Sunday, December 17, 2006

What are you wishing for in 2007? (22 software patents and counting...) 

There's obviously several things high on people's list - mine would include -

This last item is what I want to talk about here. My work includes much that is in the front line of software patents and public open standards and open source. The OASIS standards group is currently implementing new IPR modes for all its technical committees - one of which I chair.

The bottom line is that the apocolypse is when it has become impossible for software developers to write code for ANY reason - that someone else cannot claim is infringing on a patent they already have.

Folks - I now have concrete evidence we are there already and beyond - and what is more the "Big 5" - Microsoft, Oracle, IBM, Sun and BEA know we are already there - and have been acting accordingly for the past 3 or more years. One such manifestation of this admission is the new OASIS IPR policies and especially the non-assertion covenants adopted by Sun and IBM (also while these seem very philanthropic there is an under current of implied threat to transgressors).

I have two US software patents - that took me much hard effort to get awarded - because I believe they represent true inventions - and the USPTO had a hard time even understanding the ideas in them - because they are used to an entirely different patent style - written by patent attourneys - not programmers.

This is all just like playing chess - assuming you could get chess playing patents. Imagine that an invention is a chess opening - and sound new openings - well those are really tough to do - even Kasparov with all his genius has only improved on existing opening theory - not created entirely new branches. But what if you could get a software patent merely by moving just one pawn one square over and claiming you just invented something entirely new?

That is exactly the situation today. But its even worse - imagine - since obviously "pawn shifting" by itself may be fairly limited and the PTO could reject on the basis of someone else having already logged that move - but wait it gets better - what if you could reference someone elses' patent and simply switch the "move order" of the sequence of play and claim an entirely new patent?! Guess what? You can!! The PTO also employs people to review your applications who are not actually real "chess players" - so the protection afforded by the constraint - "that the claimed invention cannot be anything that is self-evident to 'average practitioners in the field'" is minimal at best.

Let's look at my specific two patents - USPTO 5,909,570 and 6,418,400 - and see this in practice. There are now 22 new patents that reference my originals. Now this is quiet flattering in one sense - until you start to examine what is really going on. I'll come back to those actual 22 in a moment. Let's look at why there are 22 references first.

The PTO finds my patents regularly during their 16 hour review process of new patents - because they search for similar and existing work - and mine happens to be extensive in scope and applicability in their particular area. So you would think this would mean the reverse - that other people would have a tough time getting awarded patents that really do not offer anything new. Presto - all you have to do is REFER to my patent - and then suggest something extra - and now you have invented it! This is like there being a patent on doors - and you being able to get new patents by just attaching different handles in different positions on the door and claiming a new invention!

This does keep patent attorneys in work - but it is having the REVERSE EFFECT of what patents are supposed to do in the first place - foster an environment where business is encouraged to innovate and develop new products and services. Unfortunately writing code is now OUT - first you have to write a patent to defend your code - and then write your code. And worse - what about developing open public standards and open source that does not infringe somewhere? Frankly this has now become impossible - hence Sun and IBM's resorting to the threat of non-assertion covenants.

Let's go back and look at some of those 22 patents referencing mine - to see just how knowledgable the USPTO is about "playing chess" or developing new software?

Would I have awarded any of these 22 additional patents referencing 5,909,570 and 6,418,400?

7,120,663 Method and apparatus for updating XML data - This is incredulous "pawn shifting" that is entirely common practice in everyones XML code - well done NEC for getting this award!
7,114,123 User controllable data grouping in structural document translation - this is even more incredulous - IBM I tip my hat - practically every XML editor uses this obvious technique.
6,915,312 Data processing environment with methods providing contemporaneous synchronization of two or more clients - Starfish Software - you guys ever owned a Palm PDA? Yeah - I thought that is where this idea came from.
6,871,187 Translator for use in an automated order entry system - Dell - you guys are amazing - you've never heard of catalogues and order processing before? I'm stunned the USPTO have not either - because they gave you this patent.
6,757,739 Method and apparatus for automatically converting the format of an electronic message - Contivo Inc - when I first met these folks I realized their product infringed on my patent - and further more - that they did not understand how to do it better - because their method is weaker. Clearly the USPTO also did not get it either - because they allowed them a patent for something mine already does.
6,457,003 Methods, systems and computer program products for logical access of data sources utilizing standard relational database management systems - this is just IBM protecting itself from patent trolls - of course logical name resolution (aliases) is a standard industry wide technique - better get a patent on it before USPTO gives it out to someone who just wants to use it to blackmail large corporations.
6,601,071 Method and system for business to business data interchange using XML - Oracle take a bow - yours is the most outrageous and laughable of the bunch - how could the USPTO have given you a patent for processing data in and out of XML?!?

I've not covered off all 22 here - but safe to say most all of these exhibit the same sad litigany of the USPTO abject incompetence in the face of manipulation by slick abusers of the process.

What is truely disturbing is that this is just 22 patents here - while there are tens of thousands of software patents issued each year!

Can we all just admit that software patents have degraded to nothing more than just sets of "chess moves" that have no real bearing on what practitioners are doing in the field - quite the reverse - just threaten this vital sector of industry and make it impossible for small inventors, standards bodies, and those truely seeking to advance the field especially to do so without threat of poverty and ruin.

Congress and the EU between them should immediately disband the issuing and practice of software patents.

At least this seems a relatively easy decision to implement compared to the other items on my 2007 wish list - so maybe I'm not dreaming when I think this might just be attainable if enough people make enough protest?

References - ACM article - The patent holder's dilemma: Buy, Sell or Troll?

"The current patent process in many ways works against IT innovation by making the road to realization too dispiriting for today's independent inventors".

This page is powered by Blogger. Isn't yours?