Not quite. In the classic example, it means you don't store a customer's address in the table of customer orders. If you do, you can wind up with several orders containing the same customer data (for a customer that has more than one order pending). If they change their address, you have to make sure to change *every* copy; worse, if the customer has no orders pending, you have *no* copies of the address.
Putting the data in third normal form would require setting up one table for customer info, another for orders, and each order would contain a foreign key -- an account number, for instance -- to link the order to a customer record. If the customer enters an address change, you update *one* record in the database, and all orders get the new information. Each bit of information is stored in one and only one place, and it will never get into an inconsistent state.
There are other "levels" of normalization, first through fifth, but 3NF is kind of a "sweet spot"; lower levels result in inconsistency in common cases, while higher orders tend to be more effort than they're worth.
Google will want its IM competitors to invest some effort into talking to Jabber servers, or at least playing nicely with inbound connectivity; but if Google is perceived as "owning" Jabber (blurring the distinction between Jabber-the-protocol and Jabber-the-company), it could scare them away from it. So I'm betting no.
In fact, I think that the whole premise that Google is gearing up for one big acquisition is wrong; a string of small acquisitions seems more in keeping with past behavior.
You see, comics are a VISUAL medium. They contain both image and story (ideally, the image IS the story). If all you care about is story, then you should make a book instead of a comic.
Movies are a visual medium too. Does story not matter there? The Dungeons & Dragons movie had some impressive visuals and sucky writing; does that make it a better movie than Casablanca? (No color, no explosions, they even had Peter Lorre's character die off-screen. What a rip!)
You see, comics are a STORYTELLING medium. They contain both image and story (and characters, and setting, and theme, and all that icky stuff they teach in English Comp). If all you care about is imagery, perhaps you should make stand-alone illustrations instead of a comic.
It's worth pointing out that some of the meaning of the original quote has been lost. In Marx's day, "opiate" didn't mean "addictive recreational drug"; it meant "powerful, potentially harmful painkiller".
Put another way, Marx saw capitalism as a disorder for which religion was a dangerous, limited remedy, and communist revolution the cure.
Yecchhh, I'm actually defending Marx. I'm going to go take a shower now...
Be aware that changing for user agent string can have unintended consequences.
True. Konqueror provides a less problematic version of this: you can change your user agent string by domain. So MSN.com sees you as IE 5 on Windows, but you identify correctly everywhere else.
It's also very handy when dealing with banks and brokerages, which tend to be picky about acceptable browsers. I just wish Mozilla and/or Galeon had this capability.
(And let's take the obligatory "why the hell would you want to go to MSN.com anyway?" comment as read.)
I had the same problem at first. After watching the show for a while, I'm completely over it. Puppetry is kind of like animation: if the only context in which you've seen it is in kids' shows and comedy, it's jarring to see it applied in a more dramatic context.
It doesn't help that Rigel, one of the puppet characters, is so frequently used for comic relief. But then the writers have Rigel torture a captive to death (he had it coming, but still...) and in so doing, extract a life-saving bit of information -- and suddenly, you no longer think about the muppetly cuteness and the hands inside making Rigel move: he's a real character.
And think about it: how do you do truly alien aliens in a live-action show on a realistic budget, or even an unrealistic one? Humans -- even extremely large or small humans in elaborate costume and makeup -- don't quite fill the bill. The alternatives are CGI and puppetry. And if the choice is between Jar Jar Binks and Rigel XVI, well...
If Saddam Hussein had been killed (by a "competitor"), the US would have been the first to celebrate.
Maybe, maybe not. but I can't see Americans celebrating the death of Iraqi civilians, even during wartime. We know who Hussein is, many believe (rightly or wrongly) that his actions deserve punishment, even death. But celebrating the death of civilians, about whom you know nothing but their nationality, is another matter entirely.
Any fanatic groups, and there are some of them in every country of the world. Search fro "waco" on google for instance.
I assume you're talking about the Branch Davidians; a Google search for "Waco" would turn up a lot of information about a city in Texas having nothing to do with fanaticism, religious or otherwise. (And as a native of Oklahoma City, I'm disappointed that my former hometown is now a popular synonym for domestic terrorism.)
It's far from proven that the Branch Davidians committed suicide. Neither were they terrorists; trouble came looking for them, not vice versa.
No, we shouldn't make assumptions about who is responsible, and unless the government has a lot more information than the public does, it's way too early for retaliation. But the Palestinians have shown that, even if they're not the ones responsible, they consider those responsible to be friends, allies, and even (retch) heroes. How could we not suspect them?
After the Murrah Building was destroyed in Oklahoma City, and the wreckage cleared away, it was decided to leave the lot permanently vacant. While I understand the motivation for this decision, I respectfully disagree. There are those (few in number, I hope) who will always see the empty spot where the Murrah building once was as a monument to McVeigh, not to his victims. He left his mark on the city. There is no way that a rebuilt Murrah Building could be so interpreted.
It's up to the New Yorkers themselves, of course, but I hope they do choose to rebuild, and build higher and stronger than before. As Heinlein said, living well is the best revenge.
There is one practical concern, though; a rebuilt WTC would be a target for new attacks. Maybe a SAM battery on the rooftop...
My theory is that the budget and/or costume technology for the original show were the primary factors in why the original klingons looked the way they did.
No shit, Sherlock?
Actually, that explanation was fine up until the DS9 characters actually noticed the change in that time-travel episode. And given that they didn't supply an explanation for it, it's inevitable that people are going to come up with ideas of their own.
Retroactive continuity may not be your idea of fun, but let's not get petty.
Perhaps the klingons from the original series, identified as being from the planet "Khlinzahi", were an offshoot bunch...
All right, time to geek out...
That explanation doesn't work. Why did Kang, Kor, and Koloth have different appearances from one series to the next? Species change operation?
Here's my thought: Klingon leadership during the original series era was obsessed with spying and infiltration, and required active-duty military personnel to modify their appearances so they could infiltrate human installations on short notice. Arne Darvin from "Tribbles" was an extreme example of this.
Between TOS and TMP, there was a change in the Klingon high command. The new bosses rejected the espionage-oriented policy as dishonorable, and the Klingon troops all gratefully switched back to their native forms. They're mortified by the whole affair to this day, so "We do not speak of it."
Then again, it's possible that Q mucked with Klingon evolution just so he could use the "macro-head with a micro-brain" line. That would be just like him, wouldn't it?
Therefore, it IS the fault of the Bush administration
How is an administration that has been in office for six months responsible for a treaty that has been signed, but gone un-ratified, for three and a half years? Get your facts straight. Long before Bush came to office, the Senate voted, 96-0, that it would not ratify any global warming treaty that let China, India, and developing countries in general off the hook.
Clinton could have tried to get it ratified, but he didn't. He could have tried to get it amended to include China, India, et al., but he didn't. As far as he was concerned, the treaty had already performed its primary function, which was to make him look good and the Repubs look bad; ratification was irrelevant.
By the time GWB came to office, Kyoto was already a dead letter; all he did was publicly acknowledge this fact.
Re:Another Scripting Language, Ho Hum
on
Why not Ruby?
·
· Score: 1
If you cannot learn a new programming language in a week or two, your computer programming education is deficient.
"Learning a new programming language" is a task without a well-defined completion condition. Yes, you can learn the core syntax of a language in a week or two, but that's just a start.
I know Perl well enough to get paid writing it for a couple of years now. But I'm still learning it, and getting noticeably better and more productive with it the longer I study and use it. It would take more than a couple of weeks' study of Ruby to approach my current level of productivity in Perl.
I can see applying myself to further study of Perl; I've been meaning to sit down and grok Parse::RecDescent in its fullness. I can see studying something wildly different from Perl, to broaden my horizons; enough people suggested studying Lisp that I'll be looking into that next. And I can see looking at tools that apply to a different problem domain; that was why I learned Perl in the first place, while I was still gainfully employed hacking in C and Java.
But Ruby uses the same basic paradigm as Perl, and applies to the same problem domain. Even if it's (by some measure, say) 20% better than Perl, it still won't make me more productive than I am now in Perl without a lot more than "a week or two" of study and practice. And no one's paying me to learn or use it. So why bother spending a couple of weeks gaining a superficial understanding of Ruby, when I could be learning something completely different, or getting better at Perl?
But if the new series is set in the 22nd Centuy, there can be no Borg or Q, and the technology will be crappy!
You say that as if it's a bad thing. But it's not a bug, it's a feature.
Q and the Borg have been done to death; it's time to try something different.
And the ultra-high tech in the 24th century version of the series actually limits the dramatic possibilities. How can anything threaten characters who have hand weapons powerful enough to vaporize a mountain? How can anyone be stranded on the surface of a hostile planet, when all you have to do is tap your comm badge, and say "Beam me up"?
You notice how many times Enterprise-D and Voyager encountered planets with funky atmospheres or force fields that made it impossible to use the transporters? They had to contrive to limit the tech, time after time until it got silly, in order to allow for some dramatic tension.
It might be nice to live in a time where your problems can be solved with the push of a button, but it makes for lousy drama.
If they're smart, the new (old?) Enterprise will have no transporter at all, or else it will be a finicky, emergencies-only, use-at-your-own-risk system.
Version control systems do a lot more that saving multiple copies of the file.
Use a central repository, rather than everyone keeping copies of everything
Store as deltas (for text files) rather than keep multiple copies of the whole file
Maintain version history with comments
Maintain multiple branches (eg., release and development branches)
Maintain a checkin/checkout discipline, so developers don't interfere with each others work
The VMS filesystem's ability to store multiple versions of files basically had the same function as the automatic "make a backup copy" capability in many text editors and word processors; it let you roll back mistakes. But you were expected to evenutally purge the old versions to keep from filling up the disk.
(You say they're calling it 'ACMS' now? Back when I was using it regularly, in '87-'93, it was just called 'CMS'.)
(alongside the original Dr. Strangelove himself, the truly sinister Edward Teller).
Not necessarily true; Wernher von Braun, and Herman Kahn of the Rand Corporation, are the more likely inspirations for the character. It's particularly unfair to compare Strangelove (a German and ex-Nazi) to Teller, who is Hungarian, and who fled the Nazi invasion of his homeland.
LISP isn't auto-obfuscated, you just hate parentheses.
I dislike LISP because my attitude towards parentheses is the same as my attitude towards people. One by one, they might be okay, but when you get a big crowd of them, they're a pain in the ass.
I saw Damian present Text::Autoformat at the Perl Conference. Briefly, it's a module that reformats emails with deeply nested replies (with several levels of ">" or ":" or "FOO>" prefixes), even to the point of aligning and outdenting bullet lists, and aligning and, if necessary, renumbering numbered lists, making intelligent guesses about whether the list is numbered in numerically, alphabetically, or with Roman numerals! Then he showed how he had mapped it to a "reformat this paragraph" key in vi...
As I watched the lecture unfold, and I understood just what a clever and puissant hack this was, I had the urge to get up and do the Wayne's World "I'm not worthy!" chant. There's nothing like the presence of greatness to make you feel like a rank amateur.
Later, he did his Quantum::Superposition talk, which was impressive whether you look at it as physics, programming, or stand-up comedy.
Now think about this: he came up with all these neat hacks literally in his spare time, between teaching C to freshmen and writing material for publication in "real" CompSci literature.
$55,000 for a year of Damian's full attention? Worth every penny IMHO (especially since that year coincides with gestation and early maturation of Perl 6).
They explored why people would devote large amounts of time to writing reviews and answering questions for complete strangers. The short answer was "egoboo" or ego boosts that came from being positively rated as a reviewer.
Working for monetary payment, and working for "payment" in the form of appreciation of your genius, are not necessarily mutually exclusive. As Heinlein said, money is the sincerest form of flattery...
Once you give money to PayPal you'll never see it again unless and until you sue them.
Has anyone actually had this happen? Anyone had any bad experiences with PayPal (or e-gold, or any of their other competitors)? Any disputes not resolved to satisfaction?
There was a reference to this anti-Moody rant; it seems that ol' Fred has had a reputation as a Microsoft shill for some time now.
It was also pointed out that Moody seems to have arrived at his Linux bug-counts by adding together the Red Hat and Linux Aggregate counts -- meaning that Red Hat bugs were counted twice. Counted correctly, NT has more Bugtraq entries than Linux.
Sorry I took so long replying; I've been out of town.
While I agree with most of your points, I disagree here. We would actually have Athena apps, with no possibility of even getting a working menu that looks different. Not putting widgets on the server was one of X's big wins, really.. I very much recommend that the designers of X's replacement not be so arrogant as to think they can dictate the design of user interfaces!
I'm not saying the designers of X, or its hypothetical replacement, should design the widget set. But neither should the designers of X applications choose the widget set, as is done now. That choice should be up to the user, by configuring the display server (i.e., the workstation).
Here's the scenario: an X application resides on a departmental server. The users of this app have various types of boxen on their desks: Linux, Windows, Mac, etc.
When the application starts, it sends a message to the workstation, saying "the application has a 'File' menu, containing the following items...": no description of how those items look, where the menu should appear, etc., just an abstract description of the menu contents.
When the application is run from the KDE box, the workstation software interprets the menu description and renders it on-screen as a KDE menu; the Windows box renders it as a Windows-style menu; and so on. In a configurable X display server, the user could change a display setting, and presto! the KDE app is magically turned into GNOME app.
Because the display software, not the application, would determine look and feel. And thus the application could adopt *any* look and feel that the user desired, without having to be ported or even recompiled.
Slashdot Mirror
Not quite. In the classic example, it means you don't store a customer's address in the table of customer orders. If you do, you can wind up with several orders containing the same customer data (for a customer that has more than one order pending). If they change their address, you have to make sure to change *every* copy; worse, if the customer has no orders pending, you have *no* copies of the address.
Putting the data in third normal form would require setting up one table for customer info, another for orders, and each order would contain a foreign key -- an account number, for instance -- to link the order to a customer record. If the customer enters an address change, you update *one* record in the database, and all orders get the new information. Each bit of information is stored in one and only one place, and it will never get into an inconsistent state.
There are other "levels" of normalization, first through fifth, but 3NF is kind of a "sweet spot"; lower levels result in inconsistency in common cases, while higher orders tend to be more effort than they're worth.
Google will want its IM competitors to invest some effort into talking to Jabber servers, or at least playing nicely with inbound connectivity; but if Google is perceived as "owning" Jabber (blurring the distinction between Jabber-the-protocol and Jabber-the-company), it could scare them away from it. So I'm betting no.
In fact, I think that the whole premise that Google is gearing up for one big acquisition is wrong; a string of small acquisitions seems more in keeping with past behavior.
Me too, please.
kevin at shaum dot com.
Thanks.
Movies are a visual medium too. Does story not matter there? The Dungeons & Dragons movie had some impressive visuals and sucky writing; does that make it a better movie than Casablanca? (No color, no explosions, they even had Peter Lorre's character die off-screen. What a rip!)
You see, comics are a STORYTELLING medium. They contain both image and story (and characters, and setting, and theme, and all that icky stuff they teach in English Comp). If all you care about is imagery, perhaps you should make stand-alone illustrations instead of a comic.
It's worth pointing out that some of the meaning of the original quote has been lost. In Marx's day, "opiate" didn't mean "addictive recreational drug"; it meant "powerful, potentially harmful painkiller".
Put another way, Marx saw capitalism as a disorder for which religion was a dangerous, limited remedy, and communist revolution the cure.
Yecchhh, I'm actually defending Marx. I'm going to go take a shower now...
It's also very handy when dealing with banks and brokerages, which tend to be picky about acceptable browsers. I just wish Mozilla and/or Galeon had this capability.
(And let's take the obligatory "why the hell would you want to go to MSN.com anyway?" comment as read.)
I had the same problem at first. After watching the show for a while, I'm completely over it. Puppetry is kind of like animation: if the only context in which you've seen it is in kids' shows and comedy, it's jarring to see it applied in a more dramatic context.
It doesn't help that Rigel, one of the puppet characters, is so frequently used for comic relief. But then the writers have Rigel torture a captive to death (he had it coming, but still...) and in so doing, extract a life-saving bit of information -- and suddenly, you no longer think about the muppetly cuteness and the hands inside making Rigel move: he's a real character.
And think about it: how do you do truly alien aliens in a live-action show on a realistic budget, or even an unrealistic one? Humans -- even extremely large or small humans in elaborate costume and makeup -- don't quite fill the bill. The alternatives are CGI and puppetry. And if the choice is between Jar Jar Binks and Rigel XVI, well...
Maybe, maybe not. but I can't see Americans celebrating the death of Iraqi civilians, even during wartime. We know who Hussein is, many believe (rightly or wrongly) that his actions deserve punishment, even death. But celebrating the death of civilians, about whom you know nothing but their nationality, is another matter entirely.
I assume you're talking about the Branch Davidians; a Google search for "Waco" would turn up a lot of information about a city in Texas having nothing to do with fanaticism, religious or otherwise. (And as a native of Oklahoma City, I'm disappointed that my former hometown is now a popular synonym for domestic terrorism.)
It's far from proven that the Branch Davidians committed suicide. Neither were they terrorists; trouble came looking for them, not vice versa.
No, we shouldn't make assumptions about who is responsible, and unless the government has a lot more information than the public does, it's way too early for retaliation. But the Palestinians have shown that, even if they're not the ones responsible, they consider those responsible to be friends, allies, and even (retch) heroes. How could we not suspect them?
After the Murrah Building was destroyed in Oklahoma City, and the wreckage cleared away, it was decided to leave the lot permanently vacant. While I understand the motivation for this decision, I respectfully disagree. There are those (few in number, I hope) who will always see the empty spot where the Murrah building once was as a monument to McVeigh, not to his victims. He left his mark on the city. There is no way that a rebuilt Murrah Building could be so interpreted.
...
It's up to the New Yorkers themselves, of course, but I hope they do choose to rebuild, and build higher and stronger than before. As Heinlein said, living well is the best revenge.
There is one practical concern, though; a rebuilt WTC would be a target for new attacks. Maybe a SAM battery on the rooftop
Actually, that explanation was fine up until the DS9 characters actually noticed the change in that time-travel episode. And given that they didn't supply an explanation for it, it's inevitable that people are going to come up with ideas of their own.
Retroactive continuity may not be your idea of fun, but let's not get petty.
That explanation doesn't work. Why did Kang, Kor, and Koloth have different appearances from one series to the next? Species change operation?
Here's my thought: Klingon leadership during the original series era was obsessed with spying and infiltration, and required active-duty military personnel to modify their appearances so they could infiltrate human installations on short notice. Arne Darvin from "Tribbles" was an extreme example of this.
Between TOS and TMP, there was a change in the Klingon high command. The new bosses rejected the espionage-oriented policy as dishonorable, and the Klingon troops all gratefully switched back to their native forms. They're mortified by the whole affair to this day, so "We do not speak of it."
Then again, it's possible that Q mucked with Klingon evolution just so he could use the "macro-head with a micro-brain" line. That would be just like him, wouldn't it?
How is an administration that has been in office for six months responsible for a treaty that has been signed, but gone un-ratified, for three and a half years? Get your facts straight. Long before Bush came to office, the Senate voted, 96-0, that it would not ratify any global warming treaty that let China, India, and developing countries in general off the hook.
Clinton could have tried to get it ratified, but he didn't. He could have tried to get it amended to include China, India, et al., but he didn't. As far as he was concerned, the treaty had already performed its primary function, which was to make him look good and the Repubs look bad; ratification was irrelevant.
By the time GWB came to office, Kyoto was already a dead letter; all he did was publicly acknowledge this fact.
"Learning a new programming language" is a task without a well-defined completion condition. Yes, you can learn the core syntax of a language in a week or two, but that's just a start. I know Perl well enough to get paid writing it for a couple of years now. But I'm still learning it, and getting noticeably better and more productive with it the longer I study and use it. It would take more than a couple of weeks' study of Ruby to approach my current level of productivity in Perl.
I can see applying myself to further study of Perl; I've been meaning to sit down and grok Parse::RecDescent in its fullness. I can see studying something wildly different from Perl, to broaden my horizons; enough people suggested studying Lisp that I'll be looking into that next. And I can see looking at tools that apply to a different problem domain; that was why I learned Perl in the first place, while I was still gainfully employed hacking in C and Java.
But Ruby uses the same basic paradigm as Perl, and applies to the same problem domain. Even if it's (by some measure, say) 20% better than Perl, it still won't make me more productive than I am now in Perl without a lot more than "a week or two" of study and practice. And no one's paying me to learn or use it. So why bother spending a couple of weeks gaining a superficial understanding of Ruby, when I could be learning something completely different, or getting better at Perl?
You say that as if it's a bad thing. But it's not a bug, it's a feature.
Q and the Borg have been done to death; it's time to try something different.
And the ultra-high tech in the 24th century version of the series actually limits the dramatic possibilities. How can anything threaten characters who have hand weapons powerful enough to vaporize a mountain? How can anyone be stranded on the surface of a hostile planet, when all you have to do is tap your comm badge, and say "Beam me up"?
You notice how many times Enterprise-D and Voyager encountered planets with funky atmospheres or force fields that made it impossible to use the transporters? They had to contrive to limit the tech, time after time until it got silly, in order to allow for some dramatic tension.
It might be nice to live in a time where your problems can be solved with the push of a button, but it makes for lousy drama.
If they're smart, the new (old?) Enterprise will have no transporter at all, or else it will be a finicky, emergencies-only, use-at-your-own-risk system.
Version control systems do a lot more that saving multiple copies of the file.
The VMS filesystem's ability to store multiple versions of files basically had the same function as the automatic "make a backup copy" capability in many text editors and word processors; it let you roll back mistakes. But you were expected to evenutally purge the old versions to keep from filling up the disk.
(You say they're calling it 'ACMS' now? Back when I was using it regularly, in '87-'93, it was just called 'CMS'.)
More on this subject here.
As I watched the lecture unfold, and I understood just what a clever and puissant hack this was, I had the urge to get up and do the Wayne's World "I'm not worthy!" chant. There's nothing like the presence of greatness to make you feel like a rank amateur.
Later, he did his Quantum::Superposition talk, which was impressive whether you look at it as physics, programming, or stand-up comedy.
Now think about this: he came up with all these neat hacks literally in his spare time, between teaching C to freshmen and writing material for publication in "real" CompSci literature.
$55,000 for a year of Damian's full attention? Worth every penny IMHO (especially since that year coincides with gestation and early maturation of Perl 6).
MSDW => Morgan Stanley Dean Witter, the firm underwriting the offering. They're also the company behind the Discover Card, if I remember right.
There was a reference to this anti-Moody rant; it seems that ol' Fred has had a reputation as a Microsoft shill for some time now.
It was also pointed out that Moody seems to have arrived at his Linux bug-counts by adding together the Red Hat and Linux Aggregate counts -- meaning that Red Hat bugs were counted twice. Counted correctly, NT has more Bugtraq entries than Linux.
Does that make NT the worst OS of all time?
I'm not saying the designers of X, or its hypothetical replacement, should design the widget set. But neither should the designers of X applications choose the widget set, as is done now. That choice should be up to the user, by configuring the display server (i.e., the workstation).
Here's the scenario: an X application resides on a departmental server. The users of this app have various types of boxen on their desks: Linux, Windows, Mac, etc.
When the application starts, it sends a message to the workstation, saying "the application has a 'File' menu, containing the following items ...": no description of how those items look, where the menu should appear, etc., just an abstract description of the menu contents.
When the application is run from the KDE box, the workstation software interprets the menu description and renders it on-screen as a KDE menu; the Windows box renders it as a Windows-style menu; and so on. In a configurable X display server, the user could change a display setting, and presto! the KDE app is magically turned into GNOME app.
Because the display software, not the application, would determine look and feel. And thus the application could adopt *any* look and feel that the user desired, without having to be ported or even recompiled.