Sunday, February 24, 2008

Peer to Patent - Good Idea or Bad? 

The ACM Communications in February carries an article by Andy Oram extolling the virtues of the new peer to patent experiment that looks to ameliorate the problems in the patent process.

Snag is I'm not sure it is helping at all for those IT and software related patents.

The basic problem is two fold.  First there really is nothing new being invented in computing - instead all the early inventions in the field dating back to the 40's through to 
the 60's are merely being applied in interesting and useful new ways that are simply applications or refinements of the technology - most all of which are intuitively obvious to any practitioner in the field.  The only exception being new hardware fabrications pushing the boundaries of practical physics - but that is really the field of applied engineering and not computer software and IT.

The second part of the problem flows from the first in that the real inhibitor to implementers is not having the "invention". Quite the contrary - ideas are aplenty and cheap - whereas the resources, time, backing and support  to bring them to successful fruition are the inhibitors.  

Moreover to compound this computing has entered every facet of modern life 
so the scope for applications of computing is pretty much endless.

Enter a rag-tag army of patent-police dedicated to stopping frivolous patents by reviewing them via Peer-To-Patent's (PTP) website (http://peertopatent.org).  Why are we wasting our time?  I'd wager $10 that 99.9% of patents are not inventions at all - but simply applications of existing infrastructure and capabilities.
Certainly from the selection I saw under review at PTP's website today - they all fit that category - and 
clearly PTP thought so too - so they'd selected them for attention.

I saw today this gem - "Method for generating mnemonic random passcodes" - I wrote something along these lines for my undergraduate work on a PDP11 30 years ago.  These types of patents are submitted by eager patent attorneys who egg on their clients - better to have it patented - should be possible to have the PTO award it...

The bottom line is that the patent office has to stop recognizing these things as inventions and worthy of patents.  They are quite clearly not.

Another example using GPS devices to track pets.  This is not an invention.  It's taking the same small format GPS device that athletes have been using for more than two years to track themselves and putting a doggie collar strap on it.

We have to go back to the original purpose behind patents. 

It was to protect the inventor who toiled in his laboratory 
to create some new breakthrough that had not been seen before
 that surmounted previous limitations and challenges and brought 
truly new capabilities and solutions to the field.

All those long hours of painstaking research, blind alleys and simply trail and error experimenting till you came up with the answers you needed.

Computing doesn't work that way.  We work on software and accompanying hardware and create a solution from the
 collection of preexisting "Lego bricks" or fabricate some new 
bricks of your own to add to the designers collections.

This may take you a few days, a few weeks or a few months - but not years - quite simply your funding would dry up long before that - in fact the bean-counters these days want tangible results in weeks only - otherwise they are generally not interested.

And worse - the current patent system has been completely subverted by large corporations and patent lawyers each responding to the other - so even if the inventor now tries to patent something - its lost in a tidal 
wave of filings from giant corporations - who can use their own phalanxes of patent portfolios to either attack of defend in the marketplace as they choose.

Then the patent office itself - that is self-funding - thanks to Ronald Reagan.  So they themselves have a huge vested interested in a torrent of endless trivial patents and awarding them so that corporations feel threatened and file yet more patents.  The last thing the executives at the PTO want is for people to stop filing trivial patents.

So whom or what is peer-to-patent helping?  Intriguingly it is sponsored by GE, HP, IBM, Microsoft and Red Hat!

Having looked at PTP's site and purpose I'm not at all enamoured of investing my valuable time in this.  Frankly the only way to solve the current patent chaos is to have the PTO continue to issue more and more increasingly stupid and worthless patents so that people can finally say "enough" and the whole rotten software patent edifice collapse under its own heaving morass.

Besides the marketplace is largely consigning all this nonsense to history
through new open source and open resource collaborative solutions
and by using new licensing techniques.  The competitive landscape has
for ever been changed by the internet and services and solutions that
are replacing traditional 1980's delivery methods that foster and attempt
to continue the "let's patent it" mindset.  The more products that are
rejected by the marketplace because they are patented and hence place
restrictions on how they can be adopted and purposed the sooner
people stop this disruptive and arcane practice that serves a tiny few
and restricts the many.

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