Your acquaintances from Oak Ridge could redact the articles on the Manhattan Project, and within a year their edits would be superseded by layers of entropic cruft.
Wikipedia is designed to encapsulate the error rate nascent in human conversation.
And worse, because its control structure (the admins) drive away honest people (by treating them like vandals at random intervals), the Wikipedia incapsulates even more error.
30 LOC is net. You spend the first 45% of a high-reliability project doing the design work, and the last 45% doing the verification. The 10% in the middle is code generation.
These guys seem to be claiming they can reduce redundancy in the design work, and rework in the verification work. They're doing it by using a design-description method that prevents unambiguity (and therefore using a team that is TRAINED to write unambiguous requirements, so their magic language may not be the key), a coding method that avoids unprovable structure (and probably eliminates a lot of other sorts of flexibility), and a verification method that first validates the design and then verifies the code as it's produced (no new value there as everything has to be touched at least once anyway, and if a big bug turns up that causes a lot of code to be redone you have to redo formal verification on those units again; something that's less likely if formal verification is delayed until full-alpha code is demonstrated, having been informally verified along the way).
Their claims of massive error reduction are, at best, anecdotal. Let's see them do this after taking over a half-coded project with minimal design requirements, a hard deadline, and a budget that can be cut by governmental forces at will.
Try it some time. Even with multiple sources, the "facts" are only true by dint of luck. Reporters are humans, and humans do not repeat things with 100% accuracy.
Besides. It's the '00s. Journalistic integrity died in the 80s when Gannett Corporation put its first steel-box newspaper rack on the corner and started selling the USA Toady.
He's probably aware that anything he'd choose is adequate for the job that he'd put it to, and he's probably right that just choosing one and saving a few tens of thousands of dollars of meeting, ordering, installing, and evaluation time is plenty better than trying to find one that will increase their total efficiency by that much.
A reorganization, starting with the non-leader at the top.
Make adminship easier to get, but even easier than that to lose.
Rotate it around, kind of like Slashdot moderation, but with some more significant criteria than gaining a few karma points.
Most of all, don't make it self-sustaining. If it takes admins to make admins, corruption is sure to follow.
And make admins follow the policy. They don't have any special editorial authority (widely ignored right now) and their opinions are not significant (also widely ignored). They are merely the intelligent automatons that apply measures to repair and stop vandalism.
The fact that there are hundreds of admins means that every act by any user can be mistaken by one admin for something malicious, and once it happens it's all but impossible to find other admins more willing to protect the user than the admin who screwed up. That's the most important thing to fix. Admins presume guilt, and clique together. They should be purely adversarial to each other and beholden to the users. Then they'll apply their powers only as policy dictates.
When admins are un-created at about the same rate they're created (with a slight inflation for the sake of keeping up with true entropy), then I'll consider the system to be reasonably constituted. Until then, it's poorly designed and guaranteed to frustrate virtually everyone who attempts to participate.
The system itself prevents perfection by causing rational people not to return.
Because there are a few real vandals, and too few admins who are too busy patting themselves on the back, the paranoid nature of adminship results in a large number of people being treated poorly while doing what is really in the best interest of the Wikipedia.
It's a complicated problem of leadership that I've posted about before, but the upshot is that as long as Wales tolerates mistreatment of users by admins, the Wikipedia will at best remain erroneous, and more likely decay into uselessness as more and more valuable people decide it's not worth their time, energy, and intellectual capital.
The admin corps is corrupt and needs to be replaced entirely.
Their abuses of power silence innocent people and create errors in the entries, guaranteeing that it will never be an accurate and trustworthy source of information.
Wales knows it. And he refuses to fix it.
So why give him more money? The problem isn't capital, it's leadership.
The New York Times published this: David Gross, a recent Nobel Prize winner and director of the Kavli Institute for Theoretical Physics in Santa Barbara, leapt into the free-for-all, saying that 80 years had not been enough time for the new concepts to sink in. "We're just too young. We should wait until 2200, when quantum mechanics is taught in kindergarten."
Did you really say that? And were you being serious? Because differential calculus is 400 years old, and we don't start teaching that until the 9th grade...
My Fundamentals of Semiconductor Physics professor, in addition to pronouncing "Schrödinger's" as something closer to "Zreniger's", made the rather obvious-in-hindight statement that, "you never really understand QM until the third time you learn it." But the third time I learned it was 20 years ago, and I've pretty much forgotten the good parts by now. I do know that the derivation of the density of states is not something that any kindergarten class is going to understand, ever, even if we can clone Richard Feynman.
So, as the man once said, "that's not wrong, it's not even right," and I hope you're chuckling over the idea that anyone took you literally.
I've just come from the Unites States Patent Office, where they have declared me the Inventor of the "Shell-company Suing Patent Infringers" business model.
I'll take that gavel, if you please.
Unless you'd maybe like to negotiate a nominal license fee...
UNIX was designed with three (3) levels of permissions:
1. Root.
2. Group.
3. User.
This has worked fine for three decades.
If you are a user and need group access or you need something done that would alter the system, then yes, of course, you need to involve someone with more authority.
Otherwise, your normal daily tasks should only require modification of files you as a user have permission to modify.
The wide availability of linux, making every installer 'root', seems to have altered the mindset somewhat. Which was a mistake. You should have been doing very little as root, and as much as possible as an ordinary user. Then you wouldn't be on a UNIX system and thinking like a Windows user...
(N.B. In most corporate shops, even that's going away; Windows users now get far less access than they used to.)
Re:Won't they get in trouble with the FDA?
on
'Intel Inside' No More
·
· Score: 0, Flamebait
But only AMD parts will put a hole through your desk (and possibly your crotch, if its your notebook that you're tweaking at the time).
From what I've seen of animated video productions on the web involving the creators' own voices, it's a fair bet that any large number of live-action videos on the web are going to be almost universally embarassing.
Remember, people blog because they don't have enough social skill to keep their lives full otherwise. I'm going to watch no video blogs. Not even Paris Hilton's.
I've said it before, I'm saying it now, I'll say it every time someone tries to enforce security on The Internet:
THE INTERNET IS NOT SECURE
By connecting to it you must expect to be probed, attacked, sniffed, decrypted, spammed, hacked, and denied service. In order to avoid these things either you must not connect to it, or you must take measures that degrade its performance in order to eliminate some of these possibilities. But you will never make it secure, because it is not secure.
If you want a secure network, you will have to start over from scratch.
Slashdot Mirror
The Internet is an enabling technology.
The Internet is not secure.
And it does not need to be.
It was not designed so that large corporations could sell security services on it.
The Internet is an open field. A common.
If you want the Cone of Silence, you know where to find it.
Your acquaintances from Oak Ridge could redact the articles on the Manhattan Project, and within a year their edits would be superseded by layers of entropic cruft.
Wikipedia is designed to encapsulate the error rate nascent in human conversation.
And worse, because its control structure (the admins) drive away honest people (by treating them like vandals at random intervals), the Wikipedia incapsulates even more error.
Then their ability to produce bug-free code depends, as usual, on control factors, not on real-world engineering.
30 LOC is net. You spend the first 45% of a high-reliability project doing the design work, and the last 45% doing the verification. The 10% in the middle is code generation.
These guys seem to be claiming they can reduce redundancy in the design work, and rework in the verification work. They're doing it by using a design-description method that prevents unambiguity (and therefore using a team that is TRAINED to write unambiguous requirements, so their magic language may not be the key), a coding method that avoids unprovable structure (and probably eliminates a lot of other sorts of flexibility), and a verification method that first validates the design and then verifies the code as it's produced (no new value there as everything has to be touched at least once anyway, and if a big bug turns up that causes a lot of code to be redone you have to redo formal verification on those units again; something that's less likely if formal verification is delayed until full-alpha code is demonstrated, having been informally verified along the way).
Their claims of massive error reduction are, at best, anecdotal. Let's see them do this after taking over a half-coded project with minimal design requirements, a hard deadline, and a budget that can be cut by governmental forces at will.
I've been told I'm always thinking with my cockles, but my hearing is starting to go...
Try it some time. Even with multiple sources, the "facts" are only true by dint of luck. Reporters are humans, and humans do not repeat things with 100% accuracy.
Besides. It's the '00s. Journalistic integrity died in the 80s when Gannett Corporation put its first steel-box newspaper rack on the corner and started selling the USA Toady.
>For all those anti-wiki naysayers, quit your bitchin and if you have a better idea lets see you put it online.
I'm working on it.
It's going to look a lot like the wikipedia.
Only without the petty tyrannies and with real instead of ersatz liberalism.
The News is expected to be foggy. Reporters can only report what they thought they heard and their editors can only correct what they know a priori.
But the Wikipedia should be "better", right? Thousands of eyes peruse and revise it every hour.
And yet, it's about as useful as an opinion column in a major-city bulldog tabloid.
Blame its lack of real leadership.
Efficient is the key word here.
He's probably aware that anything he'd choose is adequate for the job that he'd put it to, and he's probably right that just choosing one and saving a few tens of thousands of dollars of meeting, ordering, installing, and evaluation time is plenty better than trying to find one that will increase their total efficiency by that much.
Cut and paste it a few times, then come back and see me for these: {{{{}}}}
Not necessarily a death.
A reorganization, starting with the non-leader at the top.
Make adminship easier to get, but even easier than that to lose.
Rotate it around, kind of like Slashdot moderation, but with some more significant criteria than gaining a few karma points.
Most of all, don't make it self-sustaining. If it takes admins to make admins, corruption is sure to follow.
And make admins follow the policy. They don't have any special editorial authority (widely ignored right now) and their opinions are not significant (also widely ignored). They are merely the intelligent automatons that apply measures to repair and stop vandalism.
The fact that there are hundreds of admins means that every act by any user can be mistaken by one admin for something malicious, and once it happens it's all but impossible to find other admins more willing to protect the user than the admin who screwed up. That's the most important thing to fix. Admins presume guilt, and clique together. They should be purely adversarial to each other and beholden to the users. Then they'll apply their powers only as policy dictates.
When admins are un-created at about the same rate they're created (with a slight inflation for the sake of keeping up with true entropy), then I'll consider the system to be reasonably constituted. Until then, it's poorly designed and guaranteed to frustrate virtually everyone who attempts to participate.
I'd nominate that for Best of Craigslist if I could...
The problem isn't merely entropic.
The system itself prevents perfection by causing rational people not to return.
Because there are a few real vandals, and too few admins who are too busy patting themselves on the back, the paranoid nature of adminship results in a large number of people being treated poorly while doing what is really in the best interest of the Wikipedia.
It's a complicated problem of leadership that I've posted about before, but the upshot is that as long as Wales tolerates mistreatment of users by admins, the Wikipedia will at best remain erroneous, and more likely decay into uselessness as more and more valuable people decide it's not worth their time, energy, and intellectual capital.
And, oh right, error count.
Why donate to this thing?
It's provably broken.
The admin corps is corrupt and needs to be replaced entirely.
Their abuses of power silence innocent people and create errors in the entries, guaranteeing that it will never be an accurate and trustworthy source of information.
Wales knows it. And he refuses to fix it.
So why give him more money? The problem isn't capital, it's leadership.
If it (censorship, warrantless eavesdropping, sophistry-based policymaking) is good enough for America, it's good enough for China.
(Email I just sent to David Gross:)
The New York Times published this:
David Gross, a recent Nobel Prize winner and director of the Kavli Institute for Theoretical Physics in Santa Barbara, leapt into the free-for-all, saying that 80 years had not been enough time for the new concepts to sink in. "We're just too young. We should wait until 2200, when quantum mechanics is taught in kindergarten."
Did you really say that? And were you being serious? Because differential calculus is 400 years old, and we don't start teaching that until the 9th grade...
My Fundamentals of Semiconductor Physics professor, in addition to pronouncing "Schrödinger's" as something closer to "Zreniger's", made the rather obvious-in-hindight statement that, "you never really understand QM until the third time you learn it." But the third time I learned it was 20 years ago, and I've pretty much forgotten the good parts by now. I do know that the derivation of the density of states is not something that any kindergarten class is going to understand, ever, even if we can clone Richard Feynman.
So, as the man once said, "that's not wrong, it's not even right," and I hope you're chuckling over the idea that anyone took you literally.
I've just come from the Unites States Patent Office, where they have declared me the Inventor of the "Shell-company Suing Patent Infringers" business model.
I'll take that gavel, if you please.
Unless you'd maybe like to negotiate a nominal license fee...
Anyone for Windows XP SP3?
The Borg looked down on autonomous humans, too. Technically, it's known as denial.
UNIX was designed with three (3) levels of permissions:
1. Root.
2. Group.
3. User.
This has worked fine for three decades.
If you are a user and need group access or you need something done that would alter the system, then yes, of course, you need to involve someone with more authority.
Otherwise, your normal daily tasks should only require modification of files you as a user have permission to modify.
The wide availability of linux, making every installer 'root', seems to have altered the mindset somewhat. Which was a mistake. You should have been doing very little as root, and as much as possible as an ordinary user. Then you wouldn't be on a UNIX system and thinking like a Windows user...
(N.B. In most corporate shops, even that's going away; Windows users now get far less access than they used to.)
But only AMD parts will put a hole through your desk (and possibly your crotch, if its your notebook that you're tweaking at the time).
From what I've seen of animated video productions on the web involving the creators' own voices, it's a fair bet that any large number of live-action videos on the web are going to be almost universally embarassing.
Remember, people blog because they don't have enough social skill to keep their lives full otherwise. I'm going to watch no video blogs. Not even Paris Hilton's.
I've said it before, I'm saying it now, I'll say it every time someone tries to enforce security on The Internet:
THE INTERNET IS NOT SECURE
By connecting to it you must expect to be probed, attacked, sniffed, decrypted, spammed, hacked, and denied service. In order to avoid these things either you must not connect to it, or you must take measures that degrade its performance in order to eliminate some of these possibilities. But you will never make it secure, because it is not secure.
If you want a secure network, you will have to start over from scratch.
About 15 years ago bars started selling novelty shots in test tubes.
I have no idea whether it makes a difference, because once people start ordering shots, the party's only started.