Re:No Knee-jerk Privacy responses please...
on
Twist on DNA Privacy
·
· Score: 2, Interesting
Supposed a witness had pointed to someone in a line-up and said "it isn't him, but he's real close, almost like his
brother" and the police went and investigated the
brother. Nobody would have any objection.
There is a danger with retained DNA databases, but
I don't see it in this case.
As the number of "usual suspects" grows, eventually
we will have someone who has a "1 in a million" match
against the perpretators DNA.
I doubt that the prosecution will reveal to the jury that
the defendant was one of 2 two million "undesriables"
that the police keep the DNA on.
A DNA match against a sample recovered at a crime
scene is never perfect. When it matches to a high degree
of certainty someone who was already a suspect anyway
then it is a marvelous confirmation tool. (Indeed, I believe
DNA evidence has cleared more suspects than
it has convicted.)
But when DNA evidence is used to search vasts databases,
the statistical justifications that this is "evidence" rapidly dwindle, or possibly even evaporate.
And sadly, we cannot count on juries, and perhaps not
even judges, to understand the impact of how a sample
was selected in determining whether a test is statistically signifigant.
If there are 10 million people in a city, then a test that
has "less than 1 in a million chance" of a false positive
can be relied upon to falsely finger up to 10 people.
That's reasonable grounds for a search. But I sure hope
judges and juries don't buy it as 'proof'.
For better or worse, the concept of selling a closed platform is legal. This is especially true if the buyer
has adequate information to know that it is a closed
platform. I would prefer a mandatory big red
WARNING label to be affixed to all closed platforms
saying "The retail price of this unit reflects a subsidy
from the manufacturer. This subsidy is provided in
anticipation of future revenues. Therefore this unit
will only work with software lisenced by the manufacturer."
There are benefits to a closed platform to consumers.
It allows a manufacturer who is optimistic to assume the risk that there will be an inadequate supply of software
for the platform. If they don't produce their software, they
just ate their subsidy.
It allows all software to be signed and authenticated.
Responsibility for any hardware damage caused by a
software installation can be easily assigned to the
lisencee who supplied the software.
The down side is simple. The consumer is being mislead
by an artificially low up-front price into being locked into
continued payments of a monopoly tax on each piece
of software they purchase.
I believe the only solution is for the FTC to require platform vendors to offer their product in an unbundled format.
You can buy an XBOX that will run third party software,
but it might cost you $150 more.
If I read the patent correctly, the only claim of innovation
is that a non-novel process (translation) is configured
based upno session variables.
I believe that Environment variables are clearly prior
art. As are web sites that deliver language-specific
pages based on user supplied information.
An algorithm that was specifically adapted to an
interactive dialogue could be patentable. For example,
if you had specific techniques to determine whether
the word "present" was referring to a gift or to a point
in time based upon prior exchanges in the conversation.
Of course I'd hate to see what a group of punsters would
do to the poor software.
The discussion has focused solely upon patent implicaions,
ignoring the real impact of this technology.
The ability to translate an IM converstation into
a foreign language, even one that might be understood
by parents such as English, will have a profoundly chilling
effect on IM converstaions.
Yes. Any web site that remembers your preferred edition,
language whatever. Indeed the whole point of a session
is to remember status rather than redundant including
the information in each message.
Depending on how the material was actually framed,
you might or might not get away with claiming that
Amazon was acting solely as a service provider.
But even promoting material, and displaying it in an
integrated fashion is far from "publishing". If it were
every bookstore owner would be a "publisher".
Would anyone expect a book store owner to validate
that the purported author of each book had not engaged
in plagarism? Why should we expect Amazon to do
something that we would not expect a bricks and mortar book store to do?
Even a publisher is entitled to some presumption
that an author has the right to sell what they are offering.
Conspicuously absent from the charges is any basis to believe that Amazon acted irresponsibly, or was negligent
in allowing these sites to exist. What were they supposed to do, conduct an image search for all visual material offered
for sale?
One thing that I have always been puzzled about on
web privacy is the belief that explicit legislation
or new rules are required to end these abuses.
If you read RFC 1945 section 12.3, it clearly states:
12.3 Abuse of Server Log Information
A server is in the position to save personal data about a user's requests which may identify their reading patterns or subjects of interest. This information is clearly confidential in nature and its handling may be constrained by law in certain countries. People using the HTTP protocol to provide data are responsible for ensuring that such material is not distributed without the permission of any individuals that are identifiable by the published results.
Now if you advertise your URL as "http://whatever",
are you not implicitly stating that you have a server
at that address that will comply with the http protocol?
So it seems to me that anyone who publishes a URL has
already stated that they will not abuse the personal information without obtaining consent, because to
do otherwise would be non-RFC compliant.
Ok, if I as an American buy music from the iTunes store, I'll be able to use that music on up to three computers. Right?
I cannot recall any restriction saying one of those couldn't be my laptop with me on a European trip, or even a
computer that I have at a European office.
So anyone who wants to obtain this material without going
through a "European" iTunes store can simply set up an account with an American ISP and buy it there, then transfer the file themselves.
Or decide that that is too much trouble and simply
steal it the old fashioned way.
Somebodey hasn't caught on that the purpose of the
iTunes store is to allow customers who want to
be honest to pay for the music.
There are major public interest factors that have to be considered.
At some point a dominant solution shifts from being a solution to a problem, to becoming part of the definition of the problem. Hence, I do not believe the market would benefit by allowing Intel to have monopoly rights to the x86 instruction set.
Once an infrastructure has grown up around a solution, such as compilers for a machine instruction set or a large user community that knows how to use a specific GUI, allowing ownership of that interface to continue is granting a monopoly over what is now a new market. Being able to
grow that market one round strikes me as fair, and it should be enough of a reward. Being able to keep all of
your customers until you aggravate them enough to justify retraining staff strikes me as an unfair advantage.
The damage done to a dominant provided that is copied by smaller companies is minor compared to the damage down to a small upstart who has their entire advantage stolen by a large competitor.
License agreements that prohibit informed discussion by consumer's about the quality of products would clearly be against the public interest. I doubt any court would uphold a contract that prevented a wide variety of people who did not have access to naturally confidential information from sharing basic reviews of a product's
functionality.
But I agree with your fundamental analysis. The judge saw an injustice, and wanted to find a solution somehow.
All good models, standards and specifications
take great care to avoid specifying how
something is to be done, rather than on the
externally visible behavior that others
rely upon.
The key challenge is to always ask yourself "Could this be done differently without affecting others?". And even more
aggressively "How can I redefine this interface so that I no
longer care if the other side uses algorithm B, instead of A?"
A GUI Design involves specifying a set of pages,
the information on each page, its arrangement,
the buttons that can be clicked, and the resulting transitions.
You can reasonably speak of a GUI "error" if
there are obvious flaws in the navigation between
the different screen/pages. For example, if you
cannot back out of the current step without
cancelling the entire transaction. Or if the information
on a screen does not fit within the window and the
window lacks scroll bars. Or if a "more info" button
leads to a screen that is either irrelevant, or which
cannot return back to the original screen.
If these are the types of errors that actually occurred,
and they were replicated, I would have to agree that
it is very convincing that theft occurred. I just don't
agree that a EULA has the right to forbid competitive
research. Competitive research is good, it benefits
consumers and should be encouraged. Not doing
your own research and evaluation is sloppy.
Making the same mistakes is a strong indicator
that no independent research or judgemetn was
exercised. It might even be enough proof to satisfy
me as a juror if there was no contrary evidence.
My initial read is somewhat similiar. A judge has come
up with faulty reasoning to support what seems right.
But they should come up with the correct rationale.
The rationale offered would be chilling.
If customers have the right to examine products, and
determine what they like and dislike about each, then
it isn't much of a leap to say that producers have the
same right to examine their competitor's products.
But there's a line somewhere between studying what
some product does, and essentially stealing its research.
Whatever the protection mechanism should be, it should
stop lazy companies from simply stealing interface designs
from other companies rather than paying to develop
them on their own.
So it is pretty much copyright infringement, except
that some allowance has to be made for the ability
of the market to clone interfaces from dominant
providers.
Copyright also provides an excellent insight into
what contract law must not be allowed
to create here. No author is allowed to sell their
mystery to the general public except that no
other mystery writer may read it for the purpposes
of evaluating what was effective or ineffective.
Indeed many writers want to aware of what others
have written, so they can ensure that they don't
inadvertently write something too similiar to
an existing book.
The bottom line is that the term "reverse engineering"
should never be applied to observing the external
behavior of a product. To me that term implies trying to
figure out how the product works, not to trying
to figure what the product does.
Slavishly copying what a product does, before
the product has an established user base, also strikes
me as improper copying. I'll admit I do not know how
to define that line. It may be similiar to judgement calls
made on when fictional characters have achieved
"cultural icon" status.
In the 15th century both Europe and China developed
deep sea fleets. The Chinese fleet was government
run for the glory of the Emperor. The European fleet
had scattered government financing, but was basically
run by greedy mercantilists.
The Emperor lost interest. The Chinese fleet was burned.
The greedy Euroean merchants continued to expand their
fleets leading to Europe's domination over the entire world.
Do we really need another "space race"?
Or is time to allow space development to proceed
without the sponsorship of modern emperors?
If I remember correctly, I first read this argument in an
essay by Poul Anderson. Unfortunately I couldn't find
a reference to it.
If you get past posturing being above politics, you'll
quickly realize that both parties do indeed have very
serious flaws and limitations.
But they are very different flaws.
If you think Democrats are prone to immunize large corporations from truthful labeling in the marketplace
then you haven't been paying attention.
There used to be a wing of the Republican party
that really believed in the strength of the market
system. They've been gone since Reagan. So it's
true that neither party has sufficient faith in a
true free market, but the ways that they interfere
with the market are very different.
And a physical device is just exploiting the laws of physics
that everyone should be free to use.
The goal is to ensure those that find things that
were unlikely to have been found anyway are
rewarded. Thereby encouraging both their
development, and the distribution of that
knowledge.
The current system is broken. It needs fixing.
Things that are obvious are patented,
statements of the problem are patented.
But if you eliminate all protection for complex algorithm development then you'll discourage their development.
A small company will be unable to deploy a new technique
and recoup its investment if large companies can simply
reverse engineer the solution and then deploy it on a
large scale.
Without software patents, you will quickly see:
Software being disquised as hardware.
Research being financed exclusively by large
companies that knew they could afford
to bring the resulting products to market themselves.
The research that was developed being kept
as trade secrets.
The guidance the technical feed needs to be providing
lawmakers is on how to properly handle technical
intellectual property -- not eliminating it.
Eliminating 'obvious' patents would also be desirable,
but it requires more domain knowledge to make that
call. Differentiating between a problem and a solution
should be a lot easier.
I remain convinced that the only defense against
'obvious' patents is to allow a public comment
period. It would allow challenges to the basis
of a patent without having to expose yourself
or hire a lawyer.
Shorterning the period for all "high tech"
inventions is probably a good idea. But I don't
see the point in penalizing an invention just
because it was realized in software as opposed
to being realized in hardware. In many cases,
the same essential algorithm can be done
either way.
The original application date is October 14, 1996.
It's a foreign application, but I believe that it is
still the relevant date. I also believe that it predates
transparent proxying in squid.
I believe the patent predates Squid, so there could be a problem to whatever degree that Squid infringes. Just
because a later developer is open source does not mean
that the original claim was invalid.
However, reading the patent carefully, you realize that
it actually only describes a very specific solution. Specific
enough that it truly is describing a solution, not a problem.
And specific enough that it might legitimately be considered novel for the time it was filed (I really don't
have time to search the source code of all proxy servers
in the 1996 time frame -- let someone with a finanicial
stake do that).
Specifically the patent deals with websites that are identified by their IP Address and where certain content (by default all) is held in an alternate (and presumably closer)
server.
There is nothing in this patent about determining if the
content is fresh. The description presumes that the cached
copies were pushed by the server.
So this would only seem to proxy servers that are transparent to the user, but not to the servers. The proxy servers that are of most interest to an ISP would either be
transparent to the server as well, or more of an akamai
style strategy where the first-response page is localized
to directly fetch pre-positioned material from edge caches.
Interestingly, the patent seems to be worded to cover a
single box which handles both the intercept and the
decision to proxy, but does not handle the actual
proxy response. A firewall transparently redirecting a port to a proxy server is prior art. The basic claim to being
novel here is that the client does not have to be configured
to use the proxy, and diversions only take place if certain
content is requested, non-proxied sites are passed through
"unaffected" (which is a false claim, BTW, which I'll deal with
in a moment).
There are some serious omissions in the description,
would could undermine its enforceability.
It speaks about identifying "requests" and forwarding
those that are not "web requests" to their original destination "unmodified". It fails to disclose that TCP
does not naturally delimit "requests", and that identification
of a complete "web request" is a complex matter.
It does not disclose that "other requests" are not amenable to the same parsing algorithms as for "web requests", and that in fact they must be dealt with at
another protocol layer.
It does not disclose that there can only be a single
"request" per session, and a single "reply" from either the original source or the alternate.
Specifically there is no disclosure on how to splice
responses, which it obviously does not do, or on the
lifespan of a session that makes the short-version
possible.
Perhaps most importantly, the invention described here
is working as an application level gateway. It is incapable
of quickly identifying TCP connections that do not require
proxying and leaving those connection truly unaltered. Terminating a TCP connection, examing the
first request in it, and then deciding to actually
forward the request to the real server is not "transparent".
The "preferred embodiement" either a) deferred establishiing the connection until the "true source"
was to be known (clearly unacceptable, what if the
"true source" is not accepting connections?, or
b) established the connection, and then aborted
it, once the decision to substitute was made.
The implications are not discussed or disclosed.
Which isn't surprising, because this patent describes
techniques that only work for HTTP 1.0
Caching for HTTP 1.1 is a new problem. You have to
deal with caching hints, persistent connections,
cookies that might affect the material supplied, etc.
As I understand it, RFIDs identify specific units.
Barcodes identified a product.
Barcodes are easily visible, and will be removed or not
left in a visible position when a consumer wears a purhcased item. RFIDs are invisible, and could easily
be read without the consumer being aware of it.
Even without access to a central database
on who bought what shirt, RFIDs could be used to
automatically correlate data: "Customer entering in
pink blouse has been in store 7 times over the past
three weeks,"
I would not go so far as to accuse its backers of
attempting to sneak a "citizen tracking" technology
past us. The intended uses of the technology are
valid and make sense. However I believe they are
showing a callous disregard to the potential unintended
consequences on their customer's privacy.
Of course if you actualy bought anything
they could then correlate it with your credit card.
The goals of tracking inventory can easily
be met with an RFID device that automatically deactivates
at a certain date and can be told to deactivate.
Slashdot Mirror
Supposed a witness had pointed to someone in a line-up and said "it isn't him, but he's real close, almost like his brother" and the police went and investigated the brother. Nobody would have any objection.
There is a danger with retained DNA databases, but I don't see it in this case.
As the number of "usual suspects" grows, eventually we will have someone who has a "1 in a million" match against the perpretators DNA.
I doubt that the prosecution will reveal to the jury that the defendant was one of 2 two million "undesriables" that the police keep the DNA on.
A DNA match against a sample recovered at a crime scene is never perfect. When it matches to a high degree of certainty someone who was already a suspect anyway then it is a marvelous confirmation tool. (Indeed, I believe DNA evidence has cleared more suspects than it has convicted.)
But when DNA evidence is used to search vasts databases, the statistical justifications that this is "evidence" rapidly dwindle, or possibly even evaporate.
And sadly, we cannot count on juries, and perhaps not even judges, to understand the impact of how a sample was selected in determining whether a test is statistically signifigant.
If there are 10 million people in a city, then a test that has "less than 1 in a million chance" of a false positive can be relied upon to falsely finger up to 10 people.
That's reasonable grounds for a search. But I sure hope judges and juries don't buy it as 'proof'.
For better or worse, the concept of selling a closed platform is legal. This is especially true if the buyer has adequate information to know that it is a closed platform. I would prefer a mandatory big red WARNING label to be affixed to all closed platforms saying "The retail price of this unit reflects a subsidy from the manufacturer. This subsidy is provided in anticipation of future revenues. Therefore this unit will only work with software lisenced by the manufacturer."
There are benefits to a closed platform to consumers.
The down side is simple. The consumer is being mislead by an artificially low up-front price into being locked into continued payments of a monopoly tax on each piece of software they purchase.
I believe the only solution is for the FTC to require platform vendors to offer their product in an unbundled format. You can buy an XBOX that will run third party software, but it might cost you $150 more.
If I read the patent correctly, the only claim of innovation is that a non-novel process (translation) is configured based upno session variables.
I believe that Environment variables are clearly prior art. As are web sites that deliver language-specific pages based on user supplied information.
An algorithm that was specifically adapted to an interactive dialogue could be patentable. For example, if you had specific techniques to determine whether the word "present" was referring to a gift or to a point in time based upon prior exchanges in the conversation.
Of course I'd hate to see what a group of punsters would do to the poor software.
The discussion has focused solely upon patent implicaions, ignoring the real impact of this technology.
The ability to translate an IM converstation into a foreign language, even one that might be understood by parents such as English, will have a profoundly chilling effect on IM converstaions.
Yes. Any web site that remembers your preferred edition, language whatever. Indeed the whole point of a session is to remember status rather than redundant including the information in each message.
Shouldn't this have been posted from the isnt-that-horse-already-out-of-the-barn department?
Depending on how the material was actually framed, you might or might not get away with claiming that Amazon was acting solely as a service provider.
But even promoting material, and displaying it in an integrated fashion is far from "publishing". If it were every bookstore owner would be a "publisher".
Would anyone expect a book store owner to validate that the purported author of each book had not engaged in plagarism? Why should we expect Amazon to do something that we would not expect a bricks and mortar book store to do?
Even a publisher is entitled to some presumption that an author has the right to sell what they are offering.
Conspicuously absent from the charges is any basis to believe that Amazon acted irresponsibly, or was negligent in allowing these sites to exist. What were they supposed to do, conduct an image search for all visual material offered for sale?
One thing that I have always been puzzled about on web privacy is the belief that explicit legislation or new rules are required to end these abuses.
If you read RFC 1945 section 12.3, it clearly states:
Now if you advertise your URL as "http://whatever", are you not implicitly stating that you have a server at that address that will comply with the http protocol?
So it seems to me that anyone who publishes a URL has already stated that they will not abuse the personal information without obtaining consent, because to do otherwise would be non-RFC compliant.
After that, general fraud statutes should apply.
I bought and received great satisfaction from both Conflict Catcher and SoundJam Pro.
So it is sad to see Cassady & Greene's departure.
But I'm not going to complain about the fact that a) Conflict Catcher is no longer needed and b) Apple bought SoundJam for everyone.
Ok, if I as an American buy music from the iTunes store, I'll be able to use that music on up to three computers. Right?
I cannot recall any restriction saying one of those couldn't be my laptop with me on a European trip, or even a computer that I have at a European office.
So anyone who wants to obtain this material without going through a "European" iTunes store can simply set up an account with an American ISP and buy it there, then transfer the file themselves.
Or decide that that is too much trouble and simply steal it the old fashioned way.
Somebodey hasn't caught on that the purpose of the iTunes store is to allow customers who want to be honest to pay for the music.
Blocking the store won't block the music.
There are major public interest factors that have to be considered.
But I agree with your fundamental analysis. The judge saw an injustice, and wanted to find a solution somehow.
It isn't easy. But it is a common requirement.
All good models, standards and specifications take great care to avoid specifying how something is to be done, rather than on the externally visible behavior that others rely upon.
The key challenge is to always ask yourself "Could this be done differently without affecting others?". And even more aggressively "How can I redefine this interface so that I no longer care if the other side uses algorithm B, instead of A?"
A GUI Design involves specifying a set of pages, the information on each page, its arrangement, the buttons that can be clicked, and the resulting transitions.
You can reasonably speak of a GUI "error" if there are obvious flaws in the navigation between the different screen/pages. For example, if you cannot back out of the current step without cancelling the entire transaction. Or if the information on a screen does not fit within the window and the window lacks scroll bars. Or if a "more info" button leads to a screen that is either irrelevant, or which cannot return back to the original screen.
If these are the types of errors that actually occurred, and they were replicated, I would have to agree that it is very convincing that theft occurred. I just don't agree that a EULA has the right to forbid competitive research. Competitive research is good, it benefits consumers and should be encouraged. Not doing your own research and evaluation is sloppy.
Making the same mistakes is a strong indicator that no independent research or judgemetn was exercised. It might even be enough proof to satisfy me as a juror if there was no contrary evidence.
My initial read is somewhat similiar. A judge has come up with faulty reasoning to support what seems right. But they should come up with the correct rationale. The rationale offered would be chilling.
If customers have the right to examine products, and determine what they like and dislike about each, then it isn't much of a leap to say that producers have the same right to examine their competitor's products.
But there's a line somewhere between studying what some product does, and essentially stealing its research. Whatever the protection mechanism should be, it should stop lazy companies from simply stealing interface designs from other companies rather than paying to develop them on their own.
So it is pretty much copyright infringement, except that some allowance has to be made for the ability of the market to clone interfaces from dominant providers.
Copyright also provides an excellent insight into what contract law must not be allowed to create here. No author is allowed to sell their mystery to the general public except that no other mystery writer may read it for the purpposes of evaluating what was effective or ineffective.
Indeed many writers want to aware of what others have written, so they can ensure that they don't inadvertently write something too similiar to an existing book.
The bottom line is that the term "reverse engineering" should never be applied to observing the external behavior of a product. To me that term implies trying to figure out how the product works, not to trying to figure what the product does.
Slavishly copying what a product does, before the product has an established user base, also strikes me as improper copying. I'll admit I do not know how to define that line. It may be similiar to judgement calls made on when fictional characters have achieved "cultural icon" status.
In the 15th century both Europe and China developed deep sea fleets. The Chinese fleet was government run for the glory of the Emperor. The European fleet had scattered government financing, but was basically run by greedy mercantilists. The Emperor lost interest. The Chinese fleet was burned. The greedy Euroean merchants continued to expand their fleets leading to Europe's domination over the entire world.
Do we really need another "space race"?
Or is time to allow space development to proceed without the sponsorship of modern emperors?
If I remember correctly, I first read this argument in an essay by Poul Anderson. Unfortunately I couldn't find a reference to it.
If you get past posturing being above politics, you'll quickly realize that both parties do indeed have very serious flaws and limitations.
But they are very different flaws.
If you think Democrats are prone to immunize large corporations from truthful labeling in the marketplace then you haven't been paying attention.
There used to be a wing of the Republican party that really believed in the strength of the market system. They've been gone since Reagan. So it's true that neither party has sufficient faith in a true free market, but the ways that they interfere with the market are very different.
Don't defame the laissez-faire approach.
I don't endorse it, but to be fair most laissez-faire economists still believe that market participants have to label goods and services accurately.
Even a laissez-faire capitalist recognizes that selling a "CD" that will not play in a standard CD player for what it is - fraud.
Republicans on the other hand can probably come up with some idea why this is a good thing.
And a physical device is just exploiting the laws of physics that everyone should be free to use.
The goal is to ensure those that find things that were unlikely to have been found anyway are rewarded. Thereby encouraging both their development, and the distribution of that knowledge.
The current system is broken. It needs fixing. Things that are obvious are patented, statements of the problem are patented.
But if you eliminate all protection for complex algorithm development then you'll discourage their development. A small company will be unable to deploy a new technique and recoup its investment if large companies can simply reverse engineer the solution and then deploy it on a large scale.
Without software patents, you will quickly see:
The guidance the technical feed needs to be providing lawmakers is on how to properly handle technical intellectual property -- not eliminating it.
Eliminating 'obvious' patents would also be desirable, but it requires more domain knowledge to make that call. Differentiating between a problem and a solution should be a lot easier.
I remain convinced that the only defense against 'obvious' patents is to allow a public comment period. It would allow challenges to the basis of a patent without having to expose yourself or hire a lawyer.
Shorterning the period for all "high tech" inventions is probably a good idea. But I don't see the point in penalizing an invention just because it was realized in software as opposed to being realized in hardware. In many cases, the same essential algorithm can be done either way.
The issue is not whether a patent is for "software" or not. The same algorithm can be developed either in software or in hardware.
A specific algorithm that represents a truly novel solution to a problem should be patentable, no matter how it is implemented.
What needs to be stopped is the American practice of patenting the feature, rather than a specific solution.
The original application date is October 14, 1996. It's a foreign application, but I believe that it is still the relevant date. I also believe that it predates transparent proxying in squid.
It does not predate transparent proxies, however.
Akamai's paying customers are interested in reducing actual bandwidth. There's nothing virtual about it.
I believe the patent predates Squid, so there could be a problem to whatever degree that Squid infringes. Just because a later developer is open source does not mean that the original claim was invalid.
However, reading the patent carefully, you realize that it actually only describes a very specific solution. Specific enough that it truly is describing a solution, not a problem. And specific enough that it might legitimately be considered novel for the time it was filed (I really don't have time to search the source code of all proxy servers in the 1996 time frame -- let someone with a finanicial stake do that).
Specifically the patent deals with websites that are identified by their IP Address and where certain content (by default all) is held in an alternate (and presumably closer) server.
There is nothing in this patent about determining if the content is fresh. The description presumes that the cached copies were pushed by the server.
So this would only seem to proxy servers that are transparent to the user, but not to the servers. The proxy servers that are of most interest to an ISP would either be transparent to the server as well, or more of an akamai style strategy where the first-response page is localized to directly fetch pre-positioned material from edge caches.
Interestingly, the patent seems to be worded to cover a single box which handles both the intercept and the decision to proxy, but does not handle the actual proxy response. A firewall transparently redirecting a port to a proxy server is prior art. The basic claim to being novel here is that the client does not have to be configured to use the proxy, and diversions only take place if certain content is requested, non-proxied sites are passed through "unaffected" (which is a false claim, BTW, which I'll deal with in a moment).
There are some serious omissions in the description, would could undermine its enforceability.
Perhaps most importantly, the invention described here is working as an application level gateway. It is incapable of quickly identifying TCP connections that do not require proxying and leaving those connection truly unaltered. Terminating a TCP connection, examing the first request in it, and then deciding to actually forward the request to the real server is not "transparent".
The "preferred embodiement" either a) deferred establishiing the connection until the "true source" was to be known (clearly unacceptable, what if the "true source" is not accepting connections?, or b) established the connection, and then aborted it, once the decision to substitute was made.
The implications are not discussed or disclosed. Which isn't surprising, because this patent describes techniques that only work for HTTP 1.0
Caching for HTTP 1.1 is a new problem. You have to deal with caching hints, persistent connections, cookies that might affect the material supplied, etc.
Additionally, merely seeing someone download a file doesn't really prove that they knew they were violating a copyright.
RIAA could never prove the contrary. I'm not even sure they'd have reasonable grounds to search your hard drive.
But if you sending something, then you knew or should have known, what it was.
As I understand it, RFIDs identify specific units. Barcodes identified a product.
Barcodes are easily visible, and will be removed or not left in a visible position when a consumer wears a purhcased item. RFIDs are invisible, and could easily be read without the consumer being aware of it.
Even without access to a central database on who bought what shirt, RFIDs could be used to automatically correlate data: "Customer entering in pink blouse has been in store 7 times over the past three weeks,"
I would not go so far as to accuse its backers of attempting to sneak a "citizen tracking" technology past us. The intended uses of the technology are valid and make sense. However I believe they are showing a callous disregard to the potential unintended consequences on their customer's privacy.
Of course if you actualy bought anything they could then correlate it with your credit card.
The goals of tracking inventory can easily be met with an RFID device that automatically deactivates at a certain date and can be told to deactivate.