If I were Amazon's Jeff Bezos what would I not do?

At the time I - like many others - was impressed by Bezos' vision, talent, chutzpah and joviality amid the adversities that come with challenging the established players in the market, such as Barnes and Noble and the other large chains of bookstores.

At that time, too, the mainstream newspapers, magazines, TV stations, and conference organisers kept visiting, adulating, inviting and distracting him from his daily job of being smart and innovative.

The signs of being spoiled rotten started to show when Bezos switched from being an innovator to a rich investor, buying up a number of small companies with promising projects, in much the same way as AOL's Steve Case has been doing. The signs continued with some much-criticised practices such as listing books prominently on his sites without any indication that such prime location was earned not by merit but by having the money to pay for the privilege.

Nevertheless, Amazon.com kept spreading. It spawned a site in the UK and Germany and the US site itself was growing (but not really improving), adding new stores to sell music, video, toys, garden tools and kitchen stuff.

When Time magazine chose Bezos as Person of the Year, the fawning on him intensified and so did the braying laugh that no reporter failed to notice and record.

Few, however, could have understood the reason for it - but now I understand what triggered those laughs when journalists presented him as a saviour, and the poster-boy of "Interpreneurship". He must have been tickled, thinking of the patents that he filed in 1997 and which were - for him - obviously on the road to being accepted, yielding the prospects of millions of dollars in licensing fees in the future.

Little did any of us know at that time that Amazon.com had filed - among others - a patent application for its system for placing a purchase order via a communications network, with the intention of killing at least one of its competitors, as has become clear by now. This, and a series of other legal - and not so legal - actions by Amazon.com resulted in a well-organised and fairly large-scale boycott by customers, along with denunciations by industry internet pioneers and start-up companies that feel betrayed and threatened by Bezos. From Person of the Year in 1999 he is on his way to become Internet Villain of 2000. And this is not an only-in-America issue, because the European Patent Organisation (EPO) may follow suit after the revision of the European patent Convention that currently prohibits the patenting of software.

Bezos's patent grants a monopoly to Amazon for the 1-click process which makes it possible for the customer who has once provided personal information to the electronic retailer, to skip doing so the next time by entering a short identifying code (the last four digits of his or her credit card number) when placing a new order. The cookie placed on his or her computer would then forward the information needed for the purchase transaction without the user doing anything else. The process is so simple that not even the legal experts could make it incomprehensible (see box below).

I hasten to add that I am not totally and blindly against lawyers - indeed I studied law for six years to become one of them. What I am against is the abuse of law and excessive litigation. Is this 1-Click process a patentable method or device?

Most of us think it is not, but the examiner in the US Patent and Trademark Office (PTO) thought it was, and issued the patent accepting the 26 claims presented in the application for patent protection.

If you find it hard to believe how one can lay such an excessively broad set of claims on the 'invention' of the 1-Click operation you can read the claims section in all its glory for free, courtesy of IBM, the largest holder of computer-related patents at www.patents.ib.com/det ails?&pn=US05960411_.

Once a patent is issued it remains valid until challenged in court and invalidated, which is a lengthy and costly process. Exceptionally, the commissioner of the PTO may initiate the revision, as was the case in Compton's ill-fated multimedia patent that I will return to later. US patent procedures keep the application secret, in contrast to the European practice that provides for the announcement of patent applications and invites third parties to challenge patent claims before the patent is issued.

I am not merely joining now the chorus that opposes the application and the PTO decision. More than twenty years ago I wrote my doctoral dissertation about the legal protection of computer software, discussing the deficiencies of both the patent and the copyright regulations for software. I also drafted a piece of legislation specifically for computer programs, although I should confess that the proposal was never adopted.

Given the operation of the PTO, the decision is not surprising. Nearly a decade ago Congress stopped financing the PTO, forcing it to be self-supporting from the application fees.

This is the same body that the Constitution empowered through the noble provision (invoked in every second article about patents in scholarly legal journals) "to promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries".

The PTO started to grant patents in an increasing number which then encouraged applicants to give it a try for a few thousand dollars, which in turn increased the budget of the PTO. By the late 1990s patents for software may have been granted as if it was - well - a cookie cutter job.

I don't want to bore you with legal details, or numb you with convoluted legal citations formats and jargon because this is not a legal paper. Suffice it to say there are a number of criteria that inventions must satisfy to qualify for patent protection, because it gives monopolistic rights for 17 years to the inventor.

One of these is that the method or the process must be novel and not obvious. At the time Amazon applied for the patent, the cookie method, which was introduced by Netscape in 1996, was obvious and was not novel. Even without much searching I was able to find articles that mention the use of cookies to store and forward personal information about a previous visitor (see right).

The piece of code needed for the process is something that a beginner computer science student could have written in her lunch break. It is somewhat preposterous that Bezos et al. claim to have invented the method of recording and forwarding personal information for identifying returning web users in 1997 when it was widely practised in 1996. It is not unlike the bold claims that Compton's New Media made in 1993 when applied and received a patent about using multimedia technology. I happened to be at the COMDEX press conference where company executives giddily broke the news. The reaction was akin to the one that a referee would get in a soccer game when awarding a penalty kick to the visiting team for a faked fault. Days later the computer industry had a reaction similar to what we see now in the Amazon.com case. A few months later Compton's New Media went out of business. A year later the patent was annulled by the PTO which had initiated a revision in response to the uproar.

Amazon.com could probably have got away with their action as its announcement - two weeks after the patent was granted - did not generate wild and wide reaction. It was the next step that woke up journalists, competitors and individual developers to the real implication of the existence of such a patent.

In late October, Amazon.com filed a patent infringement suit against its arch-rival, barnesandnoble.com which was using a similar method to 1-click with its Express Lane service. Amazon.com sought an "immediate and permanent court-ordered halt to the defendant's copycat feature". It also asked for an unspecified amount of damages. The alleged infringer was suddenly on the receiving end of a lawsuit.

It was ironic that just when barnesandnoble.com finally decided to fight it out in the court of game instead of the court of law, it got slapped with an injunction in the federal court right in the middle of the shopping season.

This sent the alarm signal to the web developer community which was by now feeling threatened. Based on this preliminary injunction, many felt that they might be the next target. No matter how absurd it is that a patent exists for a method that is staple for any web programmer, nevertheless the justice system could crush a small start-up instantly - shooting first and asking questions later.

The last straw on the back of the camel was when Amazon.com was awarded, in late February, a patent for another 'invention': that of keeping track of referrals from associates that send customers to its site and receive a commission in return. Although no announcement is found on the Press Releases pages of the proud patent assignee, you may read it at: www.patents.ibm.com/de tails?pn=US06029141_.

So, returning to my title, if I were Jeff Bezos I would not claim that "we spent thousands of hours in developing our 1-Click process ... [and took] these kinds of risks and made these kinds of investments for customers."

There was no risk, and the thousands of hours development he claims for the 1-Click process is pathetic, and worse if it is true. Bezos himself admitted that the process would have been developed without the chance for a patent.

Nor would I apply for patents on methods or systems that have been released to the internet community by enthusiastic developers to help it flourish, because it shows disrespect for fellow entrepreneurs, greed, bad taste and bad judgement.

And I wouldn't file a lawsuit against my competitor, but would take revenge by further improving my services, and flash my better traffic statistics and quarterly statement to lure back my shareholders who deserted, and loyal customers who boycott my business and lobby the internet community to do so: www.noamazon.com.

I wouldn't pretend - with two patents and an injunction order in my pocket - to be a crusader for reducing patent protection for software to three or five years (which is an eternity in internet time) because most of those who read my open letter would know that it is unlikely to pass Congress in the foreseeable future and, even if it would, it would not be implemented retroactively

Finally, I wouldn't waste my time entangling in legal wrangling. Instead I would return to what I can do best, coming up with really new ideas and sorts of spiffy implementation that won the heart of millions.

Péter Jacsó is an associate professor at the Information and Computer Science Department of the University of Hawaii. His email is: jacso@hawaii.edu