Kodak Wins $1 Billion Java Lawsuit

nberardi writes "The Rochester Democrat & Chronicle is reporting that Eastman Kodak Company has just won a patent suit against Sun on the Java Language. According to the article Kodak owns a patent which describes a way for a piece of software to "ask for help" from another application. What they are claiming is that Sun violates this patent when Java byte code uses the Java engine to run the code. This may really upset the industry, because not only Sun uses this technology for Java but Microsoft uses this technology in .Net."

Comments

This is interesting

[vivekg]

This is interesting... let's see who is next?

Not just Java and *net

[nurb432]

Python, Pascal, most any VM based language could fall prey to this...

Sort of sucks up the 900+ mill Sun just got from Microsoft.. DOH

According to a groklaw post

[mark9white]

Authored by: Anonymous on Sunday, October 03 2004 @ 09:54 AM EDT Disclaimer: IANAL, I only reviewed this one patent, the following is merely a brief summary performed by one "skilled in the art", if you are concerned about whether any particular product (listed below or not) might infringe, you'll have to go through a court case with a lawyer(or a bunch), since there's no other option in the current system. Not all OO systems would infringe, the following is a (short) list of some systems that is likely to be instructive. These are listed in no particular order. It is possible that CICS (from the mainframe realm, and still in heavy use) would infringe. CORBA almost certainly infringes. Smalltalk (the environment, not the language) probably infringes (or represents prior art, it's hard to get the distinction straight when the patent's this bad). .Net definitely infringes. MS COM+ definitely infringes, except it got morphed into .Net. MS MTS (from NT 4.0) would definitely infringe, if it was still around, and not subsumed into COM+, and then .Net. Some parts of the Java Platform appear to infringe (RMI, combined with parts of Reflection, and a few other bits). Some systems that rely on IMS (another mainframe platform) might infringe. The average C++ program would not infringe. Bonobo (from GNOME) would definitely infringe. In order to infringe, a product would need to provide resource management, be aware of polymorphism (esp. linked polymorphism), and provide a form of rendezvous services as well (given a chunk of data with type information, and an operation to perform, find the "program" to perform the operation for that data type) I suspect that ample prior art could be culled from the PARC research that dealt with the Smalltalk environment. If that isn't enough, look into the resource managers for IMS and CICS, the old OLE technology from MS, the combined program+resource streams from the early Mac systems, and any similar products. Most of those would overlay this patent so a greater or lesser extent, and some would make the entire construct "obvious to one skilled in the art". The whole "innovation" seems to be nothing more than a resource manager that's aware of polymorphic data. Most good resource managers were aware of linked polymorphism (that's the reference-to-second-type-within-first-type part of the claims) at least 5 years before this patent was filed. I suspect this was granted because it had the word "Object" in it, and for no other reason. Replace "Object" with "Data Structure", and it's completely obvious (and there's nothing in the patent that functionally differentiates between data structures and objects)