Slashdot Mirror
Akamai Wins Lawsuit to Protect Obvious Patent
brandaman writes "Akamai, the largest content delivery network (CDN) with about 70% market share, recently won its lawsuit against the against second largest CDN - Limelight Networks. The suit asserted that Limelight was infringing on Akamai's patent which, upon examination, seems to be somewhat on the obvious side. 'In accordance with the invention, however, a base HTML document portion of a Web page is served from the Content Provider's site while one or more embedded objects for the page are served from the hosting servers, preferably, those hosting servers near the client machine. By serving the base HTML document from the Content Provider's site, the Content Provider maintains control over the content.' Limelight is obviously not pleased, and this is not the first lawsuit Akamai has won regarding its patents."
Is this patenting having the html on one server and the rest (pictures etc) on other ones?
Apparently only in the specific case of having the "other ones" be distributed across the network and with the "closest" server to the client chosen to download the content from.
I suppose that things like mirrors, etc. don't count because in that case the user typically chooses what they believe to be the closest server rather than the host or akamai.
Sorry, but you're not even close.
The way Akamai works is it distributes the "heavy duty" content like images, scripts. to its own servers all around the world. It then lets its customers (like E*Trade, to pick one actual example) modify their static HTML content to refer to those images in a special way. For example, the E*Trade home page has the following link in it for one of its images:
https://a248.e.akamai.net/n/248/1777/20080228.0/www.etrade.com/images/prospect/topGrad.gif
The url is specially encoded in such a way that when your local DNS server queries a248.e.akamai.net, the DNS server returns a server located physically near you. So if you're in England a248.e.akamai.net might resolve to an IP located in Londan, but in New York City it would resolve to an IP somewhere in New York. Then when the http request is sent, Akamais servers decode that annoyingly long URL to determine which customer of theirs it is and serve up the correct image. It's actually a fairly complex and fast process. If the server that you're directed to doesn't actually have the image locally then that Akamai server will query another nearby Akamai server. If that server also doesn't have it then it'll actually pull the image down from a master server that E*Trade uploaded the image to.
You can test this out yourself by looking up the IP address of a248.e.akamai.net yourself. Locally you'll get one IP. If you do a google search for dns lookup tools you can submit that domain name to other sites to look it up and you'll get totally different IP's that are physically close to wherever that domain lookup tool runs from.
The bottom line is that it's a prety complex process that involves both the use of DNS to ensure you download large chunks of content from physically near servers as well as some pretty sophisticated caching in the background to make sure static content is delivered rapidly no matter where in the world you are.
I used to work at Akamai so I have a pretty good firsthand knowledge of how their stuff works. I doubt a lot of their algorithms they use would pass the "obviousness" test...
As in many of these "obvious patent" trolling articles, the article/summary oversimplify the patent. The patent does not just claim click here, fetch there redirection which is used by just about every major site, but algorithms for doing the load balancing etc.
If you read some of the claims, then you'll see that various algorithms are used for load balancing and other purposes. While these might be obvious to some, they are extremely obvious to all.
The test of "obvious" is also not that clear cut. IIRC, the tests is "reasonably obvious to practitioners of the art". This test should be applied to the state of the art as at the time of the patent, because a patent "teaches" the industry and therefore after the disclosure the less-than-obvious become obvious.
Engineering is the art of compromise.
In at least two fundamental ways. First, the summary quoted the abstract of the patent, not the claims. The abstract is almost always a simplified extract of the contents of the patent and rarely has any meat to it. Of course it looks obvious.
READ THE CLAIMS TO FIND OUT WHAT IS BEING COVERED BY THE PATENT!!
Here is claim 1:
1. A distributed hosting framework operative in a computer network in which users of client machines connect to a content provider server, the framework comprising:
a routine for modifying at least one embedded object URL of a web page to include a hostname pretended to a domain name and path;
a set of content servers, distinct from the content provider server, for hosting at least some of the embedded objects of web pages that are normally hosted by the content provider server;
at least one first level name server that provides a first level domain name service (DNS) resolution; and
at least one second level name server that provides a second level domain name service (DNS) resolution;
wherein in response to requests for the web page, generated by the client machines the web page including the modified embedded object URL is served from the content provider server and the embedded object identified by the modified embedded object URL is served from a given one of the content servers as identified by the first level and second level name servers.
Doesn't seem so obvious now, does it?
The second is the fact is that Akamai is a very innovative company that has pioneered a lot of distributed content delivery starting with the early days of the internet. In my mind it is very obvious that they would have a lot of valid patent material. They are most assuredly NOT patent trolls, and in fact have brought many innovations based on some very advanced work to commercial fruition. It is insane that their work is being shown in this light by Slashdot.
The company was founded by an MIT graduate student (Dan Lewin) and an applied math professor from MIT, Tom Leighton who is currently head of the algorithms group at at MIT's Computer Science and Artificial Intelligence Laboratory. Lewin was tragically killed when AA flight 11 was crashed during the 9-11 terrorist attack.
This article is one of the most ridiculous ever posted by Slashdot.
Microsoft's use of Akamai in 2003 gained attention when it made it appear that Microsoft's web site was running on Linux. In actuality it was just the Akamai caching servers using Linux. Like Google, they've since shifted to using more of their own network as well as Limelight and Savvis (now Level 3).
RichM
Data Center Knowledge
according to patent law, algorithms are not patentable
They are now under the "business process" umbrella. The courts are accepting these so far.
Table-ized A.I.
Genuity, a web hosting company, was doing this via their "hopscotch" routing protocol in 1997. They were bought by GTE at that point but the technology had already been in development for several years. I met the founders at a conference and we exchanged some ideas on improvements based on some work I was doing for another company. Basically, though, they had connections into all of the major NAPs in the US and a dynamic cost-based routing protocol that chose which server to use for which customer. Dynamic updates to the site data (e.g. actually buying stuff) was more complex, obviously, because they had to wait for the transaction to synchronize, but at least they benefited by processing the request through the fastest pipe to the browser. Those updates and associated content came from a different server, matching the patent requirements.
I found this article ( http://findarticles.com/p/articles/mi_m0EIN/is_1997_Dec_10/ai_20053332 ) rather easily, going back to 1997.
Think about it
I've spent plenty of time, as I said in my post I've actually read all of the Ssupreme Court rulings on the subject.
A number is not an "invention". An equation is not an "invention". A calculation is not an "invention". Mental information processing is not an invention. Mental information processing does not magically become a patentable invention when you OBVIOUSLY use a calculator to accelerate/automate it. Mental information processing does not magically become a patentable invention when you OBVIOUSLY use a computer to accelerate/automate it. Mental information processing does not magically become a patentable invention when you OBVIOUSLY do it on the internet to accelerate/automate it.
The Supreme Court has explicitly ruled that no possible algorithm can ever qualify as "novel" or "non-obvious" for patent purposes. Therefore no possible software can ever qualify as novel, no possible software can ever qualify as non-obvious, no possible software can ever be an invention. There is nothing novel or non-obvious in blatantly using an ordinary computer to carry out that "non-novel" "obvious" calculation. Sticking the words "on a computer" at the end of a mathematical information manipulation does not magically turn it into a patentable invention.
It doesn't matter if you are the first person to write down some particular number, it cannot be "novel". It doesn't matter how many digits long your number is, it cannot be "non-obvious". No possible math, no possible information processing, no possible mental process, no possible algorithm, no possible software, can ever be an invention. As the Supreme Court said, they can never qualify as novel or non-obvious for the same reason that laws of physics are never treated as novel or non-obvious for patent purposes. An invention may make use of gravity, but G=M1*M2/R^2 is not an invention and it is treated as non-novel and as obvious, even if you are the first person to figure it out.
Chemistry and software are not really that different philosophically
No matter how long I *think* about a chemical reaction I will never actually make any molecules.
Physical objects and physical processes are philosophically different than math/calculations/mental-processes.
Physical objects and physical processes are concretely different than math/calculations/mental-processes.
-
- - You can't take something off the Internet! That's like trying to take pee out of a swimming pool.