Javascript, or scripting of any other sort, should be BANNEd in PDF documents. PDF means Portable Document Format. Not dynamic. Not editable. Sure, it is nice to be able to edit a PDF directly and save it, but scripting within the document is too much feature creep. Almost as retarded as having a router hijack the user's internet connection to sell software.
I'll admit, the ability to view a 3D image within a document (turning it and such) has very cool possibilities... but don't call it PDF. Don't bloat something with more than it needs. Use something that matches those goals more closely, or come out with a new document type so users aren't confronted with something they're not prepared for...</soapbox>
In my experience, the only time a support line is used is when the person having the problem doesn't know of somebody in their family/friends that could fix it. Your typical "Gramma Jane" doesn't feel comfortable talking to some 16-year-old tech support rep that doesn't know the first thing about communicating with people that have a different level of knowledge than they do.
DISCLAIMER: Anybody that knows me comes to me almost immediately to fix their computers. And I have a feeling the "No, I won't fix your computer" shirt would only make it worse.
I didn't see anything about a specific version of OpenID when I first read it... I actually had to do a CTRL+F in Firefox to find it.
That said, I'll have to admit I read the second article first and was a bit confused that Google (for some of you that's a.k.a. Saint Google) would bastardize an open protocol and make it unusable for any system but their own. Once I read the second article (and found the single reference to "OpenID 2.0")... well, I completely agree with the parent's comments.
One last thing I have heard quite a few others praise is the ability to open almost any document file type out there right out of the box, now that OOo 3.0 has Office 2007 XML support.
I've been able to open Office '07 documents (*.docx) for a while, and I'm using OOo 2.4. There's an upgrade that makes this work better?
SIDE NOTE: has anybody else expanded the contents of a docx file and wondered why it's so convoluted?
Okay, so a developer using GIT has 10 branches working on a couple of different features and bugfixes. They're long-running works, so that developer hasn't put them into the main repository yet... all of a sudden, that developer disappears: they were involved in a car accident & killed, their laptop--which contained the branches--was mercilessly burned into a pile of smoldering ashes in the process.
Now we've got a company with X developers that were waiting on those changes. They'll never see the changes, because they were destroyed along with the person that was about to implement them. So now another developer (or a few) has to devote time to fixing something that was never documented; they may not even have any idea how the afore-mentioned dead developer found the problem or how to fix it without crippling the entire program...
That seems bad. If they'd been using SVN with their own branch(es), there would be a history of it. The remaining developers could just go through and finish up those branches & merge them into the main line. Seems to me like SVN is better, but maybe I'm wrong (I've only ever used CVS & SVN, upgrading to the latter from the former).
This may already have been said (since I came into this article late), but I have to add my two cents:
Wouldn't it be better to have every check-in be stable?
Have you ever worked on a large software project? Sorry if it sounds offensive, but really, it is unlikely at best to make sure that a given commit (or "check-in") is entirely stable. The likelihood decreases exponentially as the number of target systems increases (especially if it includes both Windows and Linux).
When I setup a source control system, I use it as much for the ability to keep track of my changes as I do for maintaining the code. I've been known on occasion to make a commit that I know beforehand will destabilize the product, and then revert those changes immediately in the next commit; the log often includes "I know this is broken" and something about needing to have the change in there for historical reasons.
If there's lots of developers on a project and the leader is anal about having every commit as stable as possible, I suggest SVK. It can mirror a repository locally, and all changes can be committed whenever needed without breaking the repository (especially handy when the repo is setup so that there's no chance of branching).
Riiight... this is similar to the Prohibition (alcohol) and the steep punishments for drugs. Except instead of lessening the problem, it will simply lessen the number of people sending the stuff.
Since there will eventually be fewer willing to send SPAM, those willing to tempt fate will get paid many times more than previously, and will employ even more technology to make tracing them to their real location even more infeasible. Instead of allowing mail to be sent from their own mail servers, it will be based purely off botnets--meaning the price of botnets will skyrocket. In turn, this means the price of black market exploits will also skyrocket... this will increase the number of infected PC's on the Internet.
Now "small" botnets will number in the hundreds of thousands. Most Windows PCs will have been exploited, with each system potentially in multiple nets. More people will bitch that their systems are slow.
Here's where the geeks step in. "Hey, it'll run way faster if you just put Linux on there. Then you don't have to worry about being a part of their botnet."
Thus ushers in the year of Linux on the desktop. Let's do it.
Or the botnet is setup to continuously spew out the same email until "told" otherwise. Since there's nobody to tell it to stop, it just keeps spewing the same stuff over & over again. I mean, we are talking about spam here (most of it isn't even coherent).
Why go through a web hosting company? It's pretty easy to host your own website:
Get a broadband connection
Buy a cheap desktop/server class computer from eBay
Install Linux with Apache, PHP, etc
For handling IP changes: setup DynDNS, get a static IP from ISP, or deal with changes manually
I don't use DynDNS nor a static IP, and I hostmyown websites. I didn't bother setting up my own mailserver, as Google Apps more than fits the bill for that. So what's the problem?
Your farm, ranch, house, whatever, won't last long against a single blast from a tank. And how many ordinary citizens have access to anti-tank weapons or know the first thing about defending against bombers or attack choppers?
The problem isn't achieving the know-how to handle the armored might of the U.S. military. The problem isn't even with protecting your property against attacks from them. The problem is getting a large enough squad together to make an impact, and convincing the populace that we're fighting is for the greater good (so if a "freedom fighter" needs someplace to hide, they don't have to be afraid of their neighbors).
Guerrilla tactics aren't that hard to learn. Go on the Internet and ask Google. Homemade bombs aren't even that hard to make, provided one can get their hands on enough of the right fertilizer. But in a game of attrition, the U.S. military will win hands-down.
But let's say, just for the sake of saying it, that a large enough force has been built up to make an impact. An impact large enough for the Government, even the President, to pause. What now? Queue government-funded media spin on the "War on Terror". Now what? Storm the White House? Riiight...
The plain fact of the matter is that the Government would have to be completely overthrown in order to affect the necessary changes. And in doing so, we'd open the country up to invasion by any and all of its enemies.
Moral of the story? Subtlety doesn't work, and overthrowing the Government leaves us too open to outside attack/invasion.
Wow. That is truly an awesome observation. By their own statistics, they have no reason to use DRM. Now the question is: "why do you include DRM in your games when so few people apparently care about it?" Then the other question is, "if piracy of your game has damaged sales so much that you were 'forced' to add DRM, have you actually seen a marked improvement in sales and a decline in piracy?" (the former being far more important that the latter for determination of continued usage of DRM)
With a proper "checks and balances" system, the power to destroy evidence or alter the results can be seemlessly removed from the hands of a few:
1.) All information is encrypted in a way that it cannot be locally decrypted
2.) no information is stored locally: all responses are *immediately* sent to a remote server
3.) information is stored in multiple centralized servers: data must be verified as matching on all servers or it is either a.) flagged as tampered, or b.) discarded entirely (NOTE: discarding a vote entirely could easily be used to rig an election simply by tampering with enough votes from "the other guy"). Alerts are sent to multiple people when tampering is found, so no one person can be blamed if nothing is done.
In any given situation, given enough cooperation from enough people in power, any security system can be trumped or hacked. With enough restrictions and forethought, a system can be secured enough that there would simply be too many people that would have to cooperate and too many systems that would have to be simultaneously altered, making a hack less likely. By having enough redundant systems and proper encryption, the amount of effort involved in breaking or altering a system far outweighs any payouts... and the greater number of people that are involved, the greater chance that it won't work or that it will be exposed.
Geek: "Stop turning off the TV!" Wife: "Why?" Geek: "I told you this a hundred times. My wireless signal comes off the light waves from the TV." Wife: (turns TV back on) "There, are you happy?" Geek: "I told you this a hundred times. My wireless signal comes off the..." Wife: "YOUR STUPID TV IS ON!" Geek: "Yeah, but YOU'RE STANDING IN FRONT OF IT!"
Amazingly insightful. If I had mod points--and you hadn't posted as AC--I would have modded you up. Another problem is that the technology would take decades to be useful, and then it couldn't be completely trusted. Reliability would depend on the idea that every car on the road had that same technology... imagine what would happen if the technology actually was put into every car, and then a car show came into town...
Na. Use a picture or a small clip from "Two Girls and a Milkshake" video (just the first 30 seconds should do). Guaranteed you'll get your laptop back in 30 seconds or less... well, if they don't drop it first.
Here's a hypothetical to think about next time you have to set one of those "security questions": WebsiteA asks you to choose a password that, at a minimum:
* 8 characters in length
* 1 capital letter
* 1 lowercase letter
* 1 number
* 1 non-alphanumeric (special character)
* can't have any variation of a word associate with your name, D.O.B., home city, etc
* no dictionary words
Now they ask you a security question, "What's your mother's first name?"
Now a determined "hacker" goes and tries to brute-force your password. They make a couple of valiant efforts to brute force your password before realizing it's about to get locked-out from too many wrong passwords (or they don't bother, since it's infeasible to answer in just 2 attempts).
Then they click the "Forgot password?" link. Using the information they pulled out of your trash, they easily defeat all that security and reset your password. Seconds after logging in, they reset your security question/password as well, effectively locking you out of your account.
Oh, and if I remember correctly, most of those websites that lock your account after X failed attempts at password have no limit for responses to security questions. So why bother brute-forcing the password when they can do so to your security question which is invariably based on a real word (and orders of magnitude simpler to guess by a computer)?
Now I ask, why would I EVER give real answers to the security questions?
I absolutely agree, especially when it comes to repetitive jobs. Humans get bored and lose interest in repetitive things quickly; computers, on the other hand, are infinitely patient and don't take shortcuts after too long (well, unless they've been programmed that way/poorly... remember how that one version of Windows deleted command.com if the system was on for 90 consecutive days?)
I'm pretty sure the main source of an explosion isn't the gasoline, but the fumes it puts off. Drop a match on the ground and throw gasoline on it (don't pour beside). Now throw another match into that "empty" container. Which one makes the bigger boom? Maybe I'm wrong... (looks like I'll have to test that tonight).
So much talk about bears and wild animals thinking the wallet might be tasty, or a bag... what about the body? Seems like, after this much time, the dude's dead. The animals probably thought he was a mighty tasty treat... and the damn "dogs" buried his bones for a nice snack later.
NOTE TO SELF: remember to reschedule that class on sensitivity.
Slashdot Mirror
Javascript, or scripting of any other sort, should be BANNEd in PDF documents. PDF means Portable Document Format. Not dynamic. Not editable. Sure, it is nice to be able to edit a PDF directly and save it, but scripting within the document is too much feature creep. Almost as retarded as having a router hijack the user's internet connection to sell software.
I'll admit, the ability to view a 3D image within a document (turning it and such) has very cool possibilities... but don't call it PDF. Don't bloat something with more than it needs. Use something that matches those goals more closely, or come out with a new document type so users aren't confronted with something they're not prepared for...</soapbox>
In my experience, the only time a support line is used is when the person having the problem doesn't know of somebody in their family/friends that could fix it. Your typical "Gramma Jane" doesn't feel comfortable talking to some 16-year-old tech support rep that doesn't know the first thing about communicating with people that have a different level of knowledge than they do.
DISCLAIMER: Anybody that knows me comes to me almost immediately to fix their computers. And I have a feeling the "No, I won't fix your computer" shirt would only make it worse.
No, no. Duck Dodgers said "twenty-fourth-and-a-half century." Get it right.
That is an interesting tidbit. Do you have any sources for this information (i.e. links to the internets)?
Read the first article. They're using OpenID 2.0; just search for "OpenID 2.0" and you'll see it... maybe they should have made it more obvious.
I didn't see anything about a specific version of OpenID when I first read it... I actually had to do a CTRL+F in Firefox to find it.
That said, I'll have to admit I read the second article first and was a bit confused that Google (for some of you that's a.k.a. Saint Google) would bastardize an open protocol and make it unusable for any system but their own. Once I read the second article (and found the single reference to "OpenID 2.0")... well, I completely agree with the parent's comments.
One last thing I have heard quite a few others praise is the ability to open almost any document file type out there right out of the box, now that OOo 3.0 has Office 2007 XML support.
I've been able to open Office '07 documents (*.docx) for a while, and I'm using OOo 2.4. There's an upgrade that makes this work better?
SIDE NOTE: has anybody else expanded the contents of a docx file and wondered why it's so convoluted?
Okay, so a developer using GIT has 10 branches working on a couple of different features and bugfixes. They're long-running works, so that developer hasn't put them into the main repository yet... all of a sudden, that developer disappears: they were involved in a car accident & killed, their laptop--which contained the branches--was mercilessly burned into a pile of smoldering ashes in the process.
Now we've got a company with X developers that were waiting on those changes. They'll never see the changes, because they were destroyed along with the person that was about to implement them. So now another developer (or a few) has to devote time to fixing something that was never documented; they may not even have any idea how the afore-mentioned dead developer found the problem or how to fix it without crippling the entire program...
That seems bad. If they'd been using SVN with their own branch(es), there would be a history of it. The remaining developers could just go through and finish up those branches & merge them into the main line. Seems to me like SVN is better, but maybe I'm wrong (I've only ever used CVS & SVN, upgrading to the latter from the former).
This may already have been said (since I came into this article late), but I have to add my two cents:
Wouldn't it be better to have every check-in be stable?
Have you ever worked on a large software project? Sorry if it sounds offensive, but really, it is unlikely at best to make sure that a given commit (or "check-in") is entirely stable. The likelihood decreases exponentially as the number of target systems increases (especially if it includes both Windows and Linux).
When I setup a source control system, I use it as much for the ability to keep track of my changes as I do for maintaining the code. I've been known on occasion to make a commit that I know beforehand will destabilize the product, and then revert those changes immediately in the next commit; the log often includes "I know this is broken" and something about needing to have the change in there for historical reasons.
If there's lots of developers on a project and the leader is anal about having every commit as stable as possible, I suggest SVK. It can mirror a repository locally, and all changes can be committed whenever needed without breaking the repository (especially handy when the repo is setup so that there's no chance of branching).
Riiight... this is similar to the Prohibition (alcohol) and the steep punishments for drugs. Except instead of lessening the problem, it will simply lessen the number of people sending the stuff.
Since there will eventually be fewer willing to send SPAM, those willing to tempt fate will get paid many times more than previously, and will employ even more technology to make tracing them to their real location even more infeasible. Instead of allowing mail to be sent from their own mail servers, it will be based purely off botnets--meaning the price of botnets will skyrocket. In turn, this means the price of black market exploits will also skyrocket... this will increase the number of infected PC's on the Internet.
Now "small" botnets will number in the hundreds of thousands. Most Windows PCs will have been exploited, with each system potentially in multiple nets. More people will bitch that their systems are slow.
Here's where the geeks step in. "Hey, it'll run way faster if you just put Linux on there. Then you don't have to worry about being a part of their botnet."
Thus ushers in the year of Linux on the desktop. Let's do it.
Or the botnet is setup to continuously spew out the same email until "told" otherwise. Since there's nobody to tell it to stop, it just keeps spewing the same stuff over & over again. I mean, we are talking about spam here (most of it isn't even coherent).
Why go through a web hosting company? It's pretty easy to host your own website:
I don't use DynDNS nor a static IP, and I host my own websites. I didn't bother setting up my own mailserver, as Google Apps more than fits the bill for that. So what's the problem?
Your farm, ranch, house, whatever, won't last long against a single blast from a tank. And how many ordinary citizens have access to anti-tank weapons or know the first thing about defending against bombers or attack choppers?
The problem isn't achieving the know-how to handle the armored might of the U.S. military. The problem isn't even with protecting your property against attacks from them. The problem is getting a large enough squad together to make an impact, and convincing the populace that we're fighting is for the greater good (so if a "freedom fighter" needs someplace to hide, they don't have to be afraid of their neighbors).
Guerrilla tactics aren't that hard to learn. Go on the Internet and ask Google. Homemade bombs aren't even that hard to make, provided one can get their hands on enough of the right fertilizer. But in a game of attrition, the U.S. military will win hands-down.
But let's say, just for the sake of saying it, that a large enough force has been built up to make an impact. An impact large enough for the Government, even the President, to pause. What now? Queue government-funded media spin on the "War on Terror". Now what? Storm the White House? Riiight...
The plain fact of the matter is that the Government would have to be completely overthrown in order to affect the necessary changes. And in doing so, we'd open the country up to invasion by any and all of its enemies.
Moral of the story? Subtlety doesn't work, and overthrowing the Government leaves us too open to outside attack/invasion.
Wow. That is truly an awesome observation. By their own statistics, they have no reason to use DRM. Now the question is: "why do you include DRM in your games when so few people apparently care about it?" Then the other question is, "if piracy of your game has damaged sales so much that you were 'forced' to add DRM, have you actually seen a marked improvement in sales and a decline in piracy?" (the former being far more important that the latter for determination of continued usage of DRM)
With a proper "checks and balances" system, the power to destroy evidence or alter the results can be seemlessly removed from the hands of a few:
1.) All information is encrypted in a way that it cannot be locally decrypted
2.) no information is stored locally: all responses are *immediately* sent to a remote server
3.) information is stored in multiple centralized servers: data must be verified as matching on all servers or it is either a.) flagged as tampered, or b.) discarded entirely (NOTE: discarding a vote entirely could easily be used to rig an election simply by tampering with enough votes from "the other guy"). Alerts are sent to multiple people when tampering is found, so no one person can be blamed if nothing is done.
In any given situation, given enough cooperation from enough people in power, any security system can be trumped or hacked. With enough restrictions and forethought, a system can be secured enough that there would simply be too many people that would have to cooperate and too many systems that would have to be simultaneously altered, making a hack less likely. By having enough redundant systems and proper encryption, the amount of effort involved in breaking or altering a system far outweighs any payouts... and the greater number of people that are involved, the greater chance that it won't work or that it will be exposed.
Geek: "Stop turning off the TV!"
Wife: "Why?"
Geek: "I told you this a hundred times. My wireless signal comes off the light waves from the TV."
Wife: (turns TV back on) "There, are you happy?"
Geek: "I told you this a hundred times. My wireless signal comes off the..."
Wife: "YOUR STUPID TV IS ON!"
Geek: "Yeah, but YOU'RE STANDING IN FRONT OF IT!"
Amazingly insightful. If I had mod points--and you hadn't posted as AC--I would have modded you up. Another problem is that the technology would take decades to be useful, and then it couldn't be completely trusted. Reliability would depend on the idea that every car on the road had that same technology... imagine what would happen if the technology actually was put into every car, and then a car show came into town...
"Trust issues." Seriously, do you think any corporation would think twice about this, considering a man nearly took an entire plane down with a click from his wireless mouse?
Na. Use a picture or a small clip from "Two Girls and a Milkshake" video (just the first 30 seconds should do). Guaranteed you'll get your laptop back in 30 seconds or less... well, if they don't drop it first.
Here's a hypothetical to think about next time you have to set one of those "security questions": WebsiteA asks you to choose a password that, at a minimum:
* 8 characters in length
* 1 capital letter
* 1 lowercase letter
* 1 number
* 1 non-alphanumeric (special character)
* can't have any variation of a word associate with your name, D.O.B., home city, etc
* no dictionary words
Now they ask you a security question, "What's your mother's first name?"
Now a determined "hacker" goes and tries to brute-force your password. They make a couple of valiant efforts to brute force your password before realizing it's about to get locked-out from too many wrong passwords (or they don't bother, since it's infeasible to answer in just 2 attempts).
Then they click the "Forgot password?" link. Using the information they pulled out of your trash, they easily defeat all that security and reset your password. Seconds after logging in, they reset your security question/password as well, effectively locking you out of your account.
Oh, and if I remember correctly, most of those websites that lock your account after X failed attempts at password have no limit for responses to security questions. So why bother brute-forcing the password when they can do so to your security question which is invariably based on a real word (and orders of magnitude simpler to guess by a computer)?
Now I ask, why would I EVER give real answers to the security questions?
(Sets a laptop down on the desk, types a few sequences of keys to break into the school's "secure" wireless network)
"Just remember kids: everything is connected to the Internet. So my job is to keep this...
(a few more keystrokes. Lights go off, and a massive "Pewwwwwww" sound of systems shutting down is heard)
"...from happening. I'll take that free lunch now, Mrs. Phelps."
I can see it now:
GEEK: the ultimate prophylactic
I absolutely agree, especially when it comes to repetitive jobs. Humans get bored and lose interest in repetitive things quickly; computers, on the other hand, are infinitely patient and don't take shortcuts after too long (well, unless they've been programmed that way/poorly... remember how that one version of Windows deleted command.com if the system was on for 90 consecutive days?)
...even when the vehicle has run out of gas.
I'm pretty sure the main source of an explosion isn't the gasoline, but the fumes it puts off. Drop a match on the ground and throw gasoline on it (don't pour beside). Now throw another match into that "empty" container. Which one makes the bigger boom? Maybe I'm wrong... (looks like I'll have to test that tonight).
So much talk about bears and wild animals thinking the wallet might be tasty, or a bag... what about the body? Seems like, after this much time, the dude's dead. The animals probably thought he was a mighty tasty treat... and the damn "dogs" buried his bones for a nice snack later.
NOTE TO SELF: remember to reschedule that class on sensitivity.