|This technology is described as pinpointing the physical location of Web surfers right down to the city level, based on IP addresses |
Does anyone have a link to a more detailed description of this patent?
What if I have a database of IP ranges and their corresponding countries - does that infringe on their patent? If so, that would be going to far.
Digital Envoy have been a Google Technology partner for several years - this is more about the licence fees and the still running dispute.
A Patent would probably strengthen Digital Envoy's position should Google try to go it alone and develop an 'inhouse' replacement....
|Google currently pays US$8,000 a month for its use of the Digital Envoy technology and has offered to increase that amount by 50 percent, Kratz said. But he suspects that Google is making millions from syndicating its ads to third parties and that the court discovery process will determine just how much. That way, it can come to appropriate fees |
Here's the patent [patft.uspto.gov].
Basically, Digital Envoy's technique comprises 3 steps (my summary):
1. use nslookup to determine the hostname and hope it contains the name of the city
2. if that fails, try to figure out the city name by tracing the request -> look at the location of switches and routers (tracert)
3. store the information in a database.
These guys should be nominated for the Nobel prize.
I guess RonPK was being sarcastic. :)
The technique as summarised is nothing more than the automation of what many admins already do manually when they are curious about a referer. Credit card gateways also do this to try and match the user to a geographic area. Clearcommerce uses this technique, amongst others, to combat fraud.
The database, over time might have value, but not the technique.
Another classic for the great US Patents system. For the average internet user, this will be as remote as dinosaurs fscking. But it would not be unusual to see prior art being used in an attempt to invalidate the patent.
How's it going to target my physical location? My ISP is clear across the country - literally.... I'm in Utah, my ISP is in Rhode Island.
Hmmm. Though it could maybe get something from the local dialup number I guess. Though it doesn't include an area code....
Using a traceroute they should be able to tell where your first stop to your data is, regardless of where your ISP is centrally located. Correct me if I'm wrong, but it isn't too hard to find out where someone is connecting to the internet from.
LOL Lawyers and corporates on the make.
Digital Envoy claims they are the inventor of geo-location technology. Rubbish.
They claim a priority dating back to 1999. Rubbish.
They claim the patent was first filed before any other current geo-location company was in existence. This suggests they know full well a defunct geo-location company was in existence prior to their filing.
As I used geo-location well before 1999 I must file suits against both asap!
hdpt00 you are right in most cases but only if a web site uses complex checks and the user doesn't know how to mask where they are.
To quote Google Answers back in 2002:
Geolocation is a very inexact science, and the accuracy of the results are often fairly low, even
with the most advanced methods."
The "first stop" for me is dialup.level3.denver.net....
Mine is dialup.anywhere.wherever and I am not that IT aware!
This is silly. Its like allowing a patent on the one-click purchase.
|Mine is dialup.anywhere.wherever and I am not that IT aware! |
"nslookup to determine the hostname"
That's about what I thought the first time I heard about the lawsuit. I know for adwords... Google thinks I am where my T1 provider is. At work from one T1 they think I am in Texas and show Texas regional ads, Another T1 they they think another state. At home they always show me Virginia ads... though I am in Florida. If that's the basic extent of their patent, every sysop in the world will violate their patent in the course of their daily work. Somebody needs to tell the patent office folks how to go to ARIN and do a lookup. That seems to be the norm for the patent folk... they seem to be tech deficient.
Meaning geo-targeting is flawed. We overide it every day as we need to see what our clients target market will see. Take a Dutch client: To see how they rank in Holland we need to use a Dutch proxy.
So this free service GeoDirection [geobytes.com] (no association what-so-ever) would now be in violation of that patent?
But any proxy/cache (like AOL) will break the system of course. I wonder if AOL provides more specific location information to Google for their adwords?
you can tell much more accurate than you think but the ping rate and the additional traffic not to mention the firewall issues ...
any day now they're gonna patent hacking ...what they do is basic ..just scaled up .
|As I used geo-location well before 1999 |
I been using my 'own' geo stuff/page, probably started around the middle of 1997.
If determining where the visitor came from based on their IP is such a difficult science that it warrants a patent, someone should patent cloaking for Google ;)
Seriously, if this method is as simple as RonPK summarized, this really is silly.
As far as I understand this is a US petnt only - so just come to Europe and set up your cloaking :)
Right now I have been allocated an IP address that does not resolve to Australia (and my local ISP has lost interest in it because they haven't done "anything"). I see ads aimed at California.
The address is 220.253.nn.nnn - not a range we usually see in APAC.
It's the second occurrence in a month, but when trying to manage PPC ads aimed at specific markets (not all on Google), it makes a mockery of geo-targetting.
If I understand their conflict with Google correctly, then the patent is completely irrelevant to it. The patent covers how to collect a database. The conflict is about the use of that database.
And of course that patent is just detail to point out the complete and utter incompetence of the USPO...
Can't see, why there is a database anyway since there is this (simple) method?
And there are too often new IPs every time a client connects. Performance?