Back in 1994, the Toronto Transit Commission (TTC) introduced annual transit passes. These were credit-card sized, with a magnetic strip to operate the turnstiles and your picture on the card for boarding buses and streetcars. You needed to provide your address when getting the pass in the first place.
No smart cards or RFID needed - presumably each pass had a unique magnetic code, which the TTC can correlate with your photo and address at the time the pass is issued. I don't know if the turnstiles logged the use of the magnetic cards for any extended period of time. They do enforce the rule that the same card cannot operate the turnstiles at a station within a 5-minute time span, to avoid obvious sharing.
This type of tracking was possible only with the single-card pass. Monthly pass users have a separate magnetic card and photo-ID, which are not correlated. You buy a new magnetic card each month, write your photo-ID number on it and away you go, so the mapping between magnetics and person changes each month.
The TTC also had a monthly subscription system where you could give them a credit card number and they mail you a magnetic pass each month. Same problems.
I thought I had a comp.risks submission about this, but I can't find it in the archives.
I am using the BDW collector in an EDA tool. EDA tools store large databases of circuit connectivity, and for various reasons we don't want to be bothered with explicit memory management.
The salient points:
Destructors are not Called
If an object is allocated in collectible memory, then its destructor will not be called when the object is collected. Therefore, destructors are pretty much useless and your code must be designed to work without them.
Actually, if your object derives from class gc_cleanup, then its destructor will be called. However, due to the handling of cleanup functions in the BDW collector, cycles of such objects will never be collected. For this reason, I don't use gc_cleanup much.
Allocating Collectible Memory
By default, C++ allocates objects in the "malloc" heap. The BDW collector maintains a separate heap. In effect, there are four types of memory:
scannable and collectible (GC)
scannable, but uncollectible (NoGC)
non-scannable, but collectible (GC_atomic)
non-scannable, non-collectible (malloc)
"Scannable" refers to the property that objects in the heap are scanned for pointers. "Collectible" refers to the property that objects in the heap will be deallocated if no further references are found.
These four memory types are an issue when you interact with STL and third-party class libraries. By default, STL uses the malloc heap. If you want, say, a std::vector in collectible memory, then you need to write an allocator to get it. The most recent versions of the collector come with such a beast; the version I started using did not.
Similarly, std::string is reference-counted, and in the malloc heap. Here, rather than using an allocator to force it into the collectible heap, I wrote my own lightweight GCString class, which stores the string as an immutable object, and relies on the collector for cleanup.
Third-party class libraries such as ANTLR may use reference-counted objects; you need to bridge between GC and non-GC applications carefully.
"Music blanks" are not quite the same as "data blanks".
Superficially, they look the same, use the same dye technologies, etc.
To assist tracking, CD-R[W] media has a spiral pressed into it which the write laser follows. The return from the pressed spiral is not strong enough to pose a problem for CD players. The spiral also wobbles at a fixed rate to provide a timing reference. This feature is called "Absolute Time In Pre-groove", or ATIP.
At the beginning of the ATIP, a further sub-wobble encodes data such as the media dye type, recording speed, laser power,... and whether the media is an "audio CD" or a "data CD".
Early Philips audio CD recorders will accept only "Audio CDs" as indicated in the ATIP. They also implement the serial copy management system (SCMS) and refuse to make a second-generation copy.
Your typical burner, OTOH, doesn't give a damn about audio vs. data or SCMS, so you can burn to your heart's content.
CD players, not equipped to read the ATIP, can't tell the difference either.
The idea behind "music" vs. "data" was to charge a higher price for the music blanks, and to use the uplift to compensate the industry for piracy, while letting users of "data" blanks get away from the levy.
Of course, in reality it doesn't work that way at all. Almost no one buys music blanks any more (unless they happen to own a Philips recorder), anyone can copy a music CD using a computer data CD-R burner onto a data blank, and certain countries levy data CDs as well as audio. Oh, well...
The CDMA list applies only to members of the CDMA. It won't be abused, but it does not provide complete protection.
Nevertheless, it is worth doing. I used to get 4-5 telespam calls per day. After I put my number on the CDMA DNC, I now get maybe 1-2 a week.
There is no regulation of talk-to-human telemarketing in Canada whatsoever, other than the self-policing CDMA. Automatic dialing/announcing devices (ADAD) are regulated somewhat.
Re:Hence, GPG.
on
P2P Spam?
·
· Score: 2, Insightful
Which will be useless unless you can prove that signing cannot happen without human intervention.
If the GPG secret key is on a Windoze user's hard drive, then what stops the virus from waiting in the background, sniffing the passphrase, then invoking GPG itself?
My prediction: viruses will be used for industrial espionage:
1. Infect home PC of target, and do nothing noticeable. 2. Wait until VPN into employer comes up. 3. Fetch secret info and store on hard drive. 4. Wait until VPN link is dropped and regular Internet access is once again possible. 5. Send secret info to instigator.
Before adopting any of these ideas, one must consider the security implications of doing so.
If we assume that the filesystem is decoupled from the access control layer in the kernel, then one must ensure that any operation that potentially affects security is adequately controlled.
For example, on systems with POSIX_RESTRICTED_CHOWN, the following ought to be illegal:
cp foo/..uid bar/..uid
This can be accomplished by making the UIDs mode 444. Without POSIX_RESTRICTED_CHOWN, the UID is 644. However, we have now moved a systemwide security feature into the filesystem. If multiple filesystems are configured into one kernel, then they ought to be consistent; otherwise the security model will be flawed.
As for things such as allowing access to an environment, doesn't that break encapsulation? It means for a certain filename, the filesystem must grovel through a user-space process to find the environment. Also, if a change in some external environment immediately affects some partially-related processes (e.g. daemons started from that shell), then a whole new raft of security holes will come up based on a process' environment or filesystem layout changing unexpectedly.
Cool ideas, but let's be careful lest we make a steaming pile of Swiss cheese.
A lot of people suggest filtering, but it is not a complete solution.
If you are using your Internet connection for a variety of purposes, then some of that bandwidth is tied up by spammers. Even if your filters are perfect, you are still losing that bandwidth.
For individuals the BW loss may not be significant, but on a large corporate scale it could very well be. We need solutions that prevent the spam from getting sent in the first place.
Strangely enough, I have met the person who has this job, at least in Canada.
In Canada, condoms are medical devices, and are (at least back in 1988) regulated by the Bureau of Radiation and Medical Devices. The poor guy whose job it was to maintain rubber standards had his office in room 61A of the Health Protection Building (Building #7), Tunney's Pasture, Ottawa.
I got wind of this because that summer my job was to babysit two fax machines and a telex for the entire building. This guy slips a 20-page fax into my inbox, which I start sending... and reading. It turns out it was the testing procedure for condoms! Interesting reading...
(Test 20. If >1 leaks, fail the batch. If exactly one leaks, test another 20. Or something like that. Which goes to show that a condom alone is not adequate protection.)
Norton/Symantec/McAfee would like you to believe that $39.95/year or whatever will protect you but the truth is: these programs check against known viruses only. There is always an incubation period between the appearance of a new virus in the wild and the appearance of the update to detect and kill it. This incubation period provides a window for a real virus to do real damage.
To date, there have been no highly damaging viruses. You are lucky. Don't rely on the virus checker to protect you. Instead, look for operating systems and software having inherent immunity built into their design.
Sure, you can use the virus checkers as a secondary measure. But they won't protect you fully.
Let's say I run a 100% FreeBSD/Apache shop. I don't license MS software, so I am not bound by their license. If MS negligence causes me loss, I should be able to sue.
Let's say I am born with severe juvenile diabetes, and I am given at most 4 years to live. However, BigGenCo has a genetic therapy for juvenile diabetes, with which I am treated. I am cured and I go on to live a normal happy life.
Problem: The genetic cure is patented. However, when my cells divide, they are making copies of their genes. The patent controls the fabrication of the patented invention, which in this case is the patched gene.
Am I infringing the patent by living? What if I have children? Will I be bound by a EULA that my parents had to sign?
The theory is this: to install a program on your computer requires that you copy it (or part of it) from the CD-ROM onto your hard drive. To run the program requires that you copy (part of) it into RAM.
Both actions involve making a copy of a copyrighted work, which is illegal.
The EULA grants you an EXEMPTION from copyright law, enabling you to do certain useful things with the software, if you abide by the terms and conditions.
(Now how does this work with copyrighted documents? If I read a document, am I making a copy in my head? Must documents have EULAs to make reading them legal?)
There are 3,903 four-letter words in the National Scrabble Association (NSA) Official Club and Tournament Word List (OWL). Some of them are obscene. The OWL is available only to NSA members.
The Official Scrabble Player's Dictionary is the publication suitable for "recreational and school" play. Kiddies, this is what you can buy in your local bookstore. In this tome, words that have a usage note due to obscenity or vulgarity in Webster's are omitted.
The difference in the lists (4-letter only) is:
ABOS ARSE CUNT DAGO FART FRIG FUCK GOOK GOYS HEBE JEWS JISM KIKE MICK OFAY PEED PISS POMS SHAT SHIT SPIC SPIK TURD TWAT WOGS WOPS YIDS
Ummm... how does the RIAA expect audio CD players and CD-ROM drives to "conform" to their standard?
I briefly investigated CD copy protection a year ago and I made the following observations:
1. My Koss portable player does not mute the DACs when it encounters a data track (as identified by the bits in the Q channel). Oops - a non-conformance right there.
2. Many CD-ROM drives will refuse to read the data track (again, as defined by the Q channel) if the track is labelled as audio in the TOC. (The reverse situation is the principle behind many copy protection systems today: the drives will not rip audio if the track is labelled as data using a buggered multi-session TOC.)
So already we have two non-conformances taken from the massive installed base of optical disc readers. Way to go RIAA!
So: calling all Tcl/C programmers out there: what was YOUR most unusual Tcl application?
For me, it would have to be a CD-ROM/CD-audio copy protection testbed. Think of cdrdao, but with all low-level functionality as Tcl commands. The tool could rip audio, playing it through speakers, while simulating the effects of twiddling the P-channel bit, in real-time! Could burn it too.
Actually, Tcl can use pass-by-reference if you want it to. Tcl objects are dual-ported: they have an internal representation and a string representation. If you define your own representations, you can have the Tcl objects point directly to your internal structures without any need to do name lookups. This is VERY fast, especially if name lookups are expensive.
This reminds me of the Intel presentation at the International Solid State Circuits Conference (ISSCC) 1997. They had a paper describing a 300 MHz Pentium II. Quite an accomplishment at the time - AMD was describing a 275 MHz K6.
Only, the slide presentation was titled "A 433 MHz Out-of-Order Execution Microprocessor". 433? 433!?!?! (Remember: 1997)
The PR flack then proceeded to say "this is an Intel technology demo. This is not a product" and then presented the paper. Things got wierd during the Q&A:
Q: What is the operating temperature? A: "Colder than ice".
Q: What is the power dissipation? A: As a "technology demo" it is not appropriate to disclose such numbers as they are likely to change with actual product release.
There is a general notion among some that anti-spam laws violate constitutional guarantees on free speech. Both the U.S. and Canada have such guarantees in their constitutions. However, I don't think they should apply. Will anyone disagree with me?
The purpose of a right to free speech is to enable just that: someone standing on a street corner and speaking his mind about anything. However, spam and telemarketing are different in that they target specific persons. Sure, that spam was sent to a million people, but the copy that YOU are reading was sent to YOUR address. The telemarketer is similar - she (or the computer) dialed YOUR number. These are not public communications - they are point-to-point directed PRIVATE communications.
Point-to-point communications differ from public communication in that they require some participation from the recipient. When I get a phone call I MUST answer it sooner or later (I can let the machine take a message, but then I must delete it). When I get a spam I MUST delete it, or update my filters to stop future spams. I don't have the option of simply ignoring it, as I do with someone speaking from a soapbox on a street corner. For this reason, initiators of point-to-point communications should not enjoy constitutional protection of their activities.
This doesn't mean that spammers shouldn't be stopped, or that laws regulating spam can't be passed. It simply means that their activity cannot be constitutionally protected.
Slashdot Mirror
Dunno about NX, but the last time I tried LBX with Mentor Graphics IC Station (a VLSI layout tool), the X server rendered outside the window...
outside the framebuffer...
and stomped on other memory. Kaboom!
I don't know if the root cause of the problem was a buggy LBX or a buggy IC Station, but clearly more work was needed in that area.
Back in 1994, the Toronto Transit Commission (TTC) introduced annual transit passes. These were credit-card sized, with a magnetic strip to operate the turnstiles and your picture on the card for boarding buses and streetcars. You needed to provide your address when getting the pass in the first place.
No smart cards or RFID needed - presumably each pass had a unique magnetic code, which the TTC can correlate with your photo and address at the time the pass is issued. I don't know if the turnstiles logged the use of the magnetic cards for any extended period of time. They do enforce the rule that the same card cannot operate the turnstiles at a station within a 5-minute time span, to avoid obvious sharing.
This type of tracking was possible only with the single-card pass. Monthly pass users have a separate magnetic card and photo-ID, which are not correlated. You buy a new magnetic card each month, write your photo-ID number on it and away you go, so the mapping between magnetics and person changes each month.
The TTC also had a monthly subscription system where you could give them a credit card number and they mail you a magnetic pass each month. Same problems.
I thought I had a comp.risks submission about this, but I can't find it in the archives.
The salient points:
Destructors are not Called
If an object is allocated in collectible memory, then its destructor will not be called when the object is collected. Therefore, destructors are pretty much useless and your code must be designed to work without them.
Actually, if your object derives from class gc_cleanup, then its destructor will be called. However, due to the handling of cleanup functions in the BDW collector, cycles of such objects will never be collected. For this reason, I don't use gc_cleanup much.
Allocating Collectible Memory
By default, C++ allocates objects in the "malloc" heap. The BDW collector maintains a separate heap. In effect, there are four types of memory:
"Scannable" refers to the property that objects in the heap are scanned for pointers. "Collectible" refers to the property that objects in the heap will be deallocated if no further references are found.
These four memory types are an issue when you interact with STL and third-party class libraries. By default, STL uses the malloc heap. If you want, say, a std::vector in collectible memory, then you need to write an allocator to get it. The most recent versions of the collector come with such a beast; the version I started using did not.
Similarly, std::string is reference-counted, and in the malloc heap. Here, rather than using an allocator to force it into the collectible heap, I wrote my own lightweight GCString class, which stores the string as an immutable object, and relies on the collector for cleanup.
Third-party class libraries such as ANTLR may use reference-counted objects; you need to bridge between GC and non-GC applications carefully.
Planes are generally quiet places?
Please explain the market for active noise reduction headphones.
Alternatively, what equipment have you been flying? Perhaps Air Canada needs to buy some.
"Music blanks" are not quite the same as "data blanks".
Superficially, they look the same, use the same dye technologies, etc.
To assist tracking, CD-R[W] media has a spiral pressed into it which the write laser follows. The return from the pressed spiral is not strong enough to pose a problem for CD players. The spiral also wobbles at a fixed rate to provide a timing reference. This feature is called "Absolute Time In Pre-groove", or ATIP.
At the beginning of the ATIP, a further sub-wobble encodes data such as the media dye type, recording speed, laser power,... and whether the media is an "audio CD" or a "data CD".
Early Philips audio CD recorders will accept only "Audio CDs" as indicated in the ATIP. They also implement the serial copy management system (SCMS) and refuse to make a second-generation copy.
Your typical burner, OTOH, doesn't give a damn about audio vs. data or SCMS, so you can burn to your heart's content.
CD players, not equipped to read the ATIP, can't tell the difference either.
The idea behind "music" vs. "data" was to charge a higher price for the music blanks, and to use the uplift to compensate the industry for piracy, while letting users of "data" blanks get away from the levy.
Of course, in reality it doesn't work that way at all. Almost no one buys music blanks any more (unless they happen to own a Philips recorder), anyone can copy a music CD using a computer data CD-R burner onto a data blank, and certain countries levy data CDs as well as audio. Oh, well...
The CDMA list applies only to members of the CDMA. It won't be abused, but it does not provide complete protection.
Nevertheless, it is worth doing. I used to get 4-5 telespam calls per day. After I put my number on the CDMA DNC, I now get maybe 1-2 a week.
There is no regulation of talk-to-human telemarketing in Canada whatsoever, other than the self-policing CDMA. Automatic dialing/announcing devices (ADAD) are regulated somewhat.
Which will be useless unless you can prove that signing cannot happen without human intervention.
If the GPG secret key is on a Windoze user's hard drive, then what stops the virus from waiting in the background, sniffing the passphrase, then invoking GPG itself?
My prediction: viruses will be used for industrial espionage:
1. Infect home PC of target, and do nothing noticeable.
2. Wait until VPN into employer comes up.
3. Fetch secret info and store on hard drive.
4. Wait until VPN link is dropped and regular Internet access is once again possible.
5. Send secret info to instigator.
If you're not american and want to buy the software, be careful.
x-plane.com charges US$25 international shipping. For a $60 product? Yeah, right. Getting better, though, it used to be $30.
x-plane.org also sells the software. They have recently increased their shipping rates to Canada - to $16.25 (up from $10).
Dammit, I have a 2Mbps DSL line! Why can't I PayPal the $60 and download the thing, like I do with FreeBSD? Paying $25 to ship a CD is atrocious.
Sympatico (a Canadian ISP) uses "Pollupostage" to refer to spam. This term may also include UseNet postings.
Before adopting any of these ideas, one must consider the security implications of doing so.
If we assume that the filesystem is decoupled from the access control layer in the kernel, then one must ensure that any operation that potentially affects security is adequately controlled.
For example, on systems with POSIX_RESTRICTED_CHOWN, the following ought to be illegal:
cp foo/..uid bar/..uid
This can be accomplished by making the UIDs mode 444. Without POSIX_RESTRICTED_CHOWN, the UID is 644. However, we have now moved a systemwide security feature into the filesystem. If multiple filesystems are configured into one kernel, then they ought to be consistent; otherwise the security model will be flawed.
As for things such as allowing access to an environment, doesn't that break encapsulation? It means for a certain filename, the filesystem must grovel through a user-space process to find the environment. Also, if a change in some external environment immediately affects some partially-related processes (e.g. daemons started from that shell), then a whole new raft of security holes will come up based on a process' environment or filesystem layout changing unexpectedly.
Cool ideas, but let's be careful lest we make a steaming pile of Swiss cheese.
A lot of people suggest filtering, but it is not a complete solution.
If you are using your Internet connection for a variety of purposes, then some of that bandwidth is tied up by spammers. Even if your filters are perfect, you are still losing that bandwidth.
For individuals the BW loss may not be significant, but on a large corporate scale it could very well be. We need solutions that prevent the spam from getting sent in the first place.
(And no, laws won't work.)
Hmmm... the aRts that shipped with KDE 3.0 works fine for me on my 686A.
:-)
Of course, I use FreeBSD and had to write the 686A driver first.
Strangely enough, I have met the person who has this job, at least in Canada.
In Canada, condoms are medical devices, and are (at least back in 1988) regulated by the Bureau of Radiation and Medical Devices. The poor guy whose job it was to maintain rubber standards had his office in room 61A of the Health Protection Building (Building #7), Tunney's Pasture, Ottawa.
I got wind of this because that summer my job was to babysit two fax machines and a telex for the entire building. This guy slips a 20-page fax into my inbox, which I start sending... and reading. It turns out it was the testing procedure for condoms! Interesting reading...
(Test 20. If >1 leaks, fail the batch. If exactly one leaks, test another 20. Or something like that. Which goes to show that a condom alone is not adequate protection.)
And the tools used are 100% inorganic, sorry.
At the risk of /.ing myself, check out these items.
The object in the brushed metal case with the countdown displays is a game timer. It went through baggage X-ray without any questions asked.
The other items (in carry-on) aroused the suspicions of security personnel, and required a manual inspection.
So, where would you have hidden the bomb?
This relates to something I've said all along:
Virus checkers don't work
Norton/Symantec/McAfee would like you to believe that $39.95/year or whatever will protect you but the truth is: these programs check against known viruses only. There is always an incubation period between the appearance of a new virus in the wild and the appearance of the update to detect and kill it. This incubation period provides a window for a real virus to do real damage.
To date, there have been no highly damaging viruses. You are lucky. Don't rely on the virus checker to protect you. Instead, look for operating systems and software having inherent immunity built into their design.
Sure, you can use the virus checkers as a secondary measure. But they won't protect you fully.
If mortgage companies pay spammers $5 for every referral then why can't we spam them back?
Simply create ten million or so "honeypot" email addresses, and have an automated system have them all request information on the mortgage deal.
Once the mortgage company is on the hook for $50 million, they will think again before going to a spam outfit.
This will knock out the mortgage and credit card spams, but won't make a dent in the porn or Viagra spams, as those actually require an order.
MS' EULA shouldn't mean squat here.
Let's say I run a 100% FreeBSD/Apache shop. I don't license MS software, so I am not bound by their license. If MS negligence causes me loss, I should be able to sue.
Period.
How will patents affect any of this?
Let's say I am born with severe juvenile diabetes, and I am given at most 4 years to live. However, BigGenCo has a genetic therapy for juvenile diabetes, with which I am treated. I am cured and I go on to live a normal happy life.
Problem: The genetic cure is patented. However, when my cells divide, they are making copies of their genes. The patent controls the fabrication of the patented invention, which in this case is the patched gene.
Am I infringing the patent by living? What if I have children? Will I be bound by a EULA that my parents had to sign?
The EULA is not redundant.
The theory is this: to install a program on your computer requires that you copy it (or part of it) from the CD-ROM onto your hard drive. To run the program requires that you copy (part of) it into RAM.
Both actions involve making a copy of a copyrighted work, which is illegal.
The EULA grants you an EXEMPTION from copyright law, enabling you to do certain useful things with the software, if you abide by the terms and conditions.
(Now how does this work with copyrighted documents? If I read a document, am I making a copy in my head? Must documents have EULAs to make reading them legal?)
The Official Scrabble Player's Dictionary is the publication suitable for "recreational and school" play. Kiddies, this is what you can buy in your local bookstore. In this tome, words that have a usage note due to obscenity or vulgarity in Webster's are omitted.
The difference in the lists (4-letter only) is:
ABOS ARSE CUNT DAGO FART FRIG FUCK GOOK GOYS
HEBE JEWS JISM KIKE MICK OFAY PEED PISS POMS
SHAT SHIT SPIC SPIK TURD TWAT WOGS WOPS YIDS
I briefly investigated CD copy protection a year ago and I made the following observations:
1. My Koss portable player does not mute the DACs when it encounters a data track (as identified by the bits in the Q channel). Oops - a non-conformance right there.
2. Many CD-ROM drives will refuse to read the data track (again, as defined by the Q channel) if the track is labelled as audio in the TOC. (The reverse situation is the principle behind many copy protection systems today: the drives will not rip audio if the track is labelled as data using a buggered multi-session TOC.)
So already we have two non-conformances taken from the massive installed base of optical disc readers. Way to go RIAA!
For me, it would have to be a CD-ROM/CD-audio copy protection testbed. Think of cdrdao, but with all low-level functionality as Tcl commands. The tool could rip audio, playing it through speakers, while simulating the effects of twiddling the P-channel bit, in real-time! Could burn it too.
Actually, Tcl can use pass-by-reference if you want it to. Tcl objects are dual-ported: they have an internal representation and a string representation. If you define your own representations, you can have the Tcl objects point directly to your internal structures without any need to do name lookups. This is VERY fast, especially if name lookups are expensive.
Ahhh... "Technology demo".
This reminds me of the Intel presentation at the International Solid State Circuits Conference (ISSCC) 1997. They had a paper describing a 300 MHz Pentium II. Quite an accomplishment at the time - AMD was describing a 275 MHz K6.
Only, the slide presentation was titled "A 433 MHz Out-of-Order Execution Microprocessor". 433? 433!?!?! (Remember: 1997)
The PR flack then proceeded to say "this is an Intel technology demo. This is not a product" and then presented the paper. Things got wierd during the Q&A:
Q: What is the operating temperature?
A: "Colder than ice".
Q: What is the power dissipation?
A: As a "technology demo" it is not appropriate to disclose such numbers as they are likely to change with actual product release.
Yeah, sure.
The purpose of a right to free speech is to enable just that: someone standing on a street corner and speaking his mind about anything. However, spam and telemarketing are different in that they target specific persons. Sure, that spam was sent to a million people, but the copy that YOU are reading was sent to YOUR address. The telemarketer is similar - she (or the computer) dialed YOUR number. These are not public communications - they are point-to-point directed PRIVATE communications.
Point-to-point communications differ from public communication in that they require some participation from the recipient. When I get a phone call I MUST answer it sooner or later (I can let the machine take a message, but then I must delete it). When I get a spam I MUST delete it, or update my filters to stop future spams. I don't have the option of simply ignoring it, as I do with someone speaking from a soapbox on a street corner. For this reason, initiators of point-to-point communications should not enjoy constitutional protection of their activities.
This doesn't mean that spammers shouldn't be stopped, or that laws regulating spam can't be passed. It simply means that their activity cannot be constitutionally protected.
Opinions?