means that are compatible with individual human freedom. But for now, Google's seem to (more or less). All that could change tomorrow though.
I note, with interest, that your site chooses to use IIS. I will not be trusting any of my personal data to it.
Well, Microsoft essentially is an organized crime syndicate. They were convicted in a court of law after all, and I haven't noticed any behavior and/or attitude changes since.
Well, I have no opinion about whether or not the findings are correct or not. I imagine that study will be nit-picked to death, and there may even be some horrible major flaw in it. But...
You're average female mathematician is most likely still much brighter than your average male restaraunt manager. Just because the humps on the bell curve for some statistic as measured over the entire population happen to lie in different places for men and women still doesn't say much for what an individual you happen to meet is capable of.
I think this is a bit overly harsh. I think it's reasonable for them to want to keep secrets until they're ready to be out with them, and I don't recall any outright denials or other falsehoods, just insisting that it was a rumor. That's skirting the truth, but it's not lying.
I do think the secret keeping thing has some limits, but I don't think they came anywhere close to them with this bit about IM services.
I would've refused to use it outside school regardless. I would've left it in my locker and refused to take it home. Any computer that's running software that I don't have control over might be doing anything at all.
I don't care what they're monitoring. The computer is an active agent of the school administration, and they have no power or authority outside school grounds, and they aren't going to use some stupid laptop they're requiring me to carry to extend it there.
Hi there, please take this camera home and make sure it remains pointed at you at all times. You will not be allowed to see or edit what it has taken pictures of. If you do not point the camera at yourself, we will give you a failing grade. Any attempt to tamper with the camera or attempt to edit what it sends back will be met with felony charges.
I think the only reasonable answer to this situation is to state the school purposes can't extend beyond the school grounds.
Sure, iBook's don't have cameras built into them, but they do have microphones. Who knows what the monitoring software monitors? Maybe the iBooks were special ones with GPS devices in them.
Really, being required to take someone else's active device around isn't at all reasonable.
IMHO, they should've been able to monitor the admin from the get-go. People in positions of power should always be under much more scrutiny than the scrutiny they're allowed to put others under.
Not that I think it was OK for them to break into the computers. I think they should simply have refused to take them home.
Are the passwords to prevent co-workers from impersonating eachother, or are they to prevent outsiders from breaking in? If it's the latter, then the last one's are doing a really good job, and even the second level ones are probably significantly adding to your security.
Whenever you think about a security system, you should have the threat model in mind instead of just deciding on some simple set of rules you blanketly apply to every situation.
Personally, I feel the school has no right whatsoever to tell them that they must use a laptop that has monitoring software on it, and in other ways isn't really theirs. If the computers were in the school, that's one thing, but they were laptops.
I think it's to most places benefit to put stuff under the GPL unless there's a specific reason not to. One worry in GPLing your code is a competitor commercializing your code in a way that you can't work with. The GPL is one of the better licenses for making sure that can't happen.
Here are some reasons for this:
Those are basically the reasons. Wifi is just structured all wrong for a subscription system to work. Maybe if there were several different competing systems of providers that all blanketed a city with hotspots it would work. Then you could just pay one fee to one of the providers and have it automatically sign you in via IPSEC or something.
But the current system where I have to pay 5 different people a monthly fee so I can have coverage in the few places I sit down every month isn't reasonable.
That link was such a helpful exposition for the concept you were trying to explain.
I was noticing the error as well. I was disappointed that the article didn't say how much work (watts*time) could be extracted from a particular volume of urine. Though I would imagine that it might very well vary a lot.
If my keyboard isn't responding to input (not even Alt-Control-F1 to switch to a different virtual terminal) my mouse isn't moving, my screen isn't updating and I have no network connected, I fail to see how they're any different. Oh, you might have SETI@Home running on a work unit you downloaded when you did have a network connection, and it can continue going til it finishes even though you can't do anything else. Whee!
Yes, that happens to me very rarely on X, but it does happen. I used to even have whole system crashes or spontaneous resets even from buggy display drivers. But that hasn't been an issue now for 2-3 years.
Anyway, this is all moot. To my gf, who just wants to use the computer, she can't see much of a difference between having the X-server go funny so she has to shut it down and log back in, and having to reboot the computer. She seems more able than most to cause weird things like that to happen.