Sure, oh yes. Site's content is obviously a copyrighted material, and site's defences are to protect this material. Which makes Microsoft a company that produces technology and tools to circumvent the copyright protection. I'm holding my breath to see Ballmer arrested by FBI agents next time he goes out of Microsoft headquarters.
MSIE and Mozilla are only browsers really supporting Internet standards to some measure. Opera's working in non-Latin languages is crap (yes, Virginia, there are countries besides US and some of the still use their crappy languages and alphabets instead of using a human language like Americans do). They drop the ball here and there on CSS and HTML 4 ("we don't support this because we couldn't do it for release time").
So here I have a choice - use small fast browser that shows pages wrong or doesn't show them at all - or use bigger fatter one that does show them. I would usually choose the latter.
In other words, he was actually making a contribution to the field of encryption.
Do you think FBI agents really evaluated if Sklyarov did contribute to the field of encryption or not before arresting him? Are there many cryptography experts in that division, how do you think?
Elcomsoft's tool is to eBooks what DeCSS is to DVDs.
If one has problem with Elcomsoft, one goes to court and sues Elcomsoft. I have little sympaty to Elcomsoft as a company. But I bet you don't want to be arrested if you go to, say, Brazil and your company sales policy has some problem with some obscure Brazilian law. You probably would be outraged if you would be arrested on this basis.
Elcomsoft even advertised it as such.
Elcomsoft web pages are public. Can you quote me the place where the do it (URL, place there)? If not, please look in the mirror and say to yourself "I just lied. I am a liar. God forgive me".
If somebody had cracked CSS and made a non-licensed Linux DVD player available without providing source code or algorithms of the DeCSS part, it would have been legal under the reverse engineering clause of the DMCA.
If only they had time out of the courts to even start working on such a player, you know... In fact, as we see, one has pretty good chances to get jailed very fast after The Man knows about your research, not allowing you to produce evidence of "good faith". And, as you stated by yourself, nobody is going to believe a cryptoanalyst punk of yours that you can do this for something good.
On the other hand, one could work on such a software on the prison library computer. If one doesn't get Mitnick-like "not to come within a mile radius of any computer" judgement from his DMCA trial, of course. Then you probably could have you sentence cut or your fine to get lower, maybe?
Can you bring one link with evidence telling Sklyarov personally sold Elcomsoft products on the conference? Which titles? Anyone who saw anyone buying it?
Also, as you stated by yourself, Adobe and FBI could not know before the conference that Sklyarov is going or not going to sell anything, so obviously they were planning to arrest them for something other. The "other", as you generously quoted, is "imported and offered to the public" (via bringing in and reading his conference speach) "technology... designed or produced for the purpose of circumvention a technological measure". Note here that almost any cryptoanalyst work is such "technology", and as such bringing it in the US and publicising is now illegal. Q.E.D. - DMCA outlaws cryptographics research.
He was arrested for selling it to customers in the US over the internet and for selling in person while in the US at the conference.
That's bull. Sklyarov didn't sell anything. Elcomsoft (which Sklyarov neither owns nor manages) sells the program. Sklyarov was arrested because his copyright was on the program (i.e., for writing the program). Also, this is the first time I hear Sklyarov sold anything on the DEFCON. Did he go and sell diskettes? Or did he have Visa card device in his pockets? How exactly did he sell this program?
Also, how do you imagine some company in Russia be liable for what their clients did - importing the DMCA-prohibited wares into the USA? Elcomsoft itself is in Russia, and whatever it sells in Russia to US citizen is under Russian law, not under US law.
Also, if Some Big US Company has problem with Some Another Big US Company does, do we see employees of either company arrested by the FBI agents and rotting behind the bars? No, we see their lawyers talking. Somehow, the civilized way of settling (i.e., bringing the suit against Elcomsoft, winning it and prohibiting Elcomsoft to sell the wares) was seen unfit by Adobe and US State. They chose just to jail the first person they could lay their hands on. Very, very stupid. And very, very disgusting.
The fact that the RIAA threatened a lawsuit against Felten doesn't mean he did anything wrong.
If I threaten to kill somebody, is it wrong? Even if I do not actually do it? A lawsuit from RIAA would basically kill Felten as a researcher and maybe would also kill him economically, depending on how much money can he shell out to protect himself from frivolous lawsuits of RIAA. And you definitely know that, that being engaged in the lawsuit is a gross problem even if you win. So why you pretend like it's nothing?
I don't believe that anybody would consider a straightforward attack against a known encryption technique for the purposes of key discovery to be "good faith" research.
So what is "good faith"? Writing positive reviews in Byte? One doesn't need cryptoanalysts for this, trained monkey is enough. Cryptoanalysts work is to break the codes. Claiming otherwise would be like holding QA people liable for "breaking" the code when they find bugs. I guess if programmers were on the same dumbness and greed and power level as the RIAA, finding bugs in products or telling anyone about how some program crashed for you would be as illegal as telling someone how some encryption algorithm sucks is now.
To my knowledge, Dmitry was the only Elcomsoft employee there.
Oh. "We cannot arrest the right person, so we better arrest the innocent - at least thus we get something to report to big boys up there".
Well, then one knows what to do - don't buy such a CD. If you come to a CD that has a writing on it saying "I'm defective" in bold friendly letters - why would you buy it? Just to tell Sony it can sell you a single left shoe for the price of the pair?
Actually, 0.9 was not so very good. Quite a lot of regression bugs, including image blocking stopping to work, password management dialog broken, theme/UI problems, etc. The speed improvement was good, but the UI bugs were very unpleasant. Most of these are resting in RESOLVED/FIXED now in bugzilla, so I guess 0.9.1 is going to be yet better, unless, of course, more regressions will happen.
Now I cannot distribute work derived from code under many licenses, including GPL, unless I am agreeing to some very special conditions. So adding yet another one doesn't change anything in my stand - I still need a lawyer to understand each one of them. Unless all licensing stuff is scrapped altogether, it is going to be so forever.
Re:Well, how about not using email?
on
RFC for Spammers
·
· Score: 1
The best response for this would be for the ISP to write in bold firendly letters in their introductory account contract: "If you use this for spamming, we are going to charge you $1000". Then this way of spamming starts to be not-so-profitable.
What's bad in having 23, or, for that matter, 230 licenses? So RMS doesn't hold a monopoly on determining what open source is and what it is not, big deal. If we don't need software monopoly, why do we need license monopoly?
Attaracted by the praises sung to the Konq, I decided to use it myself. So I installed and run it. OK, first of all it brings half the KDE with it, which is very nice if I wanted it. But I *do not*! I don't want the freaking KDE, I just want to see webpages! OK, well, I have enough RAM to waste, so let it be. So I opened one of my beloved pages. And here were the point when things got worse - it showed as the row of "????"s. Yes, it was the page on Russian. There are not only US in the world. And yes, Mozilla showed this page perfectly well without needing any explanations. OK, I'm not lazy, I set the charset manually, though Konq could get it from the HTTP headers. I see the font are _ugly_. Goddamn ugly. Why again Mozilla can show it without me telling it and Konq does not?
OK, now I got page in bearable form. Now I want to get rid of those banners. Oh, no such function. Never mind. I like the site still. I bookmark it. Bookmark comes out as "????" again! Guys, ever heard of Unicode??? Ever imagined page ttles could be not in English only? OK, down with those non-Americans, the should learn to speak human language anyway. Let's continue our journey.
OK, another site - in english now. DHTML menues - do not work. It worked in Mozilla, Netscape, IE. I think I know who is not DOM-compliant here. Never mind, DHTML is for sissies. Let's try to login. Boo! "The processor for https://www.site.com protocol dies unexpectedly". I remember such phrasing from my Windows days. "Something is wrong and we don't have a damn idea what's up and now you are on your own". Well, back to Mozilla. Konq is fast, but what the use of doing the worng thing fast? I prefer doing the right thing slowly. See you in a year, Konq team.
Why not create RPM talkback builds for Linux? Installing RPM build is a lot easier, requires almost no manual work and will increase a number of users reporting bugs significantly.
The solution to this problem is simple. There should be a "corporate" top-level domain, with *very high* (like tens thousands of bucks) entry fee, which is pre-paid and trademark-checked on entrance, and there should be "small business" top-level domain, which would be everybody's else playground. Probably, there should be more levels, with prices and measure of protection varying.
The reason is simple - if you are SGI and have sgi.corp, you don't care for sgi.hobby if it exist or not - everybody who is looking for real SGI would look it in the.corp anyway. And if you are a squatter, you need to be a really reach one to pre-pay corporate rate domain only to be evicted for TM violation. And if you are just a mere online artists happened to have name like Serge G. Irving, you get your sgi.personal and are happy.
Only the commitee thinking of ICANN and their slowness on the border of brain-deadness prevent the world to have domain system fixed forever. one.com for whole world is not enough, and "smartees" in ICANN should have realised this decade ago.
Also, US should stop issuing.com's for both US and foreighn customers. There should be international corporate domains (see above), and per-country small business domains (.com.us,.co.uk,.com.de,.co.il,.com.whatever).
There were numerous times when I was told to get a life, but however hard I try to find the correct keyword for Google to find me one, all in vain. Altavista is not much of help too. Seems there's no such thing on the Internet...
Picture putting Linux on one of your sales force's desk.
If Linux had something analogoues to Word & Outlook (which it probably will in couple of years) - I see no problem with this. I've seen how sales people work, and believe me - they have no single use of the fact that that's windows or shmindows or whatever. They click icons to run certain programs, and do certain things with it. That's not to say that they are stupid - most of them are pretty smart people. They just don't care for the OS - like when you travel by public bus, you usually don't care what company made it, once it delivers you to the destination. They care for the number of simple functions that they need - emails, customer tracking, reporting, billing tracking, etc., etc. The moment programs for Linux exist (and for most purposes, they already are) to do such things, there's not problem to use it for most non-technical person.
They seem to have this stuff pretty OK as of now. At least at the first glance at local bsd.tucows.com mirror, it looks alive and well and not containing all the problems BSD Today complained about. Seems like they hired somebody with real knowledge. Good.
The first thing people should do where it happens is to require that every disk that has this "protection" should be clearly maked as being so, so that you won't accidentally buy the "protected" disk for the full price and then discover you cannot use it for half the things you are used with regular disks. If they think people would love it - they certainly won't be afraid to mark the disks. If they think people won't love it - well, tough luck, they'll have to bear the consequences.
Here I mus note that if they would spend the same thousands on getting code right and QAed on the first run, they probably would not need so much updates...
Your problems probably were not from using "wrong" eye, but from using only one of them. Human's feeling of distance - and thus velocity, etc. - is mostly based on binocular vision. The distance is determined or relatively to known-distance objects (so if something is placed before something else, you subconsciously know it's closer to you), or by difference in the angle when you see the thing by right and left eye. Usually brain combines both inputs and derives some conclusion from it. Now, if you look only with one eye, the brain has no "binocular" input, so it has to rely only on the reference to known objects. If it has no information on this (like, you concentrate you vision on the car moving before you and brain has no reference point to compare to it like other cars or border), it becomes pretty confused. As you saw, restoring binocular vision helps it to make right decision. Quality of the picture plays not the main role here - once the brain can recognize and identify the object, even with low resolution, it can take the difference between eye angles and so know the distance.
Very nice, but not for production yet. First thing it wanted when I tried to upgrade minor package on RedHat is to remove all KDE and Mesa installation altogether, without explaining reasons, and there were no way to make it to download that package (which has no relation neither to KDE not to Mesa - it was bind-util) and install it without blowing out half of my system.
So, I know it's beta and it's expected - but don't try it to manage your production sites yet, or you'll be sorry. I hope they'll fix all the things, since the tool is looking very nice and promising and as soon as they will exorcize all devils from all the small things that would be the killer tool.
XML-RPC is many things, but calling "lightweight" protocol that converts for numeric bytes into 64 textual representation bytes (not counting the whitespace) and adds XML parser pass on each function call - is an exaggeration if I ever saw it. It's definitely not lightweight protocol. Oh, and I forgot - that's an HTTP layer on top of it. Now think about you needing just to call one of your programs from another and pass a number - would you think of a protocol that adds two pretty slow layers of abstraction on top of it, one making message 8 times as big and other further duplicating its size? I would not.
Oh. So that's the cause for such laws. Now I see.
Is smoking dope mandatory for legislative bodies' members or just a traditional habit?
There's a lot of articles about "smoking dope makes your worse driver/student/whatever". Now we can make a case about "smoking dope makes you stupid legislator" too.
Well, AFAIK you can get internet access for free now. Or if you can't, most ISPs have plans for some $5/month access. Average kid spends more on chewing gum, I guess.
Slashdot Mirror
Sure, oh yes. Site's content is obviously a copyrighted material, and site's defences are to protect this material. Which makes Microsoft a company that produces technology and tools to circumvent the copyright protection. I'm holding my breath to see Ballmer arrested by FBI agents next time he goes out of Microsoft headquarters.
MSIE and Mozilla are only browsers really supporting Internet standards to some measure. Opera's working in non-Latin languages is crap (yes, Virginia, there are countries besides US and some of the still use their crappy languages and alphabets instead of using a human language like Americans do). They drop the ball here and there on CSS and HTML 4 ("we don't support this because we couldn't do it for release time").
So here I have a choice - use small fast browser that shows pages wrong or doesn't show them at all - or use bigger fatter one that does show them. I would usually choose the latter.
but rather has made the trafficking in devices using that knowledge illegal.
Devices or technology. And any scientific paper describing an attack can be considered technology to perform it, ergo...
In other words, he was actually making a contribution to the field of encryption.
Do you think FBI agents really evaluated if Sklyarov did contribute to the field of encryption or not before arresting him? Are there many cryptography experts in that division, how do you think?
Elcomsoft's tool is to eBooks what DeCSS is to DVDs.
If one has problem with Elcomsoft, one goes to court and sues Elcomsoft. I have little sympaty to Elcomsoft as a company. But I bet you don't want to be arrested if you go to, say, Brazil and your company sales policy has some problem with some obscure Brazilian law. You probably would be outraged if you would be arrested on this basis.
Elcomsoft even advertised it as such.
Elcomsoft web pages are public. Can you quote me the place where the do it (URL, place there)? If not, please look in the mirror and say to yourself "I just lied. I am a liar. God forgive me".
If somebody had cracked CSS and made a non-licensed Linux DVD player available without providing source code or algorithms of the DeCSS part, it would have been legal under the reverse engineering clause of the DMCA.
If only they had time out of the courts to even start working on such a player, you know... In fact, as we see, one has pretty good chances to get jailed very fast after The Man knows about your research, not allowing you to produce evidence of "good faith". And, as you stated by yourself, nobody is going to believe a cryptoanalyst punk of yours that you can do this for something good.
On the other hand, one could work on such a software on the prison library computer. If one doesn't get Mitnick-like "not to come within a mile radius of any computer" judgement from his DMCA trial, of course. Then you probably could have you sentence cut or your fine to get lower, maybe?
Can you bring one link with evidence telling Sklyarov personally sold Elcomsoft products on the conference? Which titles? Anyone who saw anyone buying it?
... designed or produced for the purpose of circumvention a technological measure". Note here that almost any cryptoanalyst work is such "technology", and as such bringing it in the US and publicising is now illegal. Q.E.D. - DMCA outlaws cryptographics research.
Also, as you stated by yourself, Adobe and FBI could not know before the conference that Sklyarov is going or not going to sell anything, so obviously they were planning to arrest them for something other. The "other", as you generously quoted, is "imported and offered to the public" (via bringing in and reading his conference speach) "technology
He was arrested for selling it to customers in the US over the internet and for selling in person while in the US at the conference.
That's bull. Sklyarov didn't sell anything. Elcomsoft (which Sklyarov neither owns nor manages) sells the program. Sklyarov was arrested because his copyright was on the program (i.e., for writing the program). Also, this is the first time I hear Sklyarov sold anything on the DEFCON. Did he go and sell diskettes? Or did he have Visa card device in his pockets? How exactly did he sell this program?
Also, how do you imagine some company in Russia be liable for what their clients did - importing the DMCA-prohibited wares into the USA? Elcomsoft itself is in Russia, and whatever it sells in Russia to US citizen is under Russian law, not under US law.
Also, if Some Big US Company has problem with Some Another Big US Company does, do we see employees of either company arrested by the FBI agents and rotting behind the bars? No, we see their lawyers talking. Somehow, the civilized way of settling (i.e., bringing the suit against Elcomsoft, winning it and prohibiting Elcomsoft to sell the wares) was seen unfit by Adobe and US State. They chose just to jail the first person they could lay their hands on. Very, very stupid. And very, very disgusting.
The fact that the RIAA threatened a lawsuit against Felten doesn't mean he did anything wrong.
If I threaten to kill somebody, is it wrong? Even if I do not actually do it? A lawsuit from RIAA would basically kill Felten as a researcher and maybe would also kill him economically, depending on how much money can he shell out to protect himself from frivolous lawsuits of RIAA. And you definitely know that, that being engaged in the lawsuit is a gross problem even if you win. So why you pretend like it's nothing?
I don't believe that anybody would consider a straightforward attack against a known encryption technique for the purposes of key discovery to be "good faith" research.
So what is "good faith"? Writing positive reviews in Byte? One doesn't need cryptoanalysts for this, trained monkey is enough. Cryptoanalysts work is to break the codes. Claiming otherwise would be like holding QA people liable for "breaking" the code when they find bugs. I guess if programmers were on the same dumbness and greed and power level as the RIAA, finding bugs in products or telling anyone about how some program crashed for you would be as illegal as telling someone how some encryption algorithm sucks is now.
To my knowledge, Dmitry was the only Elcomsoft employee there.
Oh. "We cannot arrest the right person, so we better arrest the innocent - at least thus we get something to report to big boys up there".
Well, then one knows what to do - don't buy such a CD. If you come to a CD that has a writing on it saying "I'm defective" in bold friendly letters - why would you buy it? Just to tell Sony it can sell you a single left shoe for the price of the pair?
Actually, 0.9 was not so very good. Quite a lot of regression bugs, including image blocking stopping to work, password management dialog broken, theme/UI problems, etc. The speed improvement was good, but the UI bugs were very unpleasant. Most of these are resting in RESOLVED/FIXED now in bugzilla, so I guess 0.9.1 is going to be yet better, unless, of course, more regressions will happen.
Now I cannot distribute work derived from code under many licenses, including GPL, unless I am agreeing to some very special conditions. So adding yet another one doesn't change anything in my stand - I still need a lawyer to understand each one of them. Unless all licensing stuff is scrapped altogether, it is going to be so forever.
The best response for this would be for the ISP to write in bold firendly letters in their introductory account contract: "If you use this for spamming, we are going to charge you $1000". Then this way of spamming starts to be not-so-profitable.
What's bad in having 23, or, for that matter, 230 licenses? So RMS doesn't hold a monopoly on determining what open source is and what it is not, big deal. If we don't need software monopoly, why do we need license monopoly?
Attaracted by the praises sung to the Konq, I decided to use it myself. So I installed and run it. OK, first of all it brings half the KDE with it, which is very nice if I wanted it. But I *do not*! I don't want the freaking KDE, I just want to see webpages! OK, well, I have enough RAM to waste, so let it be. So I opened one of my beloved pages. And here were the point when things got worse - it showed as the row of "????"s. Yes, it was the page on Russian. There are not only US in the world. And yes, Mozilla showed this page perfectly well without needing any explanations. OK, I'm not lazy, I set the charset manually, though Konq could get it from the HTTP headers. I see the font are _ugly_. Goddamn ugly. Why again Mozilla can show it without me telling it and Konq does not?
OK, now I got page in bearable form. Now I want to get rid of those banners. Oh, no such function. Never mind. I like the site still. I bookmark it. Bookmark comes out as "????" again! Guys, ever heard of Unicode??? Ever imagined page ttles could be not in English only? OK, down with those non-Americans, the should learn to speak human language anyway. Let's continue our journey.
OK, another site - in english now. DHTML menues - do not work. It worked in Mozilla, Netscape, IE. I think I know who is not DOM-compliant here. Never mind, DHTML is for sissies. Let's try to login. Boo! "The processor for https://www.site.com protocol dies unexpectedly". I remember such phrasing from my Windows days. "Something is wrong and we don't have a damn idea what's up and now you are on your own". Well, back to Mozilla. Konq is fast, but what the use of doing the worng thing fast? I prefer doing the right thing slowly. See you in a year, Konq team.
Why not create RPM talkback builds for Linux? Installing RPM build is a lot easier, requires almost no manual work and will increase a number of users reporting bugs significantly.
The solution to this problem is simple. There should be a "corporate" top-level domain, with *very high* (like tens thousands of bucks) entry fee, which is pre-paid and trademark-checked on entrance, and there should be "small business" top-level domain, which would be everybody's else playground. Probably, there should be more levels, with prices and measure of protection varying. .corp anyway. And if you are a squatter, you need to be a really reach one to pre-pay corporate rate domain only to be evicted for TM violation. And if you are just a mere online artists happened to have name like Serge G. Irving, you get your sgi.personal and are happy.
.com for whole world is not enough, and "smartees" in ICANN should have realised this decade ago.
.com's for both US and foreighn customers. There should be international corporate domains (see above), and per-country small business domains (.com.us, .co.uk, .com.de, .co.il, .com.whatever).
The reason is simple - if you are SGI and have sgi.corp, you don't care for sgi.hobby if it exist or not - everybody who is looking for real SGI would look it in the
Only the commitee thinking of ICANN and their slowness on the border of brain-deadness prevent the world to have domain system fixed forever. one
Also, US should stop issuing
There were numerous times when I was told to get a life, but however hard I try to find the correct keyword for Google to find me one, all in vain. Altavista is not much of help too. Seems there's no such thing on the Internet...
Picture putting Linux on one of your sales force's desk.
If Linux had something analogoues to Word & Outlook (which it probably will in couple of years) - I see no problem with this. I've seen how sales people work, and believe me - they have no single use of the fact that that's windows or shmindows or whatever. They click icons to run certain programs, and do certain things with it. That's not to say that they are stupid - most of them are pretty smart people. They just don't care for the OS - like when you travel by public bus, you usually don't care what company made it, once it delivers you to the destination. They care for the number of simple functions that they need - emails, customer tracking, reporting, billing tracking, etc., etc. The moment programs for Linux exist (and for most purposes, they already are) to do such things, there's not problem to use it for most non-technical person.
They seem to have this stuff pretty OK as of now. At least at the first glance at local bsd.tucows.com mirror, it looks alive and well and not containing all the problems BSD Today complained about. Seems like they hired somebody with real knowledge. Good.
The first thing people should do where it happens is to require that every disk that has this "protection" should be clearly maked as being so, so that you won't accidentally buy the "protected" disk for the full price and then discover you cannot use it for half the things you are used with regular disks. If they think people would love it - they certainly won't be afraid to mark the disks. If they think people won't love it - well, tough luck, they'll have to bear the consequences.
Here I mus note that if they would spend the same thousands on getting code right and QAed on the first run, they probably would not need so much updates...
Your problems probably were not from using "wrong" eye, but from using only one of them. Human's feeling of distance - and thus velocity, etc. - is mostly based on binocular vision. The distance is determined or relatively to known-distance objects (so if something is placed before something else, you subconsciously know it's closer to you), or by difference in the angle when you see the thing by right and left eye. Usually brain combines both inputs and derives some conclusion from it. Now, if you look only with one eye, the brain has no "binocular" input, so it has to rely only on the reference to known objects. If it has no information on this (like, you concentrate you vision on the car moving before you and brain has no reference point to compare to it like other cars or border), it becomes pretty confused. As you saw, restoring binocular vision helps it to make right decision. Quality of the picture plays not the main role here - once the brain can recognize and identify the object, even with low resolution, it can take the difference between eye angles and so know the distance.
Very nice, but not for production yet. First thing it wanted when I tried to upgrade minor package on RedHat is to remove all KDE and Mesa installation altogether, without explaining reasons, and there were no way to make it to download that package (which has no relation neither to KDE not to Mesa - it was bind-util) and install it without blowing out half of my system.
So, I know it's beta and it's expected - but don't try it to manage your production sites yet, or you'll be sorry. I hope they'll fix all the things, since the tool is looking very nice and promising and as soon as they will exorcize all devils from all the small things that would be the killer tool.
XML-RPC is many things, but calling "lightweight" protocol that converts for numeric bytes into 64 textual representation bytes (not counting the whitespace) and adds XML parser pass on each function call - is an exaggeration if I ever saw it. It's definitely not lightweight protocol. Oh, and I forgot - that's an HTTP layer on top of it. Now think about you needing just to call one of your programs from another and pass a number - would you think of a protocol that adds two pretty slow layers of abstraction on top of it, one making message 8 times as big and other further duplicating its size? I would not.
Marijuana is decriminalized and all that.
Oh. So that's the cause for such laws. Now I see.
Is smoking dope mandatory for legislative bodies' members or just a traditional habit?
There's a lot of articles about "smoking dope makes your worse driver/student/whatever". Now we can make a case about "smoking dope makes you stupid legislator" too.
Well, AFAIK you can get internet access for free now. Or if you can't, most ISPs have plans for some $5/month access. Average kid spends more on chewing gum, I guess.