I find general comments covering chunks of code to be useful when skimming through code, but agree that a lot of comments are redundant in well written code.
I definitely agree that higher level documentation is where it counts. I can read through a class and figure out what it does pretty quickly.. but figuring out how 100 classes interact is very time consuming. A quick little diagram can be invaluable.
As I understand it, their system is mostly a failure due to the cameras not really being centrally connected and matching video footage a nightmare to the point where cops don't bother. They put the cameras up thinking it would deter crime with no thought to how the video would actually be used.. but as such the cameras were so ineffective that people assume they arn't even being monitored and thus they don't deter anything.
Last I heard, the success rate was like 3% or something like that.
Mmm.. was curious after posting this and did some googling. Looks like this might not be the case everywhere. Some places they need a photo of the driver as well... which I guess would solve the "friend borrowed it" problem.
It's a true point but I think it might not be as relevant. This issue also exists with the currently in use redlight systems I mentioned.
As I understand it, even if someone has borrowed your car, legally you are still liable for anything that happens while that vehicle is on the road. It's one of the risks you take when lending someone your car. Ultimately most people you lend your car to are hopefully going going to reimburse you for a ticket they incur.
Fun fact: as I understand it, if you borrow someones car and it gets taken by the cops under the ever popular "involved in a crime == ours now!" laws, I don't think the actual owner can get it back.
Obviously we are in the theoretical here, discussing a perfect untrickable system with massive processing capabilities and ignoring the inevitable abuse that would make it impossible to implement in a useful manner.
The driving thing I see as much closer to reality. It's hard to disguise a car.. they all have a unique id plastered on them, and removing it or obscuring it is a crime in itself.. and that id is already tied to the driver. It's probably not an easy thing to do, but I imagine software could be written right now that could identify most common driving violations. You don't need to track the car, just issue the ticket automatically, same as done with some redlight camera systems.
We're obviously talking about emerging technology here and this is largely a theoretical exercise (we're also ignoring the more relevant abuse that makes this system next to impossible to actually pull off in a benificial way).
Maybe the system tracks all people all the time, and flags someone coming out of a building that it doesn't remember having gone in. Maybe it doesn't even use visuals.. maybe it uses a gravity displacement signature or some other startrek style mumbo-jumbo.
This is true, especially in the driving example. Even good drivers make mistakes and get away with them.
I suspect the system of punishment would have to be changed if this ever came to fruition. Right now it assumes your chance of getting caught is low, so the penalty is increased to even things out. You break 30 or so traffic laws.. by random chance a cop sees one of those.
Maybe crimes would need to be grouped with a certain number of freebies. Maybe you can get away with 3 "minor road violations" before getting dinged.
Maybe certain crimes that are there purely as a fallback need to be re-written or eliminated to deal with an environment where they are consistently applied (rather than being a tool the cops can bring out when needed). That might actually be better. Personally I don't like the fact that we are at the mercy of the police to not charge us with all the technically illegal stuff everyone does. Crazy subjective stuff should be refined (i.e. the difference between theft (a misdemeanor) and burglary (a felony and possible 10 years in prison).
They are closer to it, but it's the computer detecting every crime and automatic tracking that make it work. Commit a crime anywhere and the system flags it and tracks you everywhere.. eventually the police get around to picking you up. Anywhere, and every time.
Making commiting a crime an almost guarenteed arrest I imagine would prevent crime. Right now committing a crime is luck based, and sadly you have good odds. Making committing a crime the equivilant of automatic prison would be a hell of a deterant.
It's the lack of random chance that makes this effective however.
No more "did a cop see it".. or "is anyone looking". You simply get caught every single time (well, there is always going to be errors, but not enough that a criminal might be tempted to try their luck). Right now if you commit a crime, you have a good chance of getting away with it, which is probably why there are so many criminals. I imagine the crime numbers would drop if commiting a crime was an automatic arrest, every single time.
Ignoring evil government spying and abuse, and just focusing on the standard issue crime we all know and hate...
We are now near a point where we could use technology to very effectively cut down crime. The issue is no longer technological but social.
As you said, imagine a camera on every street corner. Imagine a system that constantly monitored every road for bad driving and issued immediate tickets. Cut someone off.. drive too fast.. forget your turn signal.. instant ticket. Imagine how much that would improve safety on the roads. Bad drivers would either improve or driving would become so expensive that they'd give it up.
Go forward a bit, imagine a system that can automatically detect crime. Imagine literally not being able to rob someone.. or steal anything.. because a system would immediately identify the action, and track you wherever you went until the police picked you up making it virtually impossible to escape. Imagine how much crime that would cut down on.
All at the expense of having very little privacy, and of course opening the door for massive abuse.
Do you want to live in that world? Personally I don't think I would either. Do we want to or can we find a middle ground?
That's all fine and dandy, unless of course you are living in reality.
One of the chief reasons IE6 persisted so long is the extreme prevalence of amazingly terrible “wow, we pay what for this” web apps at the heart of so many businesses. Fixing the problem of constant major browser versions is a lot simpler than fixing the "we'll take the cheapest option you have" business mentality problem.
I mostly agree, but I still think it will be a great product for many geeks (including myself).
It'll be interesting to see how long the market lasts however. Once everyone who wants one has gotten one (or in my case, probably many) I have to wonder where they will get their continued sales from. Although I guess you can say the same about most markets.
While I honestly don't see the whole revolutionizing computer education thing.. I still think this is going to be an awesome board with all kinds of uses. I can't wait to get my hands on one (or inevitably many) of these.
See that's a programmers opinion (and one that I would also intuitively come up with if asked) and probably why this is a problem.
You want a good ebook you don't get a programmer who views the book as text data, you want an artist who things the line spacing adds to the character of the book and such.
While I don't think every website is golden (or even most) right now.. I definitely see this trend.
There was a time not too long ago where people would get the "office geek" (aka the guy who knew just a little bit more about tech than everyone else) to throw a website together on their ISP supplied webspace.
At least now most companies hire someone to do their site.. even if they hire someone cheap it's usually better than Rob down in shipping hammering something out with netscape composer.
Readers are constrained by how well the protocol they read has been used.
As parent said, the format may have the utility for an author to provide information about sections and let the reader make it look good on a specific device, but just as people did when using netscape composer (and lets pause for a few minutes there and reflect.. ok.. good) a lot of people authoring the documents don't use them and instead use what they know (space bar, enter key) to make it look right on whatever they are using (which may not be their fault... it wouldn't occur intuitively to a non-geek... hell it can be hard to explain to a geek!).
I really don't know if this is a good thing. While I think I would have loved the idea while I was in school, looking back I think I would have missed out on a lot of social interaction that was probably really important.
If left to my own devices, I would have spent every hour of my free time on a computer. Luckily I had friends who dragged me to various things.. and begrudgingly I actually had a lot of fun.
In other words, I think education is only part of the education process. Social development is the other big part. Technical skills are great, but in todays work environment everything is team driven and being able to get along with people is almost (or even more) important than being able to crank out killer code.
Well that's a reasonably stupid idea. Store the password with something many users are going to carry around with their laptop...
And even if you didn't.. you forget your password on the road, then what? And this is less annoying than having to answer a previously entered question?
The problem is that it's well within the content providers ability to block legitimate sites from aggregating their content (see: robots.txt). They don't want to do that. Instead they want the benifit, _and_ want those sites to pay them.
Never subscribed to the whole "if I'm that bad, pull the plug" stuff.
Maybe I'd feel differently if I'm ever actually in that situation.. but my gut reaction has always been as George Carlin would put it "screw that, save my ass!".
I learnt on basic on a dragon32, later migrated to QBasic, then turbo c and so on until becoming what I consider to be an at least average programmer (at the very least good enough that someone continues to pay me..).
The key is migrating away from it as you advance. Much like anything, a large part of the problem around BASIC is people refusing to adapt to the proper tools and trying to use the tools they are familiar with for things that are out of scope.
I guess the question is whether BASIC is still an ok learning language. It's no longer a choice between basic or asm. There are lots of good intermediary languages that you can learn on and then keep right on using as you advance.
Wern't they the ones who used to have the "check if this domain is available" box that would then register the name and try to sell it to you at a higher price?
Slashdot Mirror
I find general comments covering chunks of code to be useful when skimming through code, but agree that a lot of comments are redundant in well written code.
I definitely agree that higher level documentation is where it counts. I can read through a class and figure out what it does pretty quickly.. but figuring out how 100 classes interact is very time consuming. A quick little diagram can be invaluable.
As I understand it, their system is mostly a failure due to the cameras not really being centrally connected and matching video footage a nightmare to the point where cops don't bother. They put the cameras up thinking it would deter crime with no thought to how the video would actually be used.. but as such the cameras were so ineffective that people assume they arn't even being monitored and thus they don't deter anything.
Last I heard, the success rate was like 3% or something like that.
Mmm.. was curious after posting this and did some googling. Looks like this might not be the case everywhere. Some places they need a photo of the driver as well... which I guess would solve the "friend borrowed it" problem.
Learn something every day :)
It's a true point but I think it might not be as relevant. This issue also exists with the currently in use redlight systems I mentioned.
As I understand it, even if someone has borrowed your car, legally you are still liable for anything that happens while that vehicle is on the road. It's one of the risks you take when lending someone your car. Ultimately most people you lend your car to are hopefully going going to reimburse you for a ticket they incur.
Fun fact: as I understand it, if you borrow someones car and it gets taken by the cops under the ever popular "involved in a crime == ours now!" laws, I don't think the actual owner can get it back.
Obviously we are in the theoretical here, discussing a perfect untrickable system with massive processing capabilities and ignoring the inevitable abuse that would make it impossible to implement in a useful manner.
The driving thing I see as much closer to reality. It's hard to disguise a car .. they all have a unique id plastered on them, and removing it or obscuring it is a crime in itself.. and that id is already tied to the driver. It's probably not an easy thing to do, but I imagine software could be written right now that could identify most common driving violations. You don't need to track the car, just issue the ticket automatically, same as done with some redlight camera systems.
We're obviously talking about emerging technology here and this is largely a theoretical exercise (we're also ignoring the more relevant abuse that makes this system next to impossible to actually pull off in a benificial way).
Maybe the system tracks all people all the time, and flags someone coming out of a building that it doesn't remember having gone in. Maybe it doesn't even use visuals.. maybe it uses a gravity displacement signature or some other startrek style mumbo-jumbo.
This is true, especially in the driving example. Even good drivers make mistakes and get away with them.
I suspect the system of punishment would have to be changed if this ever came to fruition. Right now it assumes your chance of getting caught is low, so the penalty is increased to even things out. You break 30 or so traffic laws.. by random chance a cop sees one of those.
Maybe crimes would need to be grouped with a certain number of freebies. Maybe you can get away with 3 "minor road violations" before getting dinged.
Maybe certain crimes that are there purely as a fallback need to be re-written or eliminated to deal with an environment where they are consistently applied (rather than being a tool the cops can bring out when needed). That might actually be better. Personally I don't like the fact that we are at the mercy of the police to not charge us with all the technically illegal stuff everyone does. Crazy subjective stuff should be refined (i.e. the difference between theft (a misdemeanor) and burglary (a felony and possible 10 years in prison).
They haven't implemented this.
They are closer to it, but it's the computer detecting every crime and automatic tracking that make it work. Commit a crime anywhere and the system flags it and tracks you everywhere.. eventually the police get around to picking you up. Anywhere, and every time.
Making commiting a crime an almost guarenteed arrest I imagine would prevent crime. Right now committing a crime is luck based, and sadly you have good odds. Making committing a crime the equivilant of automatic prison would be a hell of a deterant.
It's the lack of random chance that makes this effective however.
No more "did a cop see it" .. or "is anyone looking". You simply get caught every single time (well, there is always going to be errors, but not enough that a criminal might be tempted to try their luck). Right now if you commit a crime, you have a good chance of getting away with it, which is probably why there are so many criminals. I imagine the crime numbers would drop if commiting a crime was an automatic arrest, every single time.
I actually find this subject interesting.
Ignoring evil government spying and abuse, and just focusing on the standard issue crime we all know and hate...
We are now near a point where we could use technology to very effectively cut down crime. The issue is no longer technological but social.
As you said, imagine a camera on every street corner. Imagine a system that constantly monitored every road for bad driving and issued immediate tickets. Cut someone off.. drive too fast.. forget your turn signal.. instant ticket. Imagine how much that would improve safety on the roads. Bad drivers would either improve or driving would become so expensive that they'd give it up.
Go forward a bit, imagine a system that can automatically detect crime. Imagine literally not being able to rob someone.. or steal anything.. because a system would immediately identify the action, and track you wherever you went until the police picked you up making it virtually impossible to escape. Imagine how much crime that would cut down on.
All at the expense of having very little privacy, and of course opening the door for massive abuse.
Do you want to live in that world? Personally I don't think I would either. Do we want to or can we find a middle ground?
That's all fine and dandy, unless of course you are living in reality.
One of the chief reasons IE6 persisted so long is the extreme prevalence of amazingly terrible “wow, we pay what for this” web apps at the heart of so many businesses. Fixing the problem of constant major browser versions is a lot simpler than fixing the "we'll take the cheapest option you have" business mentality problem.
I mostly agree, but I still think it will be a great product for many geeks (including myself).
It'll be interesting to see how long the market lasts however. Once everyone who wants one has gotten one (or in my case, probably many) I have to wonder where they will get their continued sales from. Although I guess you can say the same about most markets.
While I honestly don't see the whole revolutionizing computer education thing .. I still think this is going to be an awesome board with all kinds of uses. I can't wait to get my hands on one (or inevitably many) of these.
See that's a programmers opinion (and one that I would also intuitively come up with if asked) and probably why this is a problem.
You want a good ebook you don't get a programmer who views the book as text data, you want an artist who things the line spacing adds to the character of the book and such.
While I don't think every website is golden (or even most) right now.. I definitely see this trend.
There was a time not too long ago where people would get the "office geek" (aka the guy who knew just a little bit more about tech than everyone else) to throw a website together on their ISP supplied webspace.
At least now most companies hire someone to do their site.. even if they hire someone cheap it's usually better than Rob down in shipping hammering something out with netscape composer.
Readers are constrained by how well the protocol they read has been used.
As parent said, the format may have the utility for an author to provide information about sections and let the reader make it look good on a specific device, but just as people did when using netscape composer (and lets pause for a few minutes there and reflect.. ok.. good) a lot of people authoring the documents don't use them and instead use what they know (space bar, enter key) to make it look right on whatever they are using (which may not be their fault... it wouldn't occur intuitively to a non-geek... hell it can be hard to explain to a geek!).
I really don't know if this is a good thing. While I think I would have loved the idea while I was in school, looking back I think I would have missed out on a lot of social interaction that was probably really important.
If left to my own devices, I would have spent every hour of my free time on a computer. Luckily I had friends who dragged me to various things.. and begrudgingly I actually had a lot of fun.
In other words, I think education is only part of the education process. Social development is the other big part. Technical skills are great, but in todays work environment everything is team driven and being able to get along with people is almost (or even more) important than being able to crank out killer code.
Well that's a reasonably stupid idea. Store the password with something many users are going to carry around with their laptop...
And even if you didn't.. you forget your password on the road, then what? And this is less annoying than having to answer a previously entered question?
The problem is that it's well within the content providers ability to block legitimate sites from aggregating their content (see: robots.txt). They don't want to do that. Instead they want the benifit, _and_ want those sites to pay them.
Never subscribed to the whole "if I'm that bad, pull the plug" stuff.
Maybe I'd feel differently if I'm ever actually in that situation.. but my gut reaction has always been as George Carlin would put it "screw that, save my ass!".
Indeed.
I learnt on basic on a dragon32, later migrated to QBasic, then turbo c and so on until becoming what I consider to be an at least average programmer (at the very least good enough that someone continues to pay me..).
The key is migrating away from it as you advance. Much like anything, a large part of the problem around BASIC is people refusing to adapt to the proper tools and trying to use the tools they are familiar with for things that are out of scope.
I guess the question is whether BASIC is still an ok learning language. It's no longer a choice between basic or asm. There are lots of good intermediary languages that you can learn on and then keep right on using as you advance.
Wern't they the ones who used to have the "check if this domain is available" box that would then register the name and try to sell it to you at a higher price?
Yup!
I don't agree with everything the EFF says by any stretch... but compared to the other extremes... they get some of my money!
By default, nmap only scans a subset of ports (first 1000 of all protocols or something).
Try explicitly telling it to scan that port (using the -p option)
Yeah I realized that like 10 seconds after I posted. Knee jerk reaction. Slow brain day :(