You said our software could NEVER beat an open source system. I don't need your test data. I know your statement is false.
I was simply asking if you were willing to back up your statement with a risk free wager.
At this point, it's apparent that all you are trying to do is get a consulting fee to prove to me that you are wrong. I have no interest in doing that as I already know you are wrong.
so you just change the subject rather than answer my simple questions.
you were the one who wrote (emphasis mine): "any commercial system is that it will NEVER do as good..."
since we can never beat you, you have NOTHING to risk by accepting my money.
Yet you refuse to accept my money. I don't understand why anyone would refuse $10,000 that is risk free. After all, by your own statement, we have NO CHANCE of winning the contest. Why do you refuse to take a risk-free $10,000??? What's the logic behind that?
I know you aren't selling anything. You just are making outrageous claims saying that it is impossible for our system to be better than open source software. I simply asked if you had any confidence in the claim which you could prove by accepting my risk free offer.
Suppose I offered you the following: "if you respond to this post, I'll pay you $10,000."
If I offered that, would you accept? There is no downside to accepting. You cannot lose money.
I'll put up $10K and ls671 will put up $10K into an escrow account. We'll run the same realtime mailstream through both systems for 24 hours and if you get a lower total error count, you win the $10K. If you don't, I get the money.
So ls671, this is easy money since you said (in another post) we'd NEVER be able to match a system constructed of free components. If that were true, you'd accept my bet instantly because you'd always win. Please accept my bet and post your acceptance here. Or post a retraction that open source will always beat commercial systems. What's your choice?
I'll put up $10K and ls671 will put up $10K into an escrow account. We'll run the same realtime mailstream through both systems for 24 hours and if you get a lower total error count, you win the $10K. If you don't, I get the money.
So ls671, this is easy money since you said we'd NEVER be able to match a system constructed of free components. If that were true, you'd accept my bet instantly because you'd always win. Please accept my bet and post your acceptance here. Or post a retraction. What's your choice?
One other thing i forgot to respond to. Someone said our system is trivial to defeat... the spammer just puts all the recipients in a bcc.
The person who wrote that assumed we work after the mail hits your MTA. We don't. We work at the SMTP level, before the mail hits your mail server. So the RCPTO information is ALWAYS there and cannot be obfuscated. If the spammer omits this, his spam won't be delivered to anyone.
So yeah, the spammer can defeat the system by not including any recipients. But that's really easy to filter out! Since there are no recipients, we eat the mail.
It isn't disclosed, because if we did, everyone would copy us. You're right though, from the description you cannot figure out how we do it.
But you know that honeypot systems work. How do they do it? The answer is very crudely..they look for content similarity. We do it in a much faster, more accurate, and elegant way.
slashdot doesn't make it easy for me to respond to each comment. I am told to "hold on cowboy" and wait between postings. So I'll answer all the rest of the comments in this email.
a couple of people said the spammers just find the good addresses and only send to them. The problem with that is that the good addresses then turn into bad addresses and the spammer loses. Fundamentally, they cannot avoid the mathematical fact that they MUST send to people who get more spam than senders who send ham. So that might work for one spammer for a few mailings, if they could pull it off, but the victory would be very short lived. And no spammer would want to limit themselves to such a small list of recipients.
one person asked about what is a sender and what is a message. That's right. That wasn't easy to figure out. Suffice it to say that the explanations on the website give you only a basic understanding. The secret sauce is secret...until the patent issues.
another person said disclose it to prove it is spammer proof. What is the economic value in doing that? then every competitor would copy it and my company would be driven out of business since the intellectual property would be then be worthless. If you want to pay us $100M, we'll publish the algorithm. That's far less than the economic value of the invention. Any takers?
that same person said it can't possibly work in the real world. That is simply ignorant of the facts in front of you. Call the customers on our website. Some have been using it for more than a year with no algorithm updates and it is working better now than a year ago. We're about to announce a major state school system has standardized on our software for all their campuses. How could that happen if our stuff doesn't work in the real world? We sure didn't give it away. All the customers pay full price or close to that. We rarely discount. And our prices are higher than our competition.
one person asked why/how Yahoo could send to a spamtrap. I just sent to a spamtrap from my yahoo account just now. You can do it too. Spammers who get Yahoo accounts do it all the time just like I proved you can do it. And there are no stats on a "generic honeypot algorithm." each implementation is different. I don't know of any that have less than 200 errors per million messages. Do you??
Finally, the last person said my system will NEVER do as well as combining the freely available tools to fight spam. This person then didn't give numbers (like I did). And I don't think this person is telling the truth either. So I challenge that person (ls671) to prove it. I'll put up $10K and ls671 will put up $10K into an escrow account. We'll run the same realtime mailstream through both systems for 24 hours and if you get a lower total error count, you win the $10K. So ls671, this is easy money since you said we'd NEVER be able to match a system constructed of free components. If that were true, you'd accept my bet instantly because you'd always win. So this is easy money. Please accept my bet and post your acceptance here. Or post a retraction. What's your choice?
The experts spent about 3 hours each going over the algorithm.
Some of our customers are listed on the website.
Your attack is clever, and of the attacks listed, it is the best thought out, but it won't work, but I can't tell you why without revealing details of the algorithm. I'm not willing to do that for competitive reasons.
Every algorithm can be attacked. I only know of one attack that would work on the core algorithm, but the attack is easy to both detect and defend.
Thank you for all the comments on the NY Times article.
It would be difficult for me to answer each and every comment, so I'll try to just hit the high points here.
It's quite easy to poke fun at an algorithm which is unknown to you as demonstrated by all the comments.
But what's more relevant is whether really smart people who know the algorithm can find fault with it. There are only two people outside of Abaca who know the algorithm: Stephen Wolfram (author of Mathematica) and University of Waterloo Professor Gordon V. Cormack (a well known figure in the anti-spam community). I picked Wolfram because he's the smartest pure math guy I know. I picked Cormack because I think is one of the smartest and most respected scientists in the spam field. You could contact either of them and ask them what they think of the approach. I can tell you what they'd say if you did that. They'd tell you it is a simple, elegant algorithm that has no obvious (to them) holes. I know that because the reason I disclosed it to them was to see if I overlooked anything. Neither found any holes. That doesn't prove that there aren't holes. All systems have holes. What this does mean is that a couple of pretty respected experts think it appears to be pretty solid logic.
In fact, Gordon was kind of enough to go even further and gave me permission to use the follow quote: "This is, by far, the most clever technique I'm aware of for spam filtering." Since Gordon is conference chair for a lot of spam conferences, this is a pretty significant endorsement from someone who KNOWS the full algorithm and who knows the spam space better than just about anyone.
I spent about 4 years studying what others had done in the space. As one commenter pointed out, the recipient reputation system can be thought of as a generalization of the honeypot technique that was first patented by Brightmail.
That's exactly right. My realization is that every email address has statistical value, not just honeypots. So instead of just "black" feedback, the system incorporates "grey" and "white" feedback; every recipient has an apriori odds associated with receiving mail. For many years, Brightmail was the "defacto" standard for spam filtering. Brightmail is just a special case of the algorithm I invented. So instead of learning from honeypots, we learn from ALL recipients and incorporate that statistical input in a mathematically rigorous way in order compute a statistical likelihood that our prediction was correct. That gives us much more input than a honeypot system: it gives us white, black, and grey values. That is critical to avoiding false positives because good sites (like Yahoo and Hotmail) send email to honeypots all the time. And we incorporate that feedback into a statistical framework that is much more accurate than what Brightmail used.
Exactly how we incorporate that input into spam scoring has not been publicly disclosed. It is not obvious.
People who say that this must be snake oil or cannot work ignore the fact that the system has been in use by real customer for more than a year. We have over 100 customers and are just annoucing our existence to the world, so that number should increase quite rapidly now that we are starting to market our product. There are customer testimonials on our website. You can contact them directly to verify that these quotes are legitimate.
Here are statistics from one of our rating servers. There were 1,380,140 messages since the last counter reset. 96% were rated spam. There were 176 false positives and 66 false negatives reported. I just grabbed those stats from one of our live servers right now as I was composing this message. Sometimes we're better, sometimes we're worse, but those numbers are pretty typical.
It's not perfect, but I think those are pretty good error rates for where we are now. And the stats always get better as we add more customers since we get more statistical input and this is just a statistical estimation problem. The more data, the more accurate
That won't work. Propel is targetted to dialup customers and we charge a fixed monthly price. It transparently goes into "bypass mode" when you are on a high speed line. This was a financial decision. When Propel provides hosting, we have to pay bandwidth charges. If your ISP provides the hosting, they incurr no incremental bandwidth charges (in fact, their bandwidth decreases). So Propel for Broadband is an ISP service. Nobody is offering this today, but they could.
We can only claim what we know, not what we don't know. We know we're the fastest of all the dialup accelerator products we know about based on their latest release that is available. So yes, our investors would be pleased to accept your donation in an attempt to disprove that statement.
That's right, ultimately are limits. The techniques used by Propel and others like it probably will not be able to achieve more than a 4X improvement (on average) from today's "unassisted" dialup. So there is still some room for further improvement. I expect we'll go from a 3X improvement to day to a 4X improvement sometime in the next 12 months.
The PacWest Dial Broadband product will gzip the HTML and compress the graphics on a remote proxy server that you directly connect to. But the performance benefits on HTML are less than what gzip gives you because you're comparing modem compression vs. modem compression of gzipped content. You also have to make and break TCP/IP connections all the time.
The two advantages to this approach are 1) that there is nothing to download and 2) it will work with browsers that support gzip on any platform.
The two downsides are that 1) you'll see less than a 2X performance gain (probably like 1.5X on average) compared to a average of 3X performance gain with Propel (and often 5X) which requires a download since it uses proprietary compression techniques and 2) if that proxy server ever crashes or is unavailable because of a network outage to the site of the proxy server, a user will be completely hosed. That never happens with Propel due to a safety bypass that is completely transparent.
If you really want speed on a dialup, the PacWest product isn't the best way to get it, but it is better than nothing.
not quite catchup. Earthlink is totally leapfrogging AOL. Propel/Earthlink is, on average, more than 3 times faster than AOL with many pages loading 5 times faster or more.
See this posting for details on how it works. AOL does nothing like this.
I'm the CEO of Propel and the inventor of our patent-pending compression algorithm. Formerly, I was CEO of Infoseek, where I designed the Infoseek search engine. I'd like to explain to all of you how Propel really works since there has been way too much misinformation on this thread. A lot of this is trade secret so I can only discuss the high level concepts, but that should be sufficient to prove that "it can be done."
First, it isn't snake oil. It really does improve the actual time to download web pages substantially. Many "real world" web pages download 3, 4, or 5 times faster than before.
How fast is it?
The eBay home page loads more than 5 times faster (from 7.75 seconds to 1.5 seconds). The CNet home page loads almost 9 times faster when Propel is used (from 36.4 seconds to 4.3 seconds). On average, that is for a wide range of web pages, you can expect that the pages will load from 2 to 5 times faster than they did before you installed Propel. You should try it and see for yourself.
Or if you are sure this can't be done, then I invite you to take advantage of the following opportunity: You bet me any amount of money up to $1M dollars. I'll show you a 5X improvement in actual web page download speed on a selection of popular web pages from top 100 websites (not some contrived test pages, but real web pages). If you can prove it is some kind of hoax, I pay you. If I get 5X or more speed improvement compared to dialup without Propel, you pay me the amount you wagered. We'll use the exact same PC for the tests with the caches in exactly the same state connected to the same ISP at the same baud rate. Modem compression on or off, your choice. No tricks. Minimum bet is $1,000. If you are confident this is a trick, this is a quick way to make a lot of money really fast. We'll invite the press to audit it. Any takers?
Or you can read the review on CNet. They tested it and wrote in their review:
When we used Propel Accelerator to download Web pages, they arrived two to three times faster than with a standard 56K connection. In some cases,
pages displayed more swiftly than on a high-speed digital subscriber line (DSL). Truly, we were amazed.
You have to remember that that review was written a long time ago. We've almost doubled our speed since that CNet review was written. That's why our claim of 2 to 5x on the latest version of Propel is consistent with the review.
How it really works
How do we do this magic? Through at least 20 techniques including persistent connections, up and down header compresssion, caching that is combined with diffing of HTML and graphics, and the compression (either lossless and lossy depending on the filetype and user settings) of filetypes that most people "think" cannot possibly be compressed any further. For example, think that Flash cannot be compressed because it already is? Think again.
Web pages consist of HTML and graphics primarily. We compress all the various datatypes and decompress on your local machine. Comments that "you can't compress graphics because they are already compressed" are from people who are misinformed. For example, Jpeg2K provides much higher compression for equivalent quality level than jpg does. And LZW is hardly the best compression scheme for GIF graphics. That technology was invented a long time ago and there are much better lossless compression algorithms for such files. For example, PNG is better than GIF but (surprise) there are propriety file formats that are superior to PNG.
If you are willing to tolerate quality degradation, you can compress even more. Propel has a slider so you can set your tolerance threshold for graphics (text and HTML are always lossless).
For HTML compression, we use our patent-pending lossless technique that makes full use of the cache in your browser by allowing us to reference text fragments o
Slashdot Mirror
the accelerator never claims to run on linux. so microsoft windows is snake oil too, right?
i'm making you the same offer for Propel then. Name your wager. Since you are right, this is a no risk deal for you.
You said our software could NEVER beat an open source system. I don't need your test data. I know your statement is false. I was simply asking if you were willing to back up your statement with a risk free wager. At this point, it's apparent that all you are trying to do is get a consulting fee to prove to me that you are wrong. I have no interest in doing that as I already know you are wrong.
No problem. I'll pay you to run the tests, but it'll be in an escrow account along with YOUR money of the equivalent amount.
If you lose, you get nothing and you lose your wager.
But you can't lose since it is IMPOSSIBLE for us to do better than open source software.
Deal?
Oh, you never said I didn't offer you enough money. So how much would you like to raise the bet to for you to accept?
so you just change the subject rather than answer my simple questions.
you were the one who wrote (emphasis mine): "any commercial system is that it will NEVER do as good..."
since we can never beat you, you have NOTHING to risk by accepting my money.
Yet you refuse to accept my money. I don't understand why anyone would refuse $10,000 that is risk free. After all, by your own statement, we have NO CHANCE of winning the contest. Why do you refuse to take a risk-free $10,000??? What's the logic behind that?
I know you aren't selling anything. You just are making outrageous claims saying that it is impossible for our system to be better than open source software. I simply asked if you had any confidence in the claim which you could prove by accepting my risk free offer.
Suppose I offered you the following: "if you respond to this post, I'll pay you $10,000."
If I offered that, would you accept? There is no downside to accepting. You cannot lose money.
I don't need you to test it. I know it works.
You claimed a commercial system could NEVER beat open source tools.
You were the one with the claim.
I asked if you'd put money behind your NEVER claim.
You declined.
The point is I'm willing to stand behind my product claims. You aren't.
those were the only ones. You didn't respond to the first one, so i posted two more.
I'll put up $10K and ls671 will put up $10K into an escrow account. We'll run the same realtime mailstream through both systems for 24 hours and if you get a lower total error count, you win the $10K. If you don't, I get the money.
So ls671, this is easy money since you said (in another post) we'd NEVER be able to match a system constructed of free components. If that were true, you'd accept my bet instantly because you'd always win. Please accept my bet and post your acceptance here. Or post a retraction that open source will always beat commercial systems. What's your choice?
I'll put up $10K and ls671 will put up $10K into an escrow account. We'll run the same realtime mailstream through both systems for 24 hours and if you get a lower total error count, you win the $10K. If you don't, I get the money.
So ls671, this is easy money since you said we'd NEVER be able to match a system constructed of free components. If that were true, you'd accept my bet instantly because you'd always win. Please accept my bet and post your acceptance here. Or post a retraction. What's your choice?
One other thing i forgot to respond to. Someone said our system is trivial to defeat... the spammer just puts all the recipients in a bcc.
The person who wrote that assumed we work after the mail hits your MTA. We don't. We work at the SMTP level, before the mail hits your mail server. So the RCPTO information is ALWAYS there and cannot be obfuscated. If the spammer omits this, his spam won't be delivered to anyone.
So yeah, the spammer can defeat the system by not including any recipients. But that's really easy to filter out! Since there are no recipients, we eat the mail.
It isn't disclosed, because if we did, everyone would copy us. You're right though, from the description you cannot figure out how we do it.
But you know that honeypot systems work. How do they do it? The answer is very crudely..they look for content similarity. We do it in a much faster, more accurate, and elegant way.
slashdot doesn't make it easy for me to respond to each comment. I am told to "hold on cowboy" and wait between postings. So I'll answer all the rest of the comments in this email.
a couple of people said the spammers just find the good addresses and only send to them. The problem with that is that the good addresses then turn into bad addresses and the spammer loses. Fundamentally, they cannot avoid the mathematical fact that they MUST send to people who get more spam than senders who send ham. So that might work for one spammer for a few mailings, if they could pull it off, but the victory would be very short lived. And no spammer would want to limit themselves to such a small list of recipients.
one person asked about what is a sender and what is a message. That's right. That wasn't easy to figure out. Suffice it to say that the explanations on the website give you only a basic understanding. The secret sauce is secret...until the patent issues.
another person said disclose it to prove it is spammer proof. What is the economic value in doing that? then every competitor would copy it and my company would be driven out of business since the intellectual property would be then be worthless. If you want to pay us $100M, we'll publish the algorithm. That's far less than the economic value of the invention. Any takers?
that same person said it can't possibly work in the real world. That is simply ignorant of the facts in front of you. Call the customers on our website. Some have been using it for more than a year with no algorithm updates and it is working better now than a year ago. We're about to announce a major state school system has standardized on our software for all their campuses. How could that happen if our stuff doesn't work in the real world? We sure didn't give it away. All the customers pay full price or close to that. We rarely discount. And our prices are higher than our competition.
one person asked why/how Yahoo could send to a spamtrap. I just sent to a spamtrap from my yahoo account just now. You can do it too. Spammers who get Yahoo accounts do it all the time just like I proved you can do it. And there are no stats on a "generic honeypot algorithm." each implementation is different. I don't know of any that have less than 200 errors per million messages. Do you??
Finally, the last person said my system will NEVER do as well as combining the freely available tools to fight spam. This person then didn't give numbers (like I did). And I don't think this person is telling the truth either. So I challenge that person (ls671) to prove it. I'll put up $10K and ls671 will put up $10K into an escrow account. We'll run the same realtime mailstream through both systems for 24 hours and if you get a lower total error count, you win the $10K. So ls671, this is easy money since you said we'd NEVER be able to match a system constructed of free components. If that were true, you'd accept my bet instantly because you'd always win. So this is easy money. Please accept my bet and post your acceptance here. Or post a retraction. What's your choice?
If it worked on a per message basis, you'd be right. But it doesn't.
We'd NEVER get those performance numbers (1 mistake per 5,000 emails) if we did it that way.
The experts spent about 3 hours each going over the algorithm.
Some of our customers are listed on the website.
Your attack is clever, and of the attacks listed, it is the best thought out, but it won't work, but I can't tell you why without revealing details of the algorithm. I'm not willing to do that for competitive reasons.
Every algorithm can be attacked. I only know of one attack that would work on the core algorithm, but the attack is easy to both detect and defend.
Thank you for all the comments on the NY Times article.
It would be difficult for me to answer each and every comment, so I'll try to just hit the high points here.
It's quite easy to poke fun at an algorithm which is unknown to you as demonstrated by all the comments.
But what's more relevant is whether really smart people who know the algorithm can find fault with it. There are only two people outside of Abaca who know the algorithm: Stephen Wolfram (author of Mathematica) and University of Waterloo Professor Gordon V. Cormack (a well known figure in the anti-spam community). I picked Wolfram because he's the smartest pure math guy I know. I picked Cormack because I think is one of the smartest and most respected scientists in the spam field. You could contact either of them and ask them what they think of the approach. I can tell you what they'd say if you did that. They'd tell you it is a simple, elegant algorithm that has no obvious (to them) holes. I know that because the reason I disclosed it to them was to see if I overlooked anything. Neither found any holes. That doesn't prove that there aren't holes. All systems have holes. What this does mean is that a couple of pretty respected experts think it appears to be pretty solid logic.
In fact, Gordon was kind of enough to go even further and gave me permission to use the follow quote: "This is, by far, the most clever technique I'm aware of for spam filtering." Since Gordon is conference chair for a lot of spam conferences, this is a pretty significant endorsement from someone who KNOWS the full algorithm and who knows the spam space better than just about anyone.
I spent about 4 years studying what others had done in the space. As one commenter pointed out, the recipient reputation system can be thought of as a generalization of the honeypot technique that was first patented by Brightmail.
That's exactly right. My realization is that every email address has statistical value, not just honeypots. So instead of just "black" feedback, the system incorporates "grey" and "white" feedback; every recipient has an apriori odds associated with receiving mail. For many years, Brightmail was the "defacto" standard for spam filtering. Brightmail is just a special case of the algorithm I invented. So instead of learning from honeypots, we learn from ALL recipients and incorporate that statistical input in a mathematically rigorous way in order compute a statistical likelihood that our prediction was correct. That gives us much more input than a honeypot system: it gives us white, black, and grey values. That is critical to avoiding false positives because good sites (like Yahoo and Hotmail) send email to honeypots all the time. And we incorporate that feedback into a statistical framework that is much more accurate than what Brightmail used.
Exactly how we incorporate that input into spam scoring has not been publicly disclosed. It is not obvious.
People who say that this must be snake oil or cannot work ignore the fact that the system has been in use by real customer for more than a year. We have over 100 customers and are just annoucing our existence to the world, so that number should increase quite rapidly now that we are starting to market our product. There are customer testimonials on our website. You can contact them directly to verify that these quotes are legitimate.
Here are statistics from one of our rating servers. There were 1,380,140 messages since the last counter reset. 96% were rated spam. There were 176 false positives and 66 false negatives reported. I just grabbed those stats from one of our live servers right now as I was composing this message. Sometimes we're better, sometimes we're worse, but those numbers are pretty typical.
It's not perfect, but I think those are pretty good error rates for where we are now. And the stats always get better as we add more customers since we get more statistical input and this is just a statistical estimation problem. The more data, the more accurate
That won't work. Propel is targetted to dialup customers and we charge a fixed monthly price. It transparently goes into "bypass mode" when you are on a high speed line. This was a financial decision. When Propel provides hosting, we have to pay bandwidth charges. If your ISP provides the hosting, they incurr no incremental bandwidth charges (in fact, their bandwidth decreases). So Propel for Broadband is an ISP service. Nobody is offering this today, but they could.
We can only claim what we know, not what we don't know. We know we're the fastest of all the dialup accelerator products we know about based on their latest release that is available. So yes, our investors would be pleased to accept your donation in an attempt to disprove that statement.
That's right, ultimately are limits. The techniques used by Propel and others like it probably will not be able to achieve more than a 4X improvement (on average) from today's "unassisted" dialup. So there is still some room for further improvement. I expect we'll go from a 3X improvement to day to a 4X improvement sometime in the next 12 months.
The two advantages to this approach are 1) that there is nothing to download and 2) it will work with browsers that support gzip on any platform.
The two downsides are that 1) you'll see less than a 2X performance gain (probably like 1.5X on average) compared to a average of 3X performance gain with Propel (and often 5X) which requires a download since it uses proprietary compression techniques and 2) if that proxy server ever crashes or is unavailable because of a network outage to the site of the proxy server, a user will be completely hosed. That never happens with Propel due to a safety bypass that is completely transparent.
If you really want speed on a dialup, the PacWest product isn't the best way to get it, but it is better than nothing.
None of the technologies used in Propel are available in modern browsers. If they were, then it would be impossible to do a 5X performance gain. See for details on how Propel achieves a 5X performance increase.
See this posting for details on how it works. AOL does nothing like this.
First, it isn't snake oil. It really does improve the actual time to download web pages substantially. Many "real world" web pages download 3, 4, or 5 times faster than before.
How fast is it?
The eBay home page loads more than 5 times faster (from 7.75 seconds to 1.5 seconds). The CNet home page loads almost 9 times faster when Propel is used (from 36.4 seconds to 4.3 seconds). On average, that is for a wide range of web pages, you can expect that the pages will load from 2 to 5 times faster than they did before you installed Propel. You should try it and see for yourself.
Or if you are sure this can't be done, then I invite you to take advantage of the following opportunity: You bet me any amount of money up to $1M dollars. I'll show you a 5X improvement in actual web page download speed on a selection of popular web pages from top 100 websites (not some contrived test pages, but real web pages). If you can prove it is some kind of hoax, I pay you. If I get 5X or more speed improvement compared to dialup without Propel, you pay me the amount you wagered. We'll use the exact same PC for the tests with the caches in exactly the same state connected to the same ISP at the same baud rate. Modem compression on or off, your choice. No tricks. Minimum bet is $1,000. If you are confident this is a trick, this is a quick way to make a lot of money really fast. We'll invite the press to audit it. Any takers?
Or you can read the review on CNet. They tested it and wrote in their review:
You have to remember that that review was written a long time ago. We've almost doubled our speed since that CNet review was written. That's why our claim of 2 to 5x on the latest version of Propel is consistent with the review.
How it really works
How do we do this magic? Through at least 20 techniques including persistent connections, up and down header compresssion, caching that is combined with diffing of HTML and graphics, and the compression (either lossless and lossy depending on the filetype and user settings) of filetypes that most people "think" cannot possibly be compressed any further. For example, think that Flash cannot be compressed because it already is? Think again.
Web pages consist of HTML and graphics primarily. We compress all the various datatypes and decompress on your local machine. Comments that "you can't compress graphics because they are already compressed" are from people who are misinformed. For example, Jpeg2K provides much higher compression for equivalent quality level than jpg does. And LZW is hardly the best compression scheme for GIF graphics. That technology was invented a long time ago and there are much better lossless compression algorithms for such files. For example, PNG is better than GIF but (surprise) there are propriety file formats that are superior to PNG.
If you are willing to tolerate quality degradation, you can compress even more. Propel has a slider so you can set your tolerance threshold for graphics (text and HTML are always lossless).
For HTML compression, we use our patent-pending lossless technique that makes full use of the cache in your browser by allowing us to reference text fragments o