There are ways around the First Amendment on that, too. One way is to define in law (the banks won't like this, of course) that any private information about someone is the intellectual property of that person. Then any private information about the relationship between two people (counting businesses are a person, here) is shared property requiring both parties to agree to any copying beyond fair use (the bank gets to use the information in its own business to properly service its customers, under fair use). Define it this way and we can use copyright laws to protect privacy.
Of course the banks, and telemarketers, and other thieves of our private intellectual property (information about ourselves), won't like this, because we will be calling the cash cow to come home.
Netscape 4 does not do that while the page is loading. The URL does flash up very briefly (too quick to be read), but then is wiped over by the status line that says how many bytes have been read at what rate. And big stories like this tend to have quite a lot of bytes and it takes a while to be loading them. And many people do know about middle-clicking on links to force a new window and get concurrent loading on 2 pages while they go get a refill on the coffee... only to come back to see that nytimes.com wants a password. Basically, you have to wait until the page finishes loading to see the mouse-over URL because some nitwit programmer at Netscape in days gone by (Netscape 3 had the same stupidity) decided it was cool to overload different status messages in the same place.
The flaw in this design is spammers will eventually figure it out. Three years ago I set up an email box on my own server with my own domain (so it would not be subject to an ISP or webmail provider giving it out). I never put the exact address online. Instead, I put a munged form of it online with things like "nospam" added. Guess what. It got spammed. Spammers figured out how to remove "nospam" from the address. That's now built in to spamware (some doesn't remove the "-" if I use name-nospam). It won't be long (maybe a year at most) after your method becomes popular for spammers to figure out how to detect and modify it to get through. And soon after that, the spamware will know how.
The only way to do this is with a scheme that makes it next to impossible to guess the base form, or an alternate form. For example, take the MD5 checksum of the date, along with a secret string you don't tell anyone, and use the first few characters as that email address. You can use it in the subdomain or the left-hand-side.
What will be needed will be a set of these pre-generated so the mail server already accepts them, and you store them in your PDA or other places where you can readily access them, and record who got which address.
The key is to provide no means for predicting what address can be used to bypass the filters.
How many times do I have to opt out if a million businesses decide to take up spamming over the course of the next year or so. Sometimes I get over a dozen different copies of exactly the same spam from exactly the same sender, sent to a dozen different email addresses. These are legitimately different addresses because they have different roles. Of course a spammer won't know they go to the same person. But sending spam to them is essentially OFF TOPIC because their role isn't to respond to advertising.
Until the FTC (and this may require Congress to do this) adopts the principle that opting *IN* is required first, and that I should not have to go to the trouble to opt out if I never opted in in the first place, then as far as I'm concerned, any actions by the FTC is misguided and useless.
To make a decent gateway out of such a box, one needs TWO ethernet ports. There is a slot there, but it would be nicer if it was integrated. So not only do I need to add the slim CDROM drive cost, but also the extra NIC card cost (which is not all that much, but now there is no more slot available).
What I'd like to know is, did all these people who come up with all these new methodologies only decide to think these up after Java was created? In other words, why weren't these methods... and all the toolkits and such that are also mentioned... introduced earlier? Why now? I mean, was it the fact that the Java language came out, or did it just take them this long to figure out these things? Some of these methods could have been useful in 1982, or even 1972, for example. But if the people who thought it up weren't even in college by then, I guess that could explain it.
My setup is a little more sophisticated. It sets the address for *.doubleclick.com (and others) to a special web server configuration which always delivers a 1x1 transparent GIF no matter what URI is requested. Bam, no tracking, and a clean substitute for ads.
Another option would be to have everything that jumps between domains (possibly for domains that are configured, or domains not configured to be exempted) have the HTTP "Referer" header suppressed, or forged. That would create the brick wall boundary between domains where information cannot as easily pass between, through your server. Cookies cannot be retrieved across domains, but by associating the cookie you get from the image with the domain in the "Referer" they can still track what domain you are surfing.
BTW, I do have cookies on, but each new instance of my browser creates a whole new context to run in (which it thinks is my home directory), which means an empty set of cookie each time. So I just make sure I start a new instance each time I go to another site.
My DNS server sends all queries for doubleclick.com and doubleclick.net (and some others) over an HTTP/HTTPS server that for any URI requested, always delivers a 1x1 transparent GIF. Bingo, no ads, and nothing tracked.
I got a 1x1 pixel transparent GIF file. But that is because I directed all queries for anything in doubleclick.com (and some others) at my DNS server over to a special IP address on which my web server always delivers that 1x1 pixel transparent GIF file no matter what URI is requested. It even does it on HTTPS (self signed cert).
Well, let me be sure to point out then that Slashdot also serves Doubleclick ads.
Well, let me be sure to point out then that Doubleclick ads are blocked here. So when my Slashdot page comes up, regardless of whether the Elite Monkeys generate it, or the Random Elephants generate it, or the Barrel of Psycho Mummies generate it, if it has images that refer to any server in the doubleclick domain (and a few others), they come up blank (a 1x1 transparent GIF is substituted). If Slashdot wants to be sure to maximize revenues, it should either be sure it charges for providing the tag, even if the image is never loaded, or make sure a different advertising source is used (which may be hard if the advertiser wants to use doubleclick... but then, those are going to be advertisers that are not going to generate as much revenue for this very reason). As I edit this comment, I'm seeing a banner ad for OSDN's PriceCompare. I may check it out later when I'm bored.
And as my luck would have it, I happen to have 5 points right now, and I see stories I'd rather post replies for. And no, I am not advertising them for sale; don't even think about it. I'll find a way to use them.
The advertising model COULD have been made to work. Perhaps it would never work quite as well as some had hoped for, but I do believe it can work better than some are now claiming.
One of the ways it failed is due to an expectation that wasn't realistic. Unlike other forms of advertising, such as radio spots, TV commercials, and blocks or pages in a newspaper or magazine, the web/internet came with technology that could gauge a response when people clicked on the ads. The reason this fails is because too many people just don't click on the ads. Now I have clicked on a few, even here on Slashdot, but that was only when it was a combination of something I was really intersted in, and I happened to be bored at the time. When I'm not bored, I have goals, such as reading the interesting article. I still see the ads, but I move on.
The correct way to do advertising on the web, and the way it will work, is to expect them to work the same way they work in other media... make impressions. Banner ads, and even the hated big box ads can have that effect. And small text ads can have that effect, too. But what the ads writers/creators have to do is make the ads impressionable. The ones that say "Click here for the best home mortgages" don't do any good to create a brand impression. Instead it should give the mortgage company name along with words that say what is being offered, e.g. lower rates, loans to those with poor credit, no down payment loans, or whatever. Don't depend on people to click through today, but make sure they know your name so they will seek you out when they are in the need for your product or service.
Because advertising and marketing executives were so interested in this new technology to allow them to track click throughs, they forgot about what makes advertising work in the first place, which is a combination of simple information and brand recognition. Making them obtrusive may have some negative impact, but when advertising is done on the old tradition impression basis (which is still going to work because the audience is still the same species that it has worked on before), even small text blocks used to separate sections or stories can have as good an impression as that big annoying box, and work out as a better compromise between an advertiser seeking more returns for a smaller investment, and a publisher seeking larger revenues while retaining and growing an audience to deliver to that advertiser.
That's kind of what I meant, though not so much in terms of the store clerk than the law enforcement. But you're right, that is certainly one aspect of it that influences how readily people might buy. That and, someone might recognize them in the store (especially in a smaller community) or their car in the parking lot. I wouldn't worry about that as much at my grocery store, whereas delivery of food by a truck meant having to be home whenever the trick might arrive.
These 3 succeed because they serve their customers
on
Web Profits in the Gutter
·
· Score: 4, Insightful
These 3 succeed because they serve their customers. The others are basically doing a terrible job at understanding what the customer wants and providing it at a reasonable price, under reasonable terms.
Porn
This is quite obvious. The customer wants it. Someone has it. And the price seems to work out. And the terms are usually pretty good, too (e.g. we're in a different country than your law enforcement, so they can take a hike).
Spam
First, understand that the customer is not the one getting the mail. The customer is the one paying the spammer to send mass mail. The price is way cheaper than snail mail spam, so the spammers are succeeding and meeting what their customers want.
Fraud
The customer in this case is the fraud perpetrator themselves. The victim is not the customer. So in a sense this works, too, when the perpetrator gets away with the booty.
Perhaps internet gambling should be included here, too. It's not as big as porn, but from what I hear, it's nearly as successful. Vice does tend to be a good business model. I'm sure once they have the technology to download matter replication (if there is some kind of digital rights management for it), there will be plenty of places to get marijuana and drugs online. They'd do it if they could, because this is a massive market. Also, knives, guns, bombs, and maybe even nuclear material might be sold this way. You can bet the government would have a fit if the technology allowed this to go on without them being able to trace it all.
Gee, thanks for whining. Apparently they heard you and changed permissions on it. Now it's denied.
Re:choice / customization is a *GOOD* thing
on
KDE Gets The Hat
·
· Score: 2
Part of the problem appears to be that such a mess, that you refer to as being the result of mixing operating systems, comes about in the first place. We should not have such messes if things worked right. While I won't deny that today there would indeed be this mess, the goal should be to make it so that we can have the mix, without the mess. If that means fixing every operating system, so be it. If one of them doesn't want to play along, that can be a problem. If Linux is more popular, then Windows, the one that probably doesn't want to play along, might, some day, have to take a back seat (well, we can wish).
I have administered a mix of Windows, Linux, and Solaris, with even some BSD machines. And not all of them were servers, either. There were some messes at first, but I removed them.
Where the cost of having a mix outweighs the cost of the benefits of invidualized computers, the one with the greater savings should win. Have you ever seen any studies on the cost savings of individualized computers to see if they are more or less than the savings of having all the same? I've never seen such a thing. Almost every manager I have met has no clue in how to measure things like increased worker productivity. Instead, the only thing they measure is the discount price they get for buying everything in one place (which certainly is an important factor, but it shows that technical issues just get thrown aside too easily).
Re:choice / customization is a *GOOD* thing
on
KDE Gets The Hat
·
· Score: 2
Making Linux "widely accepted" doesn't mean jamming Linux down people's throat. Hell, they can keep using Windows for all I care. I hate it when I find out some office happens to force everyone to use Windows. But I hate it equally when I find out some office happens to force everyone to use Red Hat. And I'd hate it equally if I find out they force everyone to use Slackware, or FreeBSD.
If Dave wants to use Debian, and Francis wants to use FreeBSD, and Steve wants to use Slackware, and Wilma wants to use Windows, and Mark wants to use a Macintosh, and if they can all be made to communicate as needed (which is very true), then why the hell not? If the office has to hire staff to support these computers, then of course they have to consider the cost of these choices. But if some of those people don't need the snot-nosed kid to come around and reinstall their computer for them, and they can run it themselves, then they should.
The goal for "acceptance" shouldn't be which one OS the office PHB accepts, but rather, that the office accepts what works best for each individual. We are not robots, and I resent corporate attempts to turn people into that.
Here is another example about how a lawyer (turned technology writer) can still be totally clueless, not only about the technology (e.g. any spammer can forge his domain to use his mail server), but even how the law applies (he gave permission to test for open relay, and it is common and standard practice for such tests to perform, in a non-malicious way, every known technique that a spammer might perform). By the way, I wrote that guy a long and detailed letter, in a non-threatening and non-abusive way, simply describing to him all the details of how it all works. To date I have not received a reply. But it's probably because his mail server is still an open relay, and still listed in one of the DNSbls I use.
Lawyers are just dumb about technology. Here is some evidence of that. In this article a lawyer turned technology writer who runs a server still has no clue about how a mail server can be an open relay, and even how the law applies (e.g. he gave permission to test and that implies using standard and conventional methods).
Slashdot Mirror
There are ways around the First Amendment on that, too. One way is to define in law (the banks won't like this, of course) that any private information about someone is the intellectual property of that person. Then any private information about the relationship between two people (counting businesses are a person, here) is shared property requiring both parties to agree to any copying beyond fair use (the bank gets to use the information in its own business to properly service its customers, under fair use). Define it this way and we can use copyright laws to protect privacy.
Of course the banks, and telemarketers, and other thieves of our private intellectual property (information about ourselves), won't like this, because we will be calling the cash cow to come home.
The links at linuxhomepage.com go direct to the article page.
Netscape 4 does not do that while the page is loading. The URL does flash up very briefly (too quick to be read), but then is wiped over by the status line that says how many bytes have been read at what rate. And big stories like this tend to have quite a lot of bytes and it takes a while to be loading them. And many people do know about middle-clicking on links to force a new window and get concurrent loading on 2 pages while they go get a refill on the coffee ... only to come back to see that nytimes.com wants a password. Basically, you have to wait until the page finishes loading to see the mouse-over URL because some nitwit programmer at Netscape in days gone by (Netscape 3 had the same stupidity) decided it was cool to overload different status messages in the same place.
The flaw in this design is spammers will eventually figure it out. Three years ago I set up an email box on my own server with my own domain (so it would not be subject to an ISP or webmail provider giving it out). I never put the exact address online. Instead, I put a munged form of it online with things like "nospam" added. Guess what. It got spammed. Spammers figured out how to remove "nospam" from the address. That's now built in to spamware (some doesn't remove the "-" if I use name-nospam). It won't be long (maybe a year at most) after your method becomes popular for spammers to figure out how to detect and modify it to get through. And soon after that, the spamware will know how.
The only way to do this is with a scheme that makes it next to impossible to guess the base form, or an alternate form. For example, take the MD5 checksum of the date, along with a secret string you don't tell anyone, and use the first few characters as that email address. You can use it in the subdomain or the left-hand-side.
What will be needed will be a set of these pre-generated so the mail server already accepts them, and you store them in your PDA or other places where you can readily access them, and record who got which address.
The key is to provide no means for predicting what address can be used to bypass the filters.
How many times do I have to opt out if a million businesses decide to take up spamming over the course of the next year or so. Sometimes I get over a dozen different copies of exactly the same spam from exactly the same sender, sent to a dozen different email addresses. These are legitimately different addresses because they have different roles. Of course a spammer won't know they go to the same person. But sending spam to them is essentially OFF TOPIC because their role isn't to respond to advertising.
Until the FTC (and this may require Congress to do this) adopts the principle that opting *IN* is required first, and that I should not have to go to the trouble to opt out if I never opted in in the first place, then as far as I'm concerned, any actions by the FTC is misguided and useless.
To make a decent gateway out of such a box, one needs TWO ethernet ports. There is a slot there, but it would be nicer if it was integrated. So not only do I need to add the slim CDROM drive cost, but also the extra NIC card cost (which is not all that much, but now there is no more slot available).
Sounds like you, too, need a bigger pipe. I know I do, too, but at least I managed to reach 93% on my 2 80GB drives.
Problem is, when PHB sees book in store, suddenly what is methodology becomes "magic potion" (at least the PHB thinks it is) in the office.
But will they work with languages like C or Fortran?
What I'd like to know is, did all these people who come up with all these new methodologies only decide to think these up after Java was created? In other words, why weren't these methods ... and all the toolkits and such that are also mentioned ... introduced earlier? Why now? I mean, was it the fact that the Java language came out, or did it just take them this long to figure out these things? Some of these methods could have been useful in 1982, or even 1972, for example. But if the people who thought it up weren't even in college by then, I guess that could explain it.
My setup is a little more sophisticated. It sets the address for *.doubleclick.com (and others) to a special web server configuration which always delivers a 1x1 transparent GIF no matter what URI is requested. Bam, no tracking, and a clean substitute for ads.
Another option would be to have everything that jumps between domains (possibly for domains that are configured, or domains not configured to be exempted) have the HTTP "Referer" header suppressed, or forged. That would create the brick wall boundary between domains where information cannot as easily pass between, through your server. Cookies cannot be retrieved across domains, but by associating the cookie you get from the image with the domain in the "Referer" they can still track what domain you are surfing.
BTW, I do have cookies on, but each new instance of my browser creates a whole new context to run in (which it thinks is my home directory), which means an empty set of cookie each time. So I just make sure I start a new instance each time I go to another site.
My DNS server sends all queries for doubleclick.com and doubleclick.net (and some others) over an HTTP/HTTPS server that for any URI requested, always delivers a 1x1 transparent GIF. Bingo, no ads, and nothing tracked.
I got a 1x1 pixel transparent GIF file. But that is because I directed all queries for anything in doubleclick.com (and some others) at my DNS server over to a special IP address on which my web server always delivers that 1x1 pixel transparent GIF file no matter what URI is requested. It even does it on HTTPS (self signed cert).
Here is my list:
Well, let me be sure to point out then that Doubleclick ads are blocked here. So when my Slashdot page comes up, regardless of whether the Elite Monkeys generate it, or the Random Elephants generate it, or the Barrel of Psycho Mummies generate it, if it has images that refer to any server in the doubleclick domain (and a few others), they come up blank (a 1x1 transparent GIF is substituted). If Slashdot wants to be sure to maximize revenues, it should either be sure it charges for providing the tag, even if the image is never loaded, or make sure a different advertising source is used (which may be hard if the advertiser wants to use doubleclick ... but then, those are going to be advertisers that are not going to generate as much revenue for this very reason). As I edit this comment, I'm seeing a banner ad for OSDN's PriceCompare. I may check it out later when I'm bored.
And as my luck would have it, I happen to have 5 points right now, and I see stories I'd rather post replies for. And no, I am not advertising them for sale; don't even think about it. I'll find a way to use them.
The advertising model COULD have been made to work. Perhaps it would never work quite as well as some had hoped for, but I do believe it can work better than some are now claiming.
One of the ways it failed is due to an expectation that wasn't realistic. Unlike other forms of advertising, such as radio spots, TV commercials, and blocks or pages in a newspaper or magazine, the web/internet came with technology that could gauge a response when people clicked on the ads. The reason this fails is because too many people just don't click on the ads. Now I have clicked on a few, even here on Slashdot, but that was only when it was a combination of something I was really intersted in, and I happened to be bored at the time. When I'm not bored, I have goals, such as reading the interesting article. I still see the ads, but I move on.
The brightest green laser pointer around is now at ThinkGeek.
The correct way to do advertising on the web, and the way it will work, is to expect them to work the same way they work in other media ... make impressions. Banner ads, and even the hated big box ads can have that effect. And small text ads can have that effect, too. But what the ads writers/creators have to do is make the ads impressionable. The ones that say "Click here for the best home mortgages" don't do any good to create a brand impression. Instead it should give the mortgage company name along with words that say what is being offered, e.g. lower rates, loans to those with poor credit, no down payment loans, or whatever. Don't depend on people to click through today, but make sure they know your name so they will seek you out when they are in the need for your product or service.
integrate. collaborate. accelerate. SourceForge 3.1, from VA Software.
Because advertising and marketing executives were so interested in this new technology to allow them to track click throughs, they forgot about what makes advertising work in the first place, which is a combination of simple information and brand recognition. Making them obtrusive may have some negative impact, but when advertising is done on the old tradition impression basis (which is still going to work because the audience is still the same species that it has worked on before), even small text blocks used to separate sections or stories can have as good an impression as that big annoying box, and work out as a better compromise between an advertiser seeking more returns for a smaller investment, and a publisher seeking larger revenues while retaining and growing an audience to deliver to that advertiser.
Smart mass putty. Bounce, stretch, contort, relieve stress. At ThinkGeek.
That's kind of what I meant, though not so much in terms of the store clerk than the law enforcement. But you're right, that is certainly one aspect of it that influences how readily people might buy. That and, someone might recognize them in the store (especially in a smaller community) or their car in the parking lot. I wouldn't worry about that as much at my grocery store, whereas delivery of food by a truck meant having to be home whenever the trick might arrive.
These 3 succeed because they serve their customers. The others are basically doing a terrible job at understanding what the customer wants and providing it at a reasonable price, under reasonable terms.
This is quite obvious. The customer wants it. Someone has it. And the price seems to work out. And the terms are usually pretty good, too (e.g. we're in a different country than your law enforcement, so they can take a hike).
First, understand that the customer is not the one getting the mail. The customer is the one paying the spammer to send mass mail. The price is way cheaper than snail mail spam, so the spammers are succeeding and meeting what their customers want.
The customer in this case is the fraud perpetrator themselves. The victim is not the customer. So in a sense this works, too, when the perpetrator gets away with the booty.
Perhaps internet gambling should be included here, too. It's not as big as porn, but from what I hear, it's nearly as successful. Vice does tend to be a good business model. I'm sure once they have the technology to download matter replication (if there is some kind of digital rights management for it), there will be plenty of places to get marijuana and drugs online. They'd do it if they could, because this is a massive market. Also, knives, guns, bombs, and maybe even nuclear material might be sold this way. You can bet the government would have a fit if the technology allowed this to go on without them being able to trace it all.
Gee, thanks for whining. Apparently they heard you and changed permissions on it. Now it's denied.
Part of the problem appears to be that such a mess, that you refer to as being the result of mixing operating systems, comes about in the first place. We should not have such messes if things worked right. While I won't deny that today there would indeed be this mess, the goal should be to make it so that we can have the mix, without the mess. If that means fixing every operating system, so be it. If one of them doesn't want to play along, that can be a problem. If Linux is more popular, then Windows, the one that probably doesn't want to play along, might, some day, have to take a back seat (well, we can wish).
I have administered a mix of Windows, Linux, and Solaris, with even some BSD machines. And not all of them were servers, either. There were some messes at first, but I removed them.
Where the cost of having a mix outweighs the cost of the benefits of invidualized computers, the one with the greater savings should win. Have you ever seen any studies on the cost savings of individualized computers to see if they are more or less than the savings of having all the same? I've never seen such a thing. Almost every manager I have met has no clue in how to measure things like increased worker productivity. Instead, the only thing they measure is the discount price they get for buying everything in one place (which certainly is an important factor, but it shows that technical issues just get thrown aside too easily).
Making Linux "widely accepted" doesn't mean jamming Linux down people's throat. Hell, they can keep using Windows for all I care. I hate it when I find out some office happens to force everyone to use Windows. But I hate it equally when I find out some office happens to force everyone to use Red Hat. And I'd hate it equally if I find out they force everyone to use Slackware, or FreeBSD.
If Dave wants to use Debian, and Francis wants to use FreeBSD, and Steve wants to use Slackware, and Wilma wants to use Windows, and Mark wants to use a Macintosh, and if they can all be made to communicate as needed (which is very true), then why the hell not? If the office has to hire staff to support these computers, then of course they have to consider the cost of these choices. But if some of those people don't need the snot-nosed kid to come around and reinstall their computer for them, and they can run it themselves, then they should.
The goal for "acceptance" shouldn't be which one OS the office PHB accepts, but rather, that the office accepts what works best for each individual. We are not robots, and I resent corporate attempts to turn people into that.
Here is another example about how a lawyer (turned technology writer) can still be totally clueless, not only about the technology (e.g. any spammer can forge his domain to use his mail server), but even how the law applies (he gave permission to test for open relay, and it is common and standard practice for such tests to perform, in a non-malicious way, every known technique that a spammer might perform). By the way, I wrote that guy a long and detailed letter, in a non-threatening and non-abusive way, simply describing to him all the details of how it all works. To date I have not received a reply. But it's probably because his mail server is still an open relay, and still listed in one of the DNSbls I use.
Lawyers are just dumb about technology. Here is some evidence of that. In this article a lawyer turned technology writer who runs a server still has no clue about how a mail server can be an open relay, and even how the law applies (e.g. he gave permission to test and that implies using standard and conventional methods).