We did it at the last place I worked at. I didn't trust their figures, and going through the logs, it was obvious they were grossly inflating their claimed click-through results - by almost 2000%. When confronted with the hard evidence, they backed down, apologized, gave us a free month, etc. There were days that they claimed we had received plenty of click-throughs, but there wasn't a single log entry. In other words, they weren't accurately parsing their log files (they were running their servers on Windows, so who knows what they were doing...) or they were committing out-and-out fraud. Our test clicks showed up in the log files as being relayed via their server, so they couldn't say that there was something wrong with our setup.
2 years later, they're out of business. I guess others cottoned on to the same thing... or (more likely) my old boss started bragging about how he caught them fudging their click-through rates and page views, and others started putting their own checks into place.
Bullshit - you have the right to break up the software into its distinct files and use them as you will. What you don't have the right to is to make multiple copies (that's what copyright is about) without permission. How you use your legal copy is entirely up to you. Its like those bogus licensing provisions that Microsoft wants you to think are legally binding - like "you must install this sticker on the computer to which the OS is installed." You don't have to. Anyone who says otherwise is a sheeple.
I said "if the proxy isn't broken" - caching proxies that serve up urls with parameters that differ for each request are clearly broken. If you're generating the parameters dynamically/pseudo-randomly (which you should be doing) they shouldn't be cached.
If you use a javascript xmlhttprequest to fetch the ad content from your server(a 2-line paste on the web page, to include your source script, which generates the pseudorrandom parameters for each request, and less than 20 lines on the host), then caching proxies don't even come into it, and you get an accurate-enough count.
"How do I know this? How many games have I played? I know it by the definition of the vast majority of games. They tend to involve (1) point and shoot in many variations and plotlines, (2) treasure or scavenger hunts, as in Myst, and (3) player control of the outcome. I don't think these attributes have much to do with art; they have more in common with sports."
How do I know this? How many paintings have I looked at? I know it by the definition of the vast majority of paintings. They tend to involve (1) strokes and colors in many variations and plotlines, (2) subjects and lighting, and (3) artist control of the outcome. I don't think these attributes have much to do with art; they have more in common with video games.
Both arguments are BS, except that, unlike Ebert, mine was crap by intent.
I give him two thumbs down. Or one finger up, if you want to be artsy-fartsy.
There are lots of ways. Web bugs (a one-pixel by 1-pixel image - = INSERT_UNIQUE_STRING_HERE> work. You know the mail has been received and which recipient looked at it.
Or serve the image from your own server, and parse your logs for requests.
Or host the text ads yourself.
Or make the text ad a block of javascript that goes to your server and serves up custom ads, like the "big boys" do, and embed it in the html of a web page.
There are dozens of ways - just ask any spammer, they have LOTS of experience with this sort of stuff.
If they did as I pointed out, the problem of caches is fixed unless the caching mechanism is broken. Putting parameters after the url is the usual way its implemented, and it tells the caching proxy that this is NOT static content, and to pass through the request instead of using the cached object. If they don't re-fetch your ad, you don't pay, since if they don't fetch it, they never see it.
Also, it doesn't matter if the average user cycles through a dozen IPs, over the course of a day - if they view your ad once, because it will come from just one of those IPs, not multiple IPs. Of course, you can also just look at the different IPs the various requests for different parts of the page are passed through, and build up a map of IPs that all front for the same user.
No solution is perfect, but this sure is better than nothing.
why not do a a mod_rewrite rule for Alexa, sending them back to alexa.com, or better yet, the DHS. Be nice to see Alexa classified as a "terr'rist". (yeah, I know there are problems with this theory, but we can dream, right:-)
If the site says "We'll get you N page views per week" for X period of weeks, and the first week comes in at N/4, you cancel the rest. Anyone relying on Alexa rather than demanding a performance guarantee is just being dumb.
"It isn't surprising that people who spend money on advertising want to have some metric by which to predict (estimate, guess, what-have-you) the impact of each dollar spent on web advertising."
There are several easy ways:
as an advertiser, host the ad on your own server, and just look in your logs..,
as an advertiser, get access to the server's banner administration system for your ad account (postnuke allows this on a per-advertiser basis)
as an advertiser, just be skeptical as all hell and don't believe 99% of the stuff you hear - its all BS anyway
If you're so naive as to not insist on hard numbers for actual views (the log files are best , you deserve to get hosed - you can analyse the log files and factor out multiple views per host ip to get the actual number of real views, and reduce fraud; ditto with geolocation of ip addresses to factor out bots in 3rd world countries; ditto for bots that crawl every link on a page; ditto for pages that are loaded then immediately dumped for another page).
As an advertiser, I'd want unique eyeballs - real human eyeballs - that can be verified.
It might not be news in the sense of something that just happened, but its still worth being reminded about once in a while.
Its true - the people who obsess over PageRank and SEO are the ones driving this - everyone already knows their own web stats just by looking at their server logs.
"If I were them I'd create a closed network with educational sites alone. "
They don't call it "sex education" for nothing, you know...
I guess someone didn't get the email - the internet is for porn. (And for those kids in Nigeria using OLPCs, they'll just search for "pr0n", and br3asts - after all, look how they spell other words, like V14GR4).
Real porn (as in obscene) to them will be seeing just how many obese people there are in the world. People who claim to be preaching th word of god, and want to end world hunger, but look like they personally caused at least one famine by eating everything in sight. Isn't gluttony a sin?
"Copying them from your Windows machine is pirating the software."
If he removes them from his Winbox to his linbox he's not pirating... I'm sure a lot of us have legit copies of older versions of windows that we don't use any more, but that we can use the codecs, fonts, etc.
Patents for software or business methods are bogus. They have zero to do with the original patent language, and are a "feature" of the USPTO that attests more to the bugginess of the process than anything else.
Copyrights for software make sense. It is code, it is copywriteable. But patented? Its not an invention.
Besides, you fail to address my main point - that because Microsoft has done nothing to reduce its alleged losses, specifically by identifying the alleged infringements, the doctrine of laches means they can't succeed with a claim any more. If they were to file a lawsuit tomorrow, they would have to prove that the lawsuit was over a patent OTHER THAN THE 235 "allegedly infringed" patents that are now lawsuit-proof. So they would ahve to name 236 patents, point to the first 235 and say, "it was not one of those", etc.
They will never successfully sue. They know it. They blew any basis for a successful leagal claim by milking it for fud value, rather than disclosing what it was (since it never existed anyway, they couldn't...).
It doesn't matter in the long run, because in the long run (20 years), Microsoft will be a niche player, probably bought out by Apple, since they can no longer produce competitive products.
"So what if I was? That has no bearing on who I am. My identity is the sum of the things I've done. Really, the only thing you have to prove is that I didn't do things under a different identity that are naughty."
It looks like we might be heading towards a different sort of singularity than was previously envisioned. A world where the US self-destructs (just look at the debt clock), ripped apart by capitalizm mixed with profiteering and greed among the well-connected, and the RotW (Rest of the World).
Some of the economic projections for later this century are downright ugly - with a US with a population of up to half a billion, but a permanently wrecked economy, as we hit the upper limits of what's possible, and the rest of the world catches up, then surpasses the US in terms of GNP. Both China and India will have larger economies by the middle of the century - and less debt.
"Uh. Rigging is the whole point of intellectual monopoly law. It's antithetical to free market capitalism. I don't know why you're surprised, america - you're about as capitalist as the soviets were communist."
People keep forgetting that the Internet routes around damage. There is no such thing as a monopoly when the whole operation can be moved to servers in another country. You're a seller in the US, and you want to sell to someone else in the US - just auction it off on ebay in the UK or Canada or france.
Do you really expect everyone who has a stake in linux NOT to join in the case so as to protect their own interests? Would IBM sit on the sidelines and let a judgement go through without unleashing a few of the nazgul?
Also remember, its a LOT cheaper to defend than to pursue.
RedHat only needs to make some motions in discovery to get Mr. Softie to STFU. "Identiy with specificity the patents that you claim are infringing, by patent number and claim number, and source code file and line number in the infringing implementations, how these supposedly infringing pieces of code are prohibited under current doctrines, the names of the original patent holders (for more discovery) and anyone else who has seen the source code for the patented materials, what damages have been incurred, and what Microsoft has done to mitigate damages."
That last bit - the obligation of the injured party to mitigate damages - will kill Mr. Softie, since they have gone out of their way NOT to mitigate damages by disclosing any infringing patents. The doctrine of laches means they are f$cked.
True identification IS impossible on the Internet. Any and all solutions have cracks in it. Heck, you can't even prove your identity in "real life" unless you've been fingerprinted at birth. Prove who you really are. Can you prove that you weren't switched at birth? That you weren't really adopted, and your adoptive parents never told you (or the converse - that your parents told you you were adopted, but lied - to cover up, for example, that your uncle is really your father...)
Sure, people have known you for years and will vouch for your identity - but people have killed other people, stolen their identities, and then after enough time passed, people vouchsafed them as the "real".
Microsoft will never sue - they know that the only thing they can do is amke noises. Actually suing would be the equivalent of a first strike in a MAD - Mutually Assured Destruction - scenario, which they would ultimately lose.
The resulting positive publicity for linux would further erode their already slipping grip on their customer base. Like it did with allthe SCO BS.
"The software would also free advertising from its traditional browser yoke. "A word processor may display a banner ad along the top of a window, similar to a toolbar, while a graphical ad may be displayed in a frame associated with the application. A digital editor for photos or movies may support video-based advertisements," the patent application says."
Prior art: The original Realplayer. Freeware products have been doing this for more than a decade. It was a dumb idea then, its a dumb idea now.
"Oh no! My Windows machine screen is full up with porn advertisements"
You mean that their "ad framework" is an unpatched copy if Internet Explorer? I think there's a LOT of prior art on that one, from drive-by installs, malware, viruses, trojans, etc.
Slashdot Mirror
We did it at the last place I worked at. I didn't trust their figures, and going through the logs, it was obvious they were grossly inflating their claimed click-through results - by almost 2000%. When confronted with the hard evidence, they backed down, apologized, gave us a free month, etc. There were days that they claimed we had received plenty of click-throughs, but there wasn't a single log entry. In other words, they weren't accurately parsing their log files (they were running their servers on Windows, so who knows what they were doing ...) or they were committing out-and-out fraud. Our test clicks showed up in the log files as being relayed via their server, so they couldn't say that there was something wrong with our setup.
2 years later, they're out of business. I guess others cottoned on to the same thing ... or (more likely) my old boss started bragging about how he caught them fudging their click-through rates and page views, and others started putting their own checks into place.
Click fraud is a problem, but so is site fraud.
Bullshit - you have the right to break up the software into its distinct files and use them as you will. What you don't have the right to is to make multiple copies (that's what copyright is about) without permission. How you use your legal copy is entirely up to you. Its like those bogus licensing provisions that Microsoft wants you to think are legally binding - like "you must install this sticker on the computer to which the OS is installed." You don't have to. Anyone who says otherwise is a sheeple.
I said "if the proxy isn't broken" - caching proxies that serve up urls with parameters that differ for each request are clearly broken. If you're generating the parameters dynamically/pseudo-randomly (which you should be doing) they shouldn't be cached.
If you use a javascript xmlhttprequest to fetch the ad content from your server(a 2-line paste on the web page, to include your source script, which generates the pseudorrandom parameters for each request, and less than 20 lines on the host), then caching proxies don't even come into it, and you get an accurate-enough count.
Calm down ... its not like there isn't a Microsoft product that doesn't leak ...
How do I know this? How many paintings have I looked at? I know it by the definition of the vast majority of paintings. They tend to involve (1) strokes and colors in many variations and plotlines, (2) subjects and lighting, and (3) artist control of the outcome. I don't think these attributes have much to do with art; they have more in common with video games.
Both arguments are BS, except that, unlike Ebert, mine was crap by intent.
I give him two thumbs down. Or one finger up, if you want to be artsy-fartsy.
Or serve the image from your own server, and parse your logs for requests.
Or host the text ads yourself.
Or make the text ad a block of javascript that goes to your server and serves up custom ads, like the "big boys" do, and embed it in the html of a web page. There are dozens of ways - just ask any spammer, they have LOTS of experience with this sort of stuff.
If they did as I pointed out, the problem of caches is fixed unless the caching mechanism is broken. Putting parameters after the url is the usual way its implemented, and it tells the caching proxy that this is NOT static content, and to pass through the request instead of using the cached object. If they don't re-fetch your ad, you don't pay, since if they don't fetch it, they never see it.
Also, it doesn't matter if the average user cycles through a dozen IPs, over the course of a day - if they view your ad once, because it will come from just one of those IPs, not multiple IPs. Of course, you can also just look at the different IPs the various requests for different parts of the page are passed through, and build up a map of IPs that all front for the same user.
No solution is perfect, but this sure is better than nothing.
" Agent: Alexa
Deny: / "
why not do a a mod_rewrite rule for Alexa, sending them back to alexa.com, or better yet, the DHS. Be nice to see Alexa classified as a "terr'rist". (yeah, I know there are problems with this theory, but we can dream, right :-)
If the site says "We'll get you N page views per week" for X period of weeks, and the first week comes in at N/4, you cancel the rest. Anyone relying on Alexa rather than demanding a performance guarantee is just being dumb.
"It isn't surprising that people who spend money on advertising want to have some metric by which to predict (estimate, guess, what-have-you) the impact of each dollar spent on web advertising."
There are several easy ways:
If you're so naive as to not insist on hard numbers for actual views (the log files are best , you deserve to get hosed - you can analyse the log files and factor out multiple views per host ip to get the actual number of real views, and reduce fraud; ditto with geolocation of ip addresses to factor out bots in 3rd world countries; ditto for bots that crawl every link on a page; ditto for pages that are loaded then immediately dumped for another page).
As an advertiser, I'd want unique eyeballs - real human eyeballs - that can be verified.
It might not be news in the sense of something that just happened, but its still worth being reminded about once in a while.
Its true - the people who obsess over PageRank and SEO are the ones driving this - everyone already knows their own web stats just by looking at their server logs.
"If I were them I'd create a closed network with educational sites alone. "
They don't call it "sex education" for nothing, you know ...
I guess someone didn't get the email - the internet is for porn. (And for those kids in Nigeria using OLPCs, they'll just search for "pr0n", and br3asts - after all, look how they spell other words, like V14GR4).
Real porn (as in obscene) to them will be seeing just how many obese people there are in the world. People who claim to be preaching th word of god, and want to end world hunger, but look like they personally caused at least one famine by eating everything in sight. Isn't gluttony a sin?
"Err, how many anti-war (or anti anything) folks are going to be transferring money overseas"
"Copying them from your Windows machine is pirating the software." If he removes them from his Winbox to his linbox he's not pirating ... I'm sure a lot of us have legit copies of older versions of windows that we don't use any more, but that we can use the codecs, fonts, etc.
"a couple gigs of RAM "
Oh, come off it ... 640 megs should be enough for anybody.
Patents for software or business methods are bogus. They have zero to do with the original patent language, and are a "feature" of the USPTO that attests more to the bugginess of the process than anything else.
Copyrights for software make sense. It is code, it is copywriteable. But patented? Its not an invention.
Besides, you fail to address my main point - that because Microsoft has done nothing to reduce its alleged losses, specifically by identifying the alleged infringements, the doctrine of laches means they can't succeed with a claim any more. If they were to file a lawsuit tomorrow, they would have to prove that the lawsuit was over a patent OTHER THAN THE 235 "allegedly infringed" patents that are now lawsuit-proof. So they would ahve to name 236 patents, point to the first 235 and say, "it was not one of those", etc.
They will never successfully sue. They know it. They blew any basis for a successful leagal claim by milking it for fud value, rather than disclosing what it was (since it never existed anyway, they couldn't ...).
It doesn't matter in the long run, because in the long run (20 years), Microsoft will be a niche player, probably bought out by Apple, since they can no longer produce competitive products.
"So what if I was? That has no bearing on who I am. My identity is the sum of the things I've done. Really, the only thing you have to prove is that I didn't do things under a different identity that are naughty."
Really ... considering that it affects what you can inherit ...
Beside, have you ever tried to prove a negative?
It looks like we might be heading towards a different sort of singularity than was previously envisioned. A world where the US self-destructs (just look at the debt clock), ripped apart by capitalizm mixed with profiteering and greed among the well-connected, and the RotW (Rest of the World).
Some of the economic projections for later this century are downright ugly - with a US with a population of up to half a billion, but a permanently wrecked economy, as we hit the upper limits of what's possible, and the rest of the world catches up, then surpasses the US in terms of GNP. Both China and India will have larger economies by the middle of the century - and less debt.
"Uh. Rigging is the whole point of intellectual monopoly law. It's antithetical to free market capitalism. I don't know why you're surprised, america - you're about as capitalist as the soviets were communist."
People keep forgetting that the Internet routes around damage. There is no such thing as a monopoly when the whole operation can be moved to servers in another country. You're a seller in the US, and you want to sell to someone else in the US - just auction it off on ebay in the UK or Canada or france.
Do you really expect everyone who has a stake in linux NOT to join in the case so as to protect their own interests? Would IBM sit on the sidelines and let a judgement go through without unleashing a few of the nazgul?
Also remember, its a LOT cheaper to defend than to pursue.
That last bit - the obligation of the injured party to mitigate damages - will kill Mr. Softie, since they have gone out of their way NOT to mitigate damages by disclosing any infringing patents. The doctrine of laches means they are f$cked.
True identification IS impossible on the Internet. Any and all solutions have cracks in it. Heck, you can't even prove your identity in "real life" unless you've been fingerprinted at birth. Prove who you really are. Can you prove that you weren't switched at birth? That you weren't really adopted, and your adoptive parents never told you (or the converse - that your parents told you you were adopted, but lied - to cover up, for example, that your uncle is really your father ...)
Sure, people have known you for years and will vouch for your identity - but people have killed other people, stolen their identities, and then after enough time passed, people vouchsafed them as the "real".
Microsoft will never sue - they know that the only thing they can do is amke noises. Actually suing would be the equivalent of a first strike in a MAD - Mutually Assured Destruction - scenario, which they would ultimately lose.
The resulting positive publicity for linux would further erode their already slipping grip on their customer base. Like it did with allthe SCO BS.
"The software would also free advertising from its traditional browser yoke. "A word processor may display a banner ad along the top of a window, similar to a toolbar, while a graphical ad may be displayed in a frame associated with the application. A digital editor for photos or movies may support video-based advertisements," the patent application says."
Prior art: The original Realplayer. Freeware products have been doing this for more than a decade. It was a dumb idea then, its a dumb idea now.
"Oh no! My Windows machine screen is full up with porn advertisements"
You mean that their "ad framework" is an unpatched copy if Internet Explorer? I think there's a LOT of prior art on that one, from drive-by installs, malware, viruses, trojans, etc.