End of the road for JPEGs?
how Forgent threw a spanner into open standards
- Forgent's own website including the original press release.
- JPEG committees website with details of Forgent's claim and other information.
What do the companies Amazon, British Telecom and Forgent Networks all have in common? The answer is that they are all claiming (or have recently claimed) controversial ownership of patents. On the 11th July Forgent, based in Texas, laid claim to a key component of the JPEG picture format. Forgent Networks are actively seeking licence agreements with any corporation producing software which utilises the JPEG format. If Forgents claim is upheld this could change the way JPEGs are used in all sorts of applications, including internet sites, mobile phones and Personal Digital Assistants (PDAs). But do Forgent have a valid claim? And if they do what does this mean for the millions of people who use computers either for recreational or business use?
Background
The JPEG file format was developed in the early 1980s by the Joint Photographic Experts Group (where JPEG gets its name from) and adopted by the International Standards Office (ISO) as the standard for photographs displayed on computers. In 1987 a patent was filed by Compression Labs Ltd who carried out most of the work on the way JPEGs are compressed. Compression Labs was subsequently bought out by Forgent in 1997.Forgent, formally known as VTEL, were once a highly profitable video-conferencing company. In 1997 it was taking in about $200 million in revenue from conferencing solutions, yet this success wasnt to last. In 2001 sales fell to $38million. A restructure followed with the loss of about 250 jobs and a new management structure. Richard Snyder took over as Chief Executive and the company began exploring its back catalogue of patents, and ownership of Intellectual Properties. Since then Forgent have been seeking licence agreements from companies who use JPEG reliant technology and exploring other possible sources of income. Snyder is quoted on Forgents website as saying:
We wanted to ensure the investment community and the general public are clear about the terms of our valuable JPEG data compression technology, one of the many technologies we have in our patent portfolio.He goes on to say that
We are in on going discussions with other manufacturers of digital still cameras, printers, scanners and other products that use JPEG technology for licensing opportunities.In other words any company who creates software which reads JPEG files will be asked to purchase a licence from Forgent. Two companies, Sony Corp and another, undisclosed, Japanese company, have reportedly already done just that. They have apparently paid somewhere between $8m and $15m each for licences to use the JPEG technology. But software patents are a very complicated issue.
For starters some have raised doubts about the legality of Forgents claim.
The Issue of Patents
US Patent number 4,698,672 (hereafter called 672) was filed by Compression Labs in 1987 with the American Patent Office but it wasnt until 2002 that Forgent decided to make licences an issue. Does it have a valid claim or is this just wishful thinking by Forgent? Many people believe that Forgents claim is worthless. To begin with the patent is due to expire in 2004 which doesnt leave Forgent much time to cash in on this means of generating income. Once the patent expires then everyone will be able to use JPEGs and not have to pay Forgent a penny. As to the legality of the situation - there opinion is divided.Forgent didnt do any of the work on JPEGs; that was all done by Compression Labs and the other companies which contributed to the work undertaken by the JPEG committee. The following statement appears on the JPEG website:
It has always been a strong goal of the JPEG committee that its standards should be implementable in their baseline form without payment of royalty and licence fees, and the committee would like to record their disappointment that some organisations appear to be working in conflict with this goal. Considerable time has been spent in committee in attempting to either arrange licensing on these terms, or in avoiding existing intellectual property, and many hundreds of organisations and academic communities have supported us in our work.But while it may be argued that Forgents claim goes against the spirit of the JPEG committee they do legally own Compression Labs so they also legally own everything Compression owned.
Part of the problem is in the issue of patents themselves. The idea of patents is a good one. They are supposed to give the inventor of a new product the right to call it his or her own and ensure that they get the proper recognition for their hard work. Patents were meant to protect the small inventor from the large corporations. Originally it was only tangible items such as machinery or computer hardware that could be patented. Computer software wasn't patentable. It wasnt until the early 1980s that American policy, under the Reagan Administration, changed to allow computer software to also be patented. It was then that problems began to occur in the way patents are granted.
Most patents dont cover the entire piece of software but a small part of the process. For example patent 672 doesnt cover JPEGs as a whole but rather the way JPEGs are compressed. This could potentially mean that one piece of software could actually infringe many different patents and therefore be subject to more than one licence. Should this happen then a domino effect occurs with the cost of development rising dramatically and the price of software increasing in order to cover the cost of multiple licensing. It is rumoured that two other companies, Phillips and Lucent, both have patents which could apply to the JPEG format. This could mean that should Forgents claim make it to the courts, and should they win their case, then many more such cases could follow.
To further complicate matters software developers will often not know when they produce a new program what, if any, patents the new software will infringe. At least not until after the software has been published. The reason for this is that, to conduct a patent search, i.e. to find out whether the good idea has been thought of before, can cost thousands of dollars and still not produce fully accurate results. To research every component of a new piece of software could cost several million dollars. This clearly puts the cost of such research out of the reach of small and independent software developers. The only solution is to publish and be damned i.e. produce the software and wait to be sued!
The above all means that patents, far from protecting the individual, are now a main source of income and method of control by large corporations. The number of patents being granted has increased dramatically in the last few years. Should the trend continue it will reach the point where only large corporations can afford to develop new pieces of software. The small, independent software developers will not be able to afford to take the risk of being taken to court over a patent they couldnt have possibly have known existed.
It has been suggested by the League for Programming Freedom that part of the problem lies with the patent office itself. Until the early 1990s the American Patent office had very few computer science graduates as examiners. They also offered a salary which was just not competitive with the computer software industry. Why would a well qualified computer science graduate work for the Patent Office when he or she could potentially earn far more working in the commercial market? What this lead to was patents being granted, or refused, by people only questionably qualified to make those judgements. What makes the task harder is the lack of published material with regards certain programming techniques. This problem arises for two main reasons; firstly the techniques may not be published because the developers dont want others finding out their secrets; nd secondly, perhaps the more important reason, is that sometimes a developer may come up with a new technique which seems so obvious that surely someone, somewhere must have already patented it - and patents can only be granted for new inventions that are not deemed too obvious.
The question remains whether patent 672 should have been granted in the first place. The JPEG committee has issued a statement on its website which says that the committee:
believes that prior art exists in areas in which the patent might claim application to ISO/IEC 10918-1.The question of prior art is an important one. If the JPEG committee can prove that examples of JPEGs existed in the public domain prior to the application for patent 672 then the patent could be invalid. On the JPEG committee's website (www.JPEG.org) they have been calling for examples of prior art to be submitted in order to attempt to persuade Forgent to back down from its patent claim.
The existence of prior art does not necessarily invalidate a patent. The League for Programming Freedom, in an article entitled Against Software Patents (1991), gives an example of prior art not helping defeat a patent. The Massachusetts Institute of Technology (MIT) produced an operating system called X Windows and distributed it free to computer companies. AT&T then sued over patent 4,555,775 which covered the use of backing store in a window system which lets multiple programs have windows. In this case 'backing store' referred to the means by which the contents of the window, that is partially hidden, are saved in off-screen memory space so they can be restored quickly if the obscuring window disappears. The technique for such backing store was used by MIT before AT&T applied for a patent but MIT didnt patent the idea, partially because the solution seemed so obvious. AT&T did apply for a patent and MIT have now to pay AT&T to use their own technology despite the existence of prior art.
One final twist in the patent tail is that it's doubtful whether patent 672 is actually relevant to JPEGs at all! Tom Lane, organiser of the Independent JPEG committee is quoted as saying that:
The patent describes an encoding method that is clearly not like what JPEG does. The patent describes a three way symbol classification; the closest analog in JPEG is a two-way classification. If the jury can count higher than two the case will fail.Other industry experts also believe that Forgents claim is either invalid or simply not relevant. But if that is the case then the question remains as to why Sony, and the other Japanese company, decided to pay for licences?
A Question of money
Both Sony and the other Japanese company have declined to comment publicly so its pure speculation as to their reasons. One viable explanation is money. It was possibly believed that the easiest, and cheapest, way to make this matter go away would be to pay the licence fee. $8m is a very small amount to a corporation the size of Sony. If, on the other hand, they decided to take Forgent to court then it could have cost them a lot more. Any company involved in legal action tends to take a hit on the share price. The combination of a reduced share price; the possibility of losing the case and being subject to expensive legal fees; and a higher licence fee should Forgent decide to make them pay for the tenacity of arguing; possibly made Sony decide that $8m wasnt really worth arguing about.So what next?
Forgent seem determined to get as much money during the remaining lifespan of the patent as possible. Should they pursue this unchecked then the ISO will have no alternative but to remove the JPEG as an international standard. The ISO will not allow a standard to be in place that a private company is charging people to use as that goes against the very nature of the ISO standard. ISO themselves cant take any action against Forgent and will simply wait for a decision to be made in the courts and then act appropriately.
As companies are being required to make money in order to discharge their obligations to their shareholders the whole matter seems to lie with legislation to clear up the mess. This is true of all patent issues including Amazon, BT and Forgent. The only way this particular matter will be resolved would be if a large enough corporation were to take Forgent to court. There are signs that this might happen. Mark Nelson, an independent expert on compression technology who is co-author of The Data Compression Handbook (M&T Books 1995) is quoted in the New York Times as saying that he has been contacted by lawyers representing a Silicon Valley company in an effort of oppose patent 672. As yet there is no further information available on this case but more is expected to appear soon and we will try to follow the action in these pages.
Should the JPEG standard become redundant then developers will have to begin to find a new standard or file type which is not JPEG or based on a JPEG derivative, i.e. doesnt use the same compression techniques as current JPEGs. While this is not an ideal solution it has been done before. When Unisys claimed to have the patent on GIFs developers switched to PNG, (pronounced PING for those who want to know); look at any website, though and you will still nd the numbers of GIFs far outstripping PNGs.
The good news for individual users of JPEGs is that whether Forgent persist in seeking licence agreements or not, there is likely to be little change. For developers the whole matter depends on Forgent and whether anyone resorts to legal action to defeat Forgents claim. The general feeling, from image specialists amongst others, is that Forgents claim will be defeated and that there are too many grey areas for it to stand up to the intense scrutiny of the law courts - but it has to reach the courts.
There isnt much else anyone, either individual users nor developers, can do. The best approach would be to take a look at the JPEG committees website and assist them wherever possible. Other than that its a wait and see scenario. This has the potential to hit small developers much harder than the likes of Microsoft and company and could very well stifle new ideas. We will keep an eye on developments and welcome comments and further news snippets.
Colin Cartmell-Browne is Staff Writer with Acorn Publisher magazine.
He achieved a masters degree in Media and Culture in 1998
following his BA in Communication Studies in 1997.
He can be contacted by emailing colin@acornpublisher.com.
References - in most cases you'll need to search for Forgent, JPEG or both:
- League for Programming Freedom
- BBC articles, Patently Absurd and "Picture Perfect Patent" by Bill Thompson
- zdnet online news journal. Forgent claims JPEG patent by Robert Lemos
- New York Times Patent Claim Strikes an Electronics Nerve. By John Markoff
- IT World online news about IT issues. Forgent claims JPEG; others cry foul By Gillian Law
- US Patent Office Details of patents online as well as general information with regards patents.
The contents of this page are copyright the author and AP On-line. This article must not be reproduced without the express permission of the publishers. Whilst we take care to ensure that the facts described in AP On-line are correct we appreciate that errors will occur and will try to ensure that such errors are amended as soon as possible after they are notified to us.