Patent search - Intellectual property

Software intellectual property is an issue for everybody involved in using or producing software. However, so far, only large firms with legal departments paid the needed attention to this issue. Smaller companies tend to see software patents as an unnecessary burden; refuse to spend time on Intellectual Property issues and when they realize that there is a danger at doing so lobby to ban software patents. We believe that it is a short-minded view for the following reasons:

• You can doubt the value of patents for the progress of useful arts or as an incentive for innovation but if you accept the patent system you cannot easily exclude software inventions from that system. Before software patents were officially allowed (1997 for USA) companies patented apparatus made of a computer and of a process or method. The software patent law probably has to evolve but it is here to last.

• Authors rightfully noted that the lack of software patents did not prevent the tremendous development of the software industry. However this development happened in a scientist community. Since that time the computer society has changed. Now it may need Intellectual Property to innovate. Most companies analyze competitor products and copy their most successful functions. The safest way to succeed is to implement, promote and support better what others have already done. This is also the easiest way. Without experimentation phase a development is easier and can be released on time. This is actually easy enough to allow countries with lower labor costs to deliver the same clones with the same high quality and tight schedules. The current lack of protection may actually kill innovation in a domain where there are still many things to invent.

In this section we focus on smaller companies. Large firms develop sensible patenting strategies with the help of lawyers and consultants. They need more data than we can possibly present here. However this section may also be useful for people who work in these large firms on a business uncovered by the patent strategy and understand that ignoring Intellectual Property for any activity of the large firm is a suicidal negligence.

A company has three ways to protect its intellectual property:

• Secrecy

• Patenting

• Publication

• Open Source

Secrecy is the default Intellectual Property policy. If nothing is done programs, documentation, ideas remain hidden in the hard disks of programmers and sometimes in the company repositories. Default secrecy is free but has several drawbacks:

1. The management does not know what are the software assets of the company

2. Code and design reuse is impossible except for the original author

3. In case of patent dispute the company may be unable to demonstrate that it knew or used the disputed invention before the patent was filed

To address the first issue the management may ask for additional reporting and for a software inventory. To address the second issue the company may put in place a document management system (DMS), forums and mailing lists with search means and build an Internet-like internal system. This system may be expensive to set up and operate and is usually ineffective for the following reasons:

• In a company where secrecy is enforced the best strategy for a programmer is to keep secret her knowledge and findings. She may be laid off. In such circumstances less she disclosed in her current job more she can reuse in her new job. She also has fewer chances to be fired if nobody can easily replace her. [I think however that her long term interest is to disclose her findings to her peers and managers because she needs feedback to make progress.]

• These DMS, forums and mailing lists are often underused. To get users these tools (1) must contain enough entries (2) answer user questions (3) not have competing means: if the person who can answer is at the next office why recording the question? Such tools are only effective when they are the most convenient medium for asking a question or sending a message because of language barriers and power distances.

To address the third issue companies can deposit their program sources and documentation in escrow.

Escrow serves three functions:

1. A customer may ask that the provider deposits the components agreed upon in the contract in escrow. Then if the provider goes bankrupt or abandons the product the customer can get the code and documentation from the software escrow. The customer may designate its escrow or agree to use the provider escrow.

2. A company can make deposits in an escrow to prove the dates when it made the inventions used in its products.

3. A company can make deposits in an escrow to prove that former employees stole a part of its intellectual property

With escrow a trusted third party can assert that the company owned sources, documentation and implied processes and methods at a given date. Because the company did not publish its processes or methods someone else can patent it. The patentee cannot ask for patent fees if the company can prove that it owned the invention before the filing date. However because it has no knowledge of the owned process or method, the patentee cannot find that the company had possession of the invention and invisibly abandon legal actions against the company. Therefore the company must address the patentee action, usually hire a lawyer and often go to court. Remember that the plaintiff complains to the richest party. If the company is a small software publisher that sells to large firms the plaintiff may choose to sue the largest customers. Therefore the company may suffer from the dispute even it is able to present convincing evidences of its ownership. Some Internet documents suggest that prior possession is not necessarily considered in all countries and that a company must either patent or make public in some places.

Escrow service is relatively inexpensive. You can find many escrows on Google with the keywords "software", "deposit" and "escrow".

A common attitude is program fetishism. Program fetishists put only sources in escrow and consider that the major risk that they face is disgruntled employees that steal programs.

• Their attitude is inconsistent: Why spending so much money in marketing and analysis if only programs matter? In reality a good product combines a good definition, a clever design and a decent coding and the company asset is made of the product, of its documentation and of a deep knowledge of the product. In that list code sources are the most tangible and the less important item.

• By default programs belong to the company and not to the programmer. Companies must ensure that employees do not reuse elsewhere programs written by other programmers. Trying to ensure that programmers do not reuse elsewhere programs they have written may have an adverse impact: companies depend on the programmer cooperation to keep the design documents and the documentation synchronized by the programs. They also depend on the programmer memory to retrieve relevant prior art. If she feels that she has not the right to reuse her code a programmer may not disclose her most important findings.

• Program fetishists may be right when the communication between management and programmers is so bad that the product definition and design is not accurately tracked. This situation may harm the company in case of dispute. A design document that describes a key patented claim can be enough to put a defendant out of danger. Proving that a piece of code actually implemented the process or method implies involving third party experts and requires more time and money.

A company must put in escrow:

• Source code

• Manuals (especially the user and reference manuals)

• Trainings

• Documentation and screenshots of the user interface, especially at an early stage when manuals and courses are not yet written

• Product definition and design documents. This is important to also put in escrow abandoned design documents.

• Analysis documents

Secrecy is what you get when you do nothing and is cheap. Secrecy is convenient to hide little secrets like "borrowed" and plagiarized code and poorly written and tested programs. More important secrecy does not help a company to assess its strengths and weaknesses and maximizes its exposure to patent infringement liability.

To patent you need competent resources or help. Patenting is not an option if

1. nobody in your company has a working knowledge of patent writing and prosecution, or

2. you do not have a legal department and a full time contract with a patent attorney, or

3. you have an incubator that does not provide such service, or

4. you do not partner with an experienced university.

If your company has a patent portfolio strategy you can patent almost everything novel and inventive and you should turn your attention at filing good quality patents in cooperation with your lawyers.

Otherwise consider that patenting like programming is a job with a sharp experience curve. Your first patent will cost you more than the second and so forth. Therefore filing a first patent is a strategic decision. For this decision to be profitable you need to file other patents and to dedicate a significant part of your resources to patenting. When we analyzed 5,838,906 we found that the assignee, a small company called Eolas owned at least seven patents and that 5,838,906 was indeed the second Eolas patent.

You must be honest about your position on the market. If you are a me-too company you probably should not patent. You need to be a technological leader that (1) by chance had to implement something substantially different of what already exists (2) has good chances to have again to implement novel processes or methods because you will need to file subsequent patents. Even if your invention is really novel, non obvious and useful the software industry can follow a different path and ignore your patent. Patenting is a risky business.

Eventually you have to go fast. We have seen that browser controls and plug-ins would have invented anyway between 1994 and 1995. This is not unique to software inventions. Without Bell and Edison the telephone and the electric bulb would have been invented anyway when they were invented. At a given time the technology is mature and different teams work independently on the same idea. You must quickly and thoroughly assess the novelty and the potential of the invention, write a patent draft, explain the invention to lawyers, check their rewording. It does not finish when the application is filed. The examiner can require changes and even reject the application. This is not easy in a small company. Filing a patent is urgent but to the opposite of product deadlines there is nobody to remind you that it is urgent.

Here is an introduction to patent writing. This explanation is useful even if you do not plan to file a patent. Before starting a new project you should look for possible infringement. For this search as we have seen in the toolkit section the best is to create a reference text that looks like a patent and to use this text to find patents with the same word frequencies. This work complements a design. It points out the differences between your design and existing products and lists the problems that you were trying to solve. You should refine your product definition according to your finding in prior art:

We may have to iterate the product definition, the design, the patent writing and the prior art search. The design answers the feasibility question and the patent search the legal question. For prior art search you do not need to write a full patent. You may focus on background, prior art and summary.

According to the USPTO guidelines a patent is made of three sections:

• An abstract

• Claims defining "in words the boundaries of the invention so that the public will know what the invention is and can avoid infringing it." Claim wording is a patent attorney job.

• A description

The reference guide for writing a patent is the Manual of Patent Examining Procedure (MPEP). Read in particular the Chapter 2100 Patentability.

We found this useful definition:

"A patent abstract is a concise statement of the technical disclosure of the patent and should include that which is new in the art to which the invention pertains. If the patent is of basic nature, the entire technical disclosure may be new in the art and the abstract should be directed to the entire disclosure. If the patent is in the nature of an improvement in an old apparatus, process, product or composition, the abstract should include the technical disclosure of the improvement. [...] If the new disclosure involves modifications or alternatives, the abstract should mention by way of example the preferred modification or alternative.

The abstract should not refer to purported merits or speculative applications of the invention and should not compare the invention with the prior art.

Where applicable, the abstract should include the following:

1. if a machine or apparatus, its organization and operation; [...]

2. if an article, its method of making;

3. if a process, the steps."

The function of claims is explained in 35 USC 112: "The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention."

Here is an example of claim:

"A computer-implemented method of searching for an item in a plurality of independently operated electronic auctions interconnected by a computer network, each electronic auction having an associated data repository, the method comprising:

• receiving input identifying an item; and

• instructing a software search agent to search for the item on the computer network in the respective data repositories of one or more of the electronic auctions."

It has a preamble and a number of elements (words or phrases) that describe the claim limitations. There is infringement when another party use a system with each and every limitation of a claim. So the inventor must include the absolute minimal set of elements that differentiate a new invention over what came before. If US patents contain many claims this is to maximize the chances to get a match. When a claim contains many limitations this is usually because the examiner rejected the first version of the claim for lack of novelty (it was anticipated by prior art) or obviousness and because the applicant had to add additional limitations to get a patent.

This claim has a function, which is to search "an item in a plurality of independently operated electronic auctions interconnected by a computer network" and two steps:

1. "receiving input identifying an item; and

2. instructing a software search agent to search for the item on the computer network in the respective data repositories of one or more of the electronic auctions."

Such claims are called means or steps plus function claims. They are so important that 35 USC 112 describes them: "An element in a claim for a combination may be expressed as a means or step for performing a specified function without the recital of structure, material, or acts in support thereof, and such claim shall be construed to cover the corresponding structure, material, or acts described in the specification and equivalents thereof."

So the precise scope of such claims is determined by the claim and by the description. There was a trial for infringement of the patent that contains the claim above. The court determined the meaning of words like item and also whether the steps recited a sequence. To learn more about this case you may look at the Business method page. For details about the construction of means plus function claims you may look at http://www.uspto.gov/go/og/con/files/cons089.htm and http://jip.kentlaw.edu/nart/2004/A-2.htm.

As implicitly stated by 35 USC 112 a claim may also recite structure, material, or acts. In this case the claim loses its function limitation. A claim for a popcorn dispenser was rejected because the claimed structure was the same as the structure of an oil can.

The description is made of:

• Title: short, brief and specific

• The field of the invention: a broad statement that should start with something like "The present invention relates to". In case of 5,838,906 it was "This invention relates generally to manipulating data in a computer network, and specifically to retrieving, presenting and manipulating embedded program objects in distributed hypermedia systems."

• Background information: what people will need to understand, search or examine your invention. Do not assume that the reader has a prior knowledge of your job. Present your activity like you would do with your neighbor or your mom.

• Prior art: Problems that inventors have faced in this area and how they have attempted to solve them. You must search patents and prior art in the area because (1) your invention may have been already invented (2) as we have seen in the reference metric discussion you must cite the prior art and patents you are aware about. This is important because a patent shall be presumed valid as said in 35 USC 282 but "while the presentation at trial of a reference that was not before the examiner does not change the presumption of validity, the alleged infringer’s burden may be more easily carried because of this additional reference." So an alleged infringer can more easily demonstrate that your patent is invalid if you omit relevant prior art in your reference list. In the prior art section briefly describe how these prior art and patents are working and the problems that they do not solve.

• Summary of the invention: How your invention solves one or several of these problems. What you are trying to show is how your invention is new and different.

• Drawing and screenshots

• Detailed description: Must describe the invention parts in sufficient detail so that someone could reproduce at least one version of your invention. Describe each part of the invention, how parts fit together and how they work together. For a process describe each step, the initial state and the result.

• Example of an intended use for your invention (best mode)

When you write your first patent document the most important problems you face is that you do not have the right mindset and you were not prepared to describe your work in such terms. A patent is neither a marketing nor a analysis document. Use the same sentence structure and the same verbs as in IBM patents (on the USPTO advanced search an/IBM). Check that you do not use words like innovative, powerful, successful, synergy, strategy and project. Do not use the word product outside the background and the prior art section.

If patenting is not an option you can make your invention public. Your competitors will be able to use it but not to patent it because a public process or method is prior art. This can actually help in your business. Software patents are repellent to competitors and if you are the only company doing things in one way your product will not sell. If you made your invention public competitors will use it and because you have some advance you should increase your market shares.

As we have seen in the 5,838,906 case a public mailing list whose readers are concerned by the invention matter is sufficient because "An electronic publication, including an on-line database or Internet publication, is considered to be a 'printed publication' within the meaning of 35 U.S.C. 102(a) and (b) provided the publication was accessible to persons concerned with the art to which the document relates." A mail is probably the cheapest and fastest way to make public an invention.

On the other hand the court of MercExchange v. eBay found that a newsgroup posting does not have the status of a printed publication. It gave two reasons:

1. There was no guarantee that any party would have been granted permission to join in the newsgroup. In our view this reason is not valid for newsgroups but it may be for some forums.

2. Newsgroups have a retention period after which they are removed. This period depends on the disk space allocated to the newsgroup and is therefore variable. So it is difficult to find a posteriori how long a message remained posted.

So to make public an invention we recommend to not use newsgroups and forums whose read access is password protected.

According to most national laws a mailing list can be in a country and a language different of the court country and language. However, to be on the safe side, make your disclosures in English.

There are two other solutions that you can combine with mails to mailing lists:

1. Publish an article on a newspaper or on an electronic publication. Writing an article is not obvious but is simpler and faster than writing a patent. An article has to be a story of around 3000 words that must capture the reader attention at the beginning and keep it up to the end. You can use a plan background - prior art - description. There are side benefits at writing articles. A customer who has liked your paper is more likely to buy your product. In case of newspaper publication you can use reprints to target technicians who are generally insensitive to traditional marketing material. In case of electronic publication you can link your site to the article.

2. Present the invention to a conference. It may imply joining an organization and paying for membership. You also have travel expenses. The presentation cost is not necessarily lower than the patenting cost. If you already participate to an organization the presentation can be a good solution because it allows more flexibility than an article. You can also use the background - prior art – description plan.

Both solutions produce prior art of premium quality but they have a common drawback, which is to take time. You usually submit a proposal either to the publisher or to the organization. Your proposal may be approved with some amendments one month later. Then you send your article or presentation. Next the publisher or organization has to find a slot for you in its next issues or meetings. In case of newspaper article or presentation it takes between four and six months. Electronic publication is faster. According to most national laws the article can be written or the presentation given in a country and language different of the court country and language. However, to be on the safe side, make your disclosures in English.

A publication on your Web site is usually not considered as an Internet publication because you cannot prove the last modification date of your publication. There are however a couple of cases worth to consider:

• If your Web site is well indexed by search engines and is making 100,000 visits per month when your competitor files its patent, there are good chances that (1) at least in USA, you competitor will have to cite your site and show how its invention is different of your process (2) the examiner will check your site and reject the application. In both case you avoid a lawsuit.

• There are Internet archives. Maybe the most famous is http://www.archive.org/. If your site is popular your pages may be recorded every month on this site and this facility is routinely used by courts and examiners. Take care of not changing the page locations because this can confuse the Internet archivists.

A popular and content-rich web site is not only a contribution to Internet knowledge and a way to attract programmers. It can also limit your exposure to patent lawsuits.

Some companies offer a technical disclosure service, typically on Internet. An example is http://www.ip.com. For a single publication the price is $155 (November 2003). This kind of service is interesting because you do not have the travel and membership fee associated to a presentation and you do not need the writing skills needed for an article. You can just publish a cleaned up version of a design document.

The most radical solution is to release your core code with an Open Source or free (GPL) license.

The best location for this is probably SourceForge because this site is well known of persons concerned with most computer arts. Two key points are:

• "Only Open Source projects may be hosted on SourceForge.net. Projects are expected to meet the requirements of the Open Source Definition and abide by the spirit of that definition. With this in mind, it is important to understand that projects hosted on SourceForge.net are expected to release their source code (as per Clause 2 of the Open Source Definition) -- projects may not make binary-only releases."

• "When registering your project, you will be expected to provide a full description of your project, information about which Open Source license under which your project will be released, and the name of your project."

SourceForge registration and hosting are free.

According to the OSI definition an Open Source license must conform to the following rules:

1. Free Redistribution. The license shall not restrict any party from selling or giving away the software.

2. Source Code. The license must allow distribution in source code as well as compiled form. Where some form of a product is not distributed with source code, there must be a well-publicized means of obtaining the source code for no more than a reasonable reproduction cost preferably, downloading via the Internet without charge.

3. Derived Works. The license must allow modifications and derived works, and must allow them to be distributed under the same terms as the license of the original software.

4. Integrity of The Author's Source Code

5. No Discrimination Against Persons or Groups

6. No Discrimination Against Fields of Endeavor. This clause exists to prohibit license traps that prevent open source from being used commercially

7. Distribution of License. This clause is intended to forbid closing up software by indirect means such as requiring a non-disclosure agreement.

8. License Must Not Be Specific to a Product

9. The License Must Not Restrict Other Software. GPL is conformant with this requirement. Software linked with GPL libraries only inherits the GPL if it forms a single work, not any software with which they are merely distributed.

10. The License must be technology-neutral. This clause actually means that the product can be distributed in FTP or on CDROM.

The OSI definition comes actually from Debian. Debian is a Linux distribution and the authors of the license definition had in mind to give a definition enabling a business including:

1. Stable rules

2. Third-party distribution

3. Newspaper distribution

4. A compelling offer to users. People want free products and at the same time to release commercial products.

This section aims to help you choosing a license. The two most important families of Open Source licenses are BSD-based licenses are GNU licenses.

The Berkeley University had developed a number of programs under this license that were used in institutions and industry. This license addresses three needs:

1. Disclaimer: Berkeley doesn't want to be liable to prosecution for bugs in its programs

2. Copyright: Berkeley wants to enforce its copyright to show that it wisely use public funds

3. The name of the university and of the author may not be used to endorse or promote products derived from this software

These needs were typical of 70’s and 80’s University needs. The following products have a BSD-based license:

• Java

• Apache products

• XFree 86 and other licenses used with XFree 86

Quite interestingly the BSD license is titled "Berkeley-based copyrights". The most important constraint of this license is that you have to advertise the contribution of the publisher. However there are even more user-friendly licenses like the ACE / TAO and the MIT license used with some changes by Expat. Use a BSD-based license when you want to facilitate the life of your users, for instance:

• You offer a Web service. You want to provide free tools to your users for using your service.

• A competitor took all the market but this market is so important for you or so profitable for your competitor that you are ready to give away your code to come back on this market

As far as Intellectual Property is concerned you do not need to use a BSD-based license and GNU licenses usually serve better your needs because they prevent users to derive non-free products from your code. We use free here because the father of GNU, Richard Stallman rejects the "Open Source" term and prefers the "free" word: "When we speak of free software, we are referring to freedom of use, not price." Note that in the rest of this document when we talk about Open Source we mean Open Source or free. The two most important GNU licenses are the GNU General Public license, GPL in short and the Lesser General Public License, LGPL in short. These licenses enforce the copyleft principle, which states that you must release a modified [L]GPL program with the same license as the original program. Things are however more complex because of derivative works.

First the difference between GPL and LGPL is that the LGPL license allows linking non-free programs to LGPL libraries whereas GPL doesn't allow it. Here is the rationale: "When a program is linked with a library, whether statically or using a shared library, the combination of the two is legally speaking a combined work, a derivative of the original library. The ordinary General Public License therefore permits such linking only if the entire combination fits its criteria of freedom. The Lesser General Public License permits more lax criteria for linking other code with the library." More precisely the LGPL permits to programs containing no derivative of any portion of the Library, but designed to work with the Library by being compiled or dynamically linked with it, to be a non-free product. An executable statically linked with the library contains the library and therefore should be released under a [L]GPL license but objects used to build the executable that don't contain portions of the library don't need to.

GNU license put no restriction on the program use. The GPL license says: "Activities other than copying, distribution and modification are not covered by this License; they are outside its scope. The act of running the Program is not restricted, and the output from the Program is covered only if its contents constitute a work based on the Program (independent of having been made by running the Program)." The restriction on the output is clarified by the GNU FAQ: "If the user uses your program to enter or convert his own data, the copyright on the output belongs to him, not you. More generally, when a program translates its input into some other form, the copyright status of the output inherits that of the input it was generated from. So the only way you have a say in the use of the output is if substantial parts of the output are copied (more or less) from text in your program. For instance, part of the output of Bison (see above) would be covered by the GNU GPL, if we had not made an exception in this specific case."

GPL is the license of GCC and most Linux programs. LGPL is the license of the GCC library.

The GNU wrote a couple of documents of interest for choosing a license:

• An list of Open Source and free licenses called Various Licenses and Comments about Them

• Explanations about free software and OSI that you find here

• A comment about the BSD license

In some circumstances releasing your core code with an Open Source or free (GPL) license is a rational choice from an economical and legal point of view. When a new market emerges a patent and publication race starts. If you never patented and never published you have almost no chance to win. If you distribute your core product and documentation with an Open Source license starting with the alpha version you have a good chance to win the race even if you have less resources than your competitors. In that respect Open Source acts as publication. Your competitors will be able to use your invention but not to patent it because your implementation is prior art. Obviously your competitors can also use your implementation. This is not necessarily a bad thing:

• You should be able to convince your customers that you know better your code and that you can support it

• If your competitors use your product you control a part of the market

Open Source has other strong points:

• When you release your core product with an Open Source you make a superior proposal to your customer. As we have seen above customers often require a source deposit in escrow. The escrow checks that the deposit conforms to the contract but the customer has no access to the code.

• Customer’s programmers may test your core product and use it to make some prototypes and development before signing a contract with you.

You still can sell a product that uses the Open Source part and contains other facilities like configuration and administration user interfaces. The differences between the core part and the whole product are the prerequisites. The core part requires an intimate knowledge of the product technology whereas the whole product can be administrated and used by almost anybody. While they do not usually contain a significant amount of Intellectual property the extra facilities are essential to the product success and frequently more expensive to implement that the core product.

There are other solutions:

1. You can use a license that does not grant the right to use the product for free when you get paid for using it. An example is the Caucho Developer Source License used for the Resin Application server. Note however that such licence may not conform to the OSI definition.

2. You can use a dual license like MySQL. On the download page of MySQL users are given the choice between a GPL version provided "as is" without any warranty and a commercial version the user needs to make commercial derivatives of MySQL or to get a warranty or support. The CEO of Trolltech, an early adopter of dual licenses, explained: "Berkeley DB was originally available from UC Berkeley under the popular open source Berkeley Software Distribution (BSD) license. When we decided to form a business around Berkeley DB, we looked at the BSD license closely. The BSD license would give our customers flexibility. It would allow them to freely embed Berkeley DB in hardware or software products and distribute their product without restriction. However, the BSD license would also allow our customers or any developer to take our code and create their own products without our permission or payment. We would create our own competition. We decided to begin the business with a dual-license strategy [GPL and commercial]." Dual license works well in the common case in which the GPLed product has a very small developer community outside the company that support it.

You may read that Open Source and free software are untested in court and therefore unproven. As we have seen in the Patent issues section software Intellectual property will continue to evolve, especially regarding how much care must be taken by a company to get an enforceable patent. Furthermore this is only half the truth. From time to time free software goes to court. Before considering Open Source legal issues we must first briefly explain what Copyright and trademark are.

A copyright comes into effect immediately, as soon as something that can be protected is created and "fixed" in some way. A copyright is an unregistered right. In most country the copyright notice is not even necessary. Writing the copyright symbol (c) followed by the company or organization name, the date and "all rights reserved" warns others against copying the copyrighted stuff, which may strengthen your position in case of dispute.

There is no incompatibility between Open Source licenses and Copyright. You actually must claim to be the owner of your code to license it with an Open Source license.

You can find the Copyright law of USA at http://www.copyright.gov/title17/. Copyright law is covered by Trade-Related Aspects of Intellectual Property Rights (TRIPs) like patent law. Governments try to harmonize their Copyright law like their patent law.

A trademark is any sign that can distinguish the goods and services of one trader from those of another. For more information about trademark you can read our trademark page.

The case opposed MySQLAB (plaintiff) to the NuSphere subsidiary of Progress (defendant). The two companies had formed a "strategic" alliance. Here is an excerpt from the press release: "To create an open source business, Progress Software has formed NuSphere, a Progress Software Company, to distribute and support the NuSphere MySQL open source database. Progress Software will be providing up to $2.5 million (US) to help further the development of MySQL. NuSphere will extend the core development team's efforts by creating NuSphere MySQL; a multi-platform integrated distribution of MySQL and related open source products. NuSphere will provide commercial support, consulting, and training for customers deploying MySQL as part of their web infrastructure. As a single, global source of integrated services for MySQL, NuSphere will provide support from three main support centers in the U.S., the Netherlands and Australia. NuSphere chose to support MySQL because of its position as the market-leading open source database for web infrastructure. MySQL is especially well suited for capturing web site traffic, providing authentication and authorization services, and content personalization. The MySQL database is the first widely used database product released under the Free Software Foundation's General Public License (GPL)."

We understand that at some stage NuSphere released a database product called Gemini statically linked to MySQL's code without the accompanying source code. NuSphere did another thing that may have triggered the dispute. Though MySQL has a GPL license the original MySQL Web site was http://www.mysql.com. NuSphere opened a http://www.mysql.org Web site.

MySQL asked to a Boston court to issue a preliminary injunction to bar Progress, and its subsidiary NuSphere Corp., from using MySQL's trademarks in their products and from distributing any software that included MySQL's self-titled database program. MySQL alleged that Progress failed to include the source code for a software tool NuSphere built and included in its MySQL Advantage package, which included MySQL's database. GPL requires that any software built using components released under the GPL must also include the source code of the new application. Eben Moglen, General Counsel to the Free Software Foundation (FSF) testified as an expert witness. Here is his affidavit. Ed Kelly and John Palfrey reported the meeting to Tim O’Reilly: "Judge Saris was surprised at how many people were there to hear the arguments on a motion for a preliminary injunction barring use of the MySQL mark by NuSphere. She noted the almost religious zeal of the open-source community. She also noted that both parties came loaded for bear and that MYSQL had flown in David Axmark from Sweden [MySQL is a Swedish company] and both parties had expert witnesses prepared to testify."

Judge Sarris refused to allow arguments to expand beyond the trademark dispute. Though she did not issue any orders in court, Saris made clear that she intended to issue a preliminary injunction against NuSphere's use of the marks in question and that she was not going to issue a preliminary injunction in the more complicated matter of the use of the MySQL code and terms the GPL. She urged both parties to come to an out-of-court settlement. From our perspective the most important is her comments about the GPL license:

• GPL was treated as any other license would be in a software context

• Judge Saris seemed to focus on the question of whether Gemini constitutes an independent or a derivative work (does "linking" to another program equate to creation of a derivative work?) and whether the harm caused to MySQL met the irreparable standard.

ComputerWire reported the story with different words and a different title: Judge blocks route to GPL legal test case.

Then Judge Saris issued a preliminary injunction. In the ruling she barred Progress and NuSphere from advertising or selling any products using the MySQL name and from registering or operating any Web sites that include the MySQL name. The ruling did allow Progress to state that its products interoperate with MySQL. Saris, however, declined to rule on the GPL issue, stating that MySQL had not shown a likelihood of success in the matter or irreparable harm, two prerequisites for the issuance of a preliminary injunction. MySQL seems to have the better argument in the matter, Saris wrote, but said that the release of source code by Progress in July 2001 may have solved the problem. Because of MySQL's central place in NuSphere's product line, barring the company from using the software could substantially hurt NuSphere, she said.

Eventually MySql and NuSphere concluded a settlement on November 7, 2002. We read the settlement that way: "The settlement resolves all outstanding issues between the two companies including [...] assignment to MySQL AB of copyrights for all NuSphere contributions to the MySQL program", which means that (1) the GPL license was extended to NuSphere contributions (2) the copyright of NuSphere contributions was transferred to MySQL AB. Again ComputerWire reported the settlement is a different way in MySQL, NuSphere Settle GPL Contract Dispute.

This case shows that:

• The irreparable harm prerequisite may bar preliminary injunctions in further cases. If you give an application for free you cannot claim that you suffer irreparable harm when a user infringes your license whereas the user can prove that the injunction hurts its business.

• The GPL may be found partially illegal. Much of Saris questioning surrounded whether the application could operate without MySQL. A company could for instance develop an application that can use MySQL, PostGres and Oracle for database access. Then according to the copyleft principle this company should release its code under GPL but the judge would probably find that the company does not have to. Richard Stallman wrote: "Using the ordinary GPL is not advantageous for every library. There are reasons that can make it better to use the Library GPL in certain cases. The most common case is when a free library's features are readily available for proprietary software through other alternative libraries. In that case, the library cannot give free software any particular advantage, so it is better to use the Library GPL for that library. This is why we used the Library GPL for the GNU C library. [...] However, when a library provides a significant unique capability, like GNU Readline, that's a horse of a different color. The Readline library implements input editing and history for interactive programs, and that's a facility not generally available elsewhere. Releasing it under the GPL and limiting its use to free programs gives our community a real boost." From our point of view this is not only a matter of interest. If users have an alternative to your product for addressing a certain need then you should use LGPL.

SCO filed a suit against IBM in the district court of Salt Lake City through its Caldera subsidiary. SCO wrote: "IBM has breached its own obligations to SCO [the Unix owner], induced and encouraged others to breach their obligations to SCO, interfered with SCO’s business, and engaged in unfair competition with SCO, including by misusing and misappropriating SCO’s proprietary software, incorporating (and inducing, encouraging, and enabling others to incorporate) SCO’s proprietary software into open source software offerings and incorporating (and inducing, encouraging, and enabling others to incorporate) SCO’s proprietary software into open source software offerings.

The proprietary code SCO is talking about is primarily UNIX and the Open Source software SCO is talking about is primarily Linux. SCO, summarizes the UNIX story: "AT&T Technologies originally licensed the UNIX operating system software code to approximately 30,000 software licensees, including defendant IBM, for the UNIX operating system software source code, object code and related schematics, documentation and derivative works (collectively, the UNIX Software Code). To protect the confidential and proprietary source code information, these license agreements, as detailed below, contained strict limitations on use and dissemination of UNIX Software Code. When SCO acquired the UNIX assets from Novell in 1995, it acquired rights in and to all (1) underlying, original UNIX software code developed by AT&T Bell Laboratories, including all claims against any parties relating to any right, property or asset used in the business of developing UNIX and UnixWare; (2) the sale of binary and source code licenses to various versions of UNIX and UnixWare; (3) the support of such products and (4) the sale of other products that are directly related to UNIX and UnixWare."

SCO explained the IBM strategy in this way: "In the process of moving from product offerings to services offerings, IBM dramatically increased its staff of systems integrators to 120,000 strong under the marketing brand IBM Global Services. [...] IBM sought to move the corporate enterprise computing market to a services model based on free software on Intel processors. By undermining and destroying the entire marketplace value of UNIX in the enterprise market, IBM would gain even greater advantage over all its competitors whose revenue model was based on licensing of software rather than sale of services. To accomplish the end of transforming the enterprise software market to a services-driven market, IBM set about to deliberately and improperly destroy the economic value of UNIX and particularly the economic value of UNIX on Intel-based processors."

About Linux SCO wrote: "Prior to IBM involvement, Linux was the software equivalent of a bicycle. UNIX was the software equivalent of a luxury car. To make Linux of necessary quality for use by enterprise customers, it must be re-designed so that Linux also becomes the software equivalent of a luxury car. This re-design is not technologically feasible or even possible at the enterprise level without (1) a high degree of design coordination, (2) access to expensive and sophisticated design and testing equipment; (3) access to UNIX code, methods and concepts; (4) UNIX architectural experience; and (5) a very significant financial investment."

SCO explained: "This case is not about the debate about the relative merits of proprietary versus open source software. Nor is this case about IBM right to develop and promote open source software if it decides to do so in furtherance of its independent business objectives, so long as it does so without SCO proprietary information."

You can find the reaction of the OSI. A technician may also disagree: SCO, the second-hand buyer of UNIX, did not significantly enhance UNIX. Why investing on something that does not make money? Actually ATT and Novell sold UNIX for the same reason. In early 90’s ATT had a successful Operating System but they did not find the way to convert this technical success into money. Therefore Unix licensees had to substantially invest to port and optimize UNIX performance on their hardware and to support their devices. IBM did this effort twice, one for AIX with the UNIX code, one for Linux with Linux code. A technician would not understand either how a company that releases a Linux distribution (Caldera) can find that Linux contains proprietary code.

So far this story was just another example of weak to strong action in Intellectual Property. However on October 24, 2003 in an answer to IBM SCO wrote: "The GPL violates the US Constitution together with copyright, antitrust and export control laws" (eighth affirmative defense). An article of LinuxWorld reports this news.

Before this answer Darl Mc Bride, the CEO of SCO wrote an open letter to the open source community: "In copyright law, ownership cannot be transferred without express, written authority of a copyright holder. Some have claimed that, because SCO software code was present in software distributed under the GPL, SCO has forfeited its rights to this code. Not so – SCO never gave permission, or granted rights, for this to happen. Transfer of copyright ownership without express written authority of all proper parties is null and void. Use of derivative rights in copyrighted material is defined by the scope of a license grant. An authorized derivative work may not be used beyond the scope of a license grant. License grants regarding derivative works vary from license to license – some are broad and some are narrow. In other words, the license itself defines the scope of permissive use, and licensees agree to be bound by that definition."

This letter triggered the comment from Grocklaw, another SCO group company: "You have continued to distribute the Linux kernel, despite alleging that it contains infringing source code. Simultaneously, you are attempting to compel purchase of "Linux Intellectual Property" licenses for binary-only use, the terms of which are incompatible with freedoms granted under the GPL. According to the GPL, any violation of its license terms automatically and immediately terminates your permission to modify or distribute the software or derivative works. Note the wording of the GPL:

4. You may not copy, modify, sublicense, or distribute the Program except as expressly provided under this License. Any attempt otherwise to copy, modify, sublicense or distribute the Program is void, and will automatically terminate your rights under this License. However, parties who have received copies, or rights, from you under this License will not have their licenses terminated so long as such parties remain in full compliance.

5. You are not required to accept this License, since you have not signed it. However, nothing else grants you permission to modify or distribute the Program or its derivative works. These actions are prohibited by law if you do not accept this License. Therefore, by modifying or distributing the Program (or any work based on the Program), you indicate your acceptance of this License to do so, and all its terms and conditions for copying, distributing or modifying the Program or works based on it.

Releasing software under the GPL is not the same as releasing it into the public domain. Authors retain their copyrights to software licensed under the GPL. Even when authors assign their copyrights to someone else, such as to the Free Software Foundation, the copyrights remain valid, but with the new owner. Therefore, subsequent to termination of your permissions under the GPL, you are in the unhappy position of violating the copyrights of the software authors, if you continue to distribute their software. Under copyright law, you are not allowed to distribute at all without their permission -- and they have chosen to grant that permission only by means of the GPL."

In its answer to IBM SCO does not give details about its assertion that GPL infringes the Copyright law. LinuxWorld believes that the argument is that under the GPL, mere human beings - American citizens - arrogate to themselves the right to give permission to copy, modify, or redistribute intellectual property, while that kind of regulation of copyright, according to SCO's contention, can and must only be done - per Article 1 Section 8 of the US copyright law - by Congress itself. This explanation is consistent with the open letter of Darl Mc Bride.

The US copyright law (circular 92) says: "unless authorized by the owner of copyright in a computer program any person in possession of a particular copy of a computer program may, for the purposes of direct or indirect commercial advantage, dispose of, or authorize the disposal of that computer program by rental, lease, or lending, or by any other act or practice in the nature of rental, lease, or lending" and "Any exact copies prepared in accordance with the provisions of this section may be leased, sold, or otherwise transferred, along with the copy from which such copies were prepared, only as part of the lease, sale, or other transfer of all rights in the program. Adaptations so prepared may be transferred only with the authorization of the copyright owner." Our understanding is that authorizations granted by the GPL license are well described in the Copyright law and that the issue rose by SCO is whether implicit authorization is legal or not. From our point this is legal because, for instance, opening a CDROM seal is already considered as an implicit approval of license terms. However we think that this is time to clarify this issue in a judgement.

We have seen above that Intellectual Property can no longer be ignored. We may also add that this is no longer only the job of specialists. In that respect Intellectual Property follows the evolution of the Society. Since the beginning of the 60’s people are massively educated. Up to the 80’s computers were black boxes operated by aliens talking a language that nobody else understood. Now the whole workforce has a coarse understanding on computer matters and everyone talk the computing jargon. We wrote Air transport because we wanted to illustrate the close integration of all parties involved in providing a service, companies of different size and government agencies, which is, we believe, one of the deepest changes in the Society and because we were asked to. Now people also understand what revenue management and landing slots are. We see that not-so-easy-to-understand concepts of the Air Transport and of the banking system are commonly understood. Intellectual property will become a part of our common culture. Now managers understand that in the next decade they will spend a third of their time on Intellectual property issues. In this section we explain how this time should be spent:

• Intellectual Property out: the information disclosed by a company or organization,

• Intellectual Property in: the information a company should collect.

A company or organization communicates with the rest of the world and therefore constantly discloses information about its technical choices, methods and processes through mails, messages posted on public forums, Web sites, press releases, presentations, documentation, trainings, RFP, products and patents. Employees of large firms frequently get more and better data about their company on Internet than using the firm internal communication systems. This is not always a drawback. The real issue is that Murphy’s Law also applies here: things that you would like to keep hidden quickly spread whereas things that you would like to spread are unnoticed.

Our experience is that mails and messages are less important:

1. Mails and messages frequently make sense years after their writing when a new product is released. A prototype can generate a large number of messages and be abandoned later.

2. Mails are usually about technical issues like errors, bugs, coding and administration problems.

3. Authors are usually hard to identify.

Very early in a project a company should take a final decision about its Intellectual Property policy regarding this project:

1. Secrecy

2. Patent protection

3. Publication

4. Open source

Procrastination is not an option because the date when a publication was made or a patent was filed determines who legally invented a process or method. When you take your time to decide you have to keep your innovations secret during the while to keep the choice open. If you opt for publication or for open source sooner you disclose better it is.

When you chose secrecy or patent protection you have to handle three sorts of disclosure:

1. Disclosure to customers such as documentation, trainings and products

2. Disclosure to prospects such as RFP answers and presentations and to partners

3. Disclosure to public such as Web site, press releases, brochures and some presentations (for instance in conferences)

We take the example of a software publisher.

The license agreement should include a confidentiality clause.

If you sell service with the product the service contract should also include a confidentiality clause like "Each party may disclose confidential information to the other party. The parties shall preserve of the confidentiality of such information during the performance of the service and during sometime [to be defined] as from the Agreement expiry."

Documentation manuals and training support documents should include a page with something like: "This documentation may not be copied, transferred, reproduced, disclosed or duplicated in whole or in part, without the prior consent of [your company]. This documentation is proprietary information of [your company] and protected by the applicable law of [your country] and international treaties." This page should refer to the license agreement. It may also refer to a restricted rights definition. The industry commonly follows the definition used by government for its own purchases. Magic keywords to find samples on the Web are 48 CFR Section 12.212 and 48 CFR 227.7202. 48 stands for Federal Acquisition Regulations System, 12 for Acquisition of commercial items, 212 for Computer software, 227 for Patents, data, and copyrights, 7202 for Statutes pertaining to administrative claims of infringement. You may find these Federal regulations on http://www.access.gpo.gov/nara/cfr/cfr-table-search.html.

A secret shared by thousand or more customer employees is longer a secret (a user may copy a product method if this method is not patented and competitors have your documentation) but if you follow the guidelines above this is still a legal secret: you still can file a patent and a competitor cannot show your user manual as evidence in an infringement prosecution or reference your manual in its patent applications.

You keep your intellectual property secret though a non-disclosure agreement (NDA). An NDA is a contract whereby the signer agrees not to disclose certain information, except under terms as described in the contract. In a one-way NDA one party agrees to give to a second party confidential information and the second party agrees not to share this information with anyone else for a specified period of time. In a two-way NDA both parties agree to give confidential information and to not share this information with anyone else for a specified period of time.

A one-way NDA may be fine for the first presentation to a prospect. If they have to explain how they intent to use the product, then the prospect will probably ask for a two-way NDA. In case of answer to an RFP or of partnership you also need a two-way NDA. The problem is that each party usually has its own NDA. The stronger party and when parties are of similar strength the customer tends to impose its NDA or to refuse to sign the NDA of the weaker party. This is especially true in countries and businesses in which actors are not fully aware of Intellectual Property issues. NDA negotiation is an extra burden for an already tough job. Furthermore when a sale or an RFP cycle is three to six months, when a partnership lasts two years, the consequence of negligence in an NDA may be fully appreciated eight or ten years later, which is more than the average experience of the actors in their job and sometimes more than the time they will work for their company.

You can find examples of NDA at http://www.uiausa.com/Nondisclosure.htm and http://www.uiausa.com/TutleNDA.htm.

Here we consider Web site, press releases, brochures and public presentations. Intellectually property is made of functions whose combination solves a set of problems and of how these functions are implemented. Look at the description part and at the drawings of a software patent. You will find screenshots, block diagrams and relatively high-level explanations. The description of a patent contains the same data as an article.

A site that does not disclose intellectual property is a site that does not answer to the how question. Such site does not attract visitors because it contains no information and is poorly indexed by search engines because it has no content. If you choose secrecy or patenting you cannot effectively use one of the cheapest and most effective communication means. You may however design a site with a public part and a private, password protected, part. When you can identify the users you are in the same situation as for paper documentation: the content is not public but at the same time the very last persons you would like to see on your site will be your most regular visitors. You may allow self-registration. In that case ask the user to acknowledge a confidentiality clause. You can check her identity by sending a mail with a confirmation number. Oracle uses such a system. You can find in Oracle documents: "The Programs (which include both the software and documentation) contain proprietary information of Oracle Corporation; they are provided under a license agreement containing restrictions on use and disclosure and are also protected by copyright, patent and other intellectual and industrial property laws. Reverse engineering, disassembly or decompilation of the Programs, except to the extent required to obtain interoperability with other independently created software or as specified by law, is prohibited." Only highly motivated users accept to register, which is both a drawback (few visitors) and a benefit (good quality visitors).

A press release is a news exercise and because of that an unlikely place for intellectual property disclosure. Neither the format nor the reader expectations, nor the press release goal really enable disclosures. Furthermore in a company that has software intellectual property people who create and understand this intellectual property are in development teams and not a communication services. We only mention press releases because we were asked if a press release could prevent patenting. To check if a press release contains intellectual property answer the following questions:

• Does it explain how something is working?

• Does it contain diagrams?

• Does it contain screenshots and explanations?

If you answer no to all of these questions then do not worry about the press release. This also works for brochures. A public technical presentation is a much more dangerous exercise for two reasons:

• People attend a presentation to learn something

• The objective of the presentation is to show that the company is skilled and experienced but to achieve this objective the speaker needs to have some matter. Less the speaker is experienced in the difficult art of communication more she needs to disclose and she is frequently inexperienced. This is another illustration of the type 3 work of Society and Computing. Press release writing and brochureware are comfortable jobs. There is no shortage of readers, suggestions and enhancements for these documents. Presentation is hard work implying a strong involvement of the speaker. Therefore technical presentations are frequently delegated to junior programmers who are not even given the time to properly prepare their performance.

This is difficult to not disclose intellectual property. In an ideal world the company would define what it is ready to disclose and record what and to whom that it has disclosed. Things made public still have a value. We can illustrate it with a simple example. Your competitor patents a method. If you can show that you owned the method before, you still can use the method but this is bad for your business and for your customers. If you can prove that you presented the method to the public the patent will be rejected.

Patenting has the merit of forcing people to think about these issues. Most of the time what we call secrecy should be called "not care". Open Sourcing your development implies taking care to not infringe copyright of others. Publishing requires extra work. Secret is definitely the cheapest option. For instance it makes sense to keep internal developments for internal users secret. Even the most dedicated patent portfolio makers realize that not everything can be patented: there is always novelty in a large project but this novelty may have a so narrow scope that this is not worth to protect it. Large firms use all methods:

• Patent portfolios

• Open Source

• Publication

• Secrecy

The most obvious example is IBM, which has the biggest patent portfolio and is also one of the biggest Open Source publishers. Companies like BEA, Cisco and Oracle publish large amount of content on their Web sites. Even a small company probably has to make some things public in a conscious way.

Intellectual Property in is obviously related to competitive analysis and market intelligence. However whereas the latter aims to answer the questions: Who are our competitors? How well are they doing? What are their plans? Intellectual Property in aims to answer a single question: Are we allowed to do that?

In some way competitive analysis and market intelligence are the other side of Intellectual Property out and suffers the same problem. Marketing, sales forces and everybody who meet customers is doing market intelligence. A company collects a huge amount of market data but these data are disseminated. Furthermore it is difficult to measure the relative importance of these data, which are in different formats, more or less reliable and hard to link together. However despite their flaws competitive analysis and market intelligence are the first input of Intellectual Property in. They provide raw data such as competitor names, market and mind shares.

Understanding the "that" in "are we allowed to do that?" is another difficulty. Even if you do not plan to patent you must follow the principle exposed above in Patent writing: the product design is the other input of Intellectual Property in. You use this input to write a reference text, which is a patent subset:

• The field of the product

• Background information: what people will need to understand the product

• Prior art: Problems faced in the product area and existing solutions. Briefly describe how these solutions are working and the problems that they do not solve.

• Summary of the product design: How the product solves one or several of these problems

Use the reference text to make searches in the patent databases with the toolkit presented above. You can use the patent class of the found patents to broaden the search scope. Look at the referenced patents. Look at the patents referencing the patents. At this stage you may have found forty or more patents. Then consolidate your results with the list of competitors identified by the competitive analysis and market intelligence. A direct competitor will go to court with a case weaker than a company that is on a different market or that is also your provider, customer or partner. It has also more chance to be granted an injunction, in which case you will have to stop selling or using your product. On the other hand patent search may identify new entrants on your market. In that respect Intellectual Property in also feeds competitive analysis and market intelligence. You must also consider patent families: a patent filed in a country may have been promoted to other countries through PCT, which implies a stronger commitment of the patentee. Such patent may represent a stronger threat for your product.

Next compare the independent claims of each remaining patent with the product design. If you find a match between an independent claim and a part of the design ask for a redesign. You should also send a document presenting the most threatening patents and claims to analysts and programmers. Then throughout the project you should periodically check for new patent applications. This is a tough and time-consuming job. If you frequently file patents and if you never stop being sued (which means that you make money and have an enviable position on your market) you represent a sizeable and regular source of revenue for your lawyers and it makes sense to ask their help. A software patent attorney usually knows something about software, which is not always the case of legal departments. On the other hand we do not recommend hiring a patent attorney only to assess legal risks related to a particular project:

• The attorney must understand the activity of the company and know its market and competitors

• The attorney must check for new patent applications throughout the project development

• This is a time-consuming task that highly depends on the project. An assessment may require ten to hundred and more hours of research. Such assessment cannot be cheap. Furthermore the customer has no means to check how extensive and thorough the research was.

Therefore it can make sense to do the search yourself. This is a senior programmer or designer job that requires a holistic view of the company and market. This document may help starting in this activity. Public databases and our toolkit are probably enough to identify threats and find corrective actions. You can also use http://www.freepatentsonline.com/ as an alternative to the USPTO web site. freepatentsonline allows downloading patent files in PDF format, which may be preferred notably for printing. Today freepatentsonline only lists granted patents. In the future they should also list patent applications. You may need to check other sources of information. The Patent Information Users Group site may be a good starting point: "This Patent Information Users Group, Inc. (PIUG), the International Society for Patent Information is a not-for-profit organization for individuals having a professional, scientific or technical interest in patent information." This site contains a good FAQ. You can also use commercial databases and tools. You can find a list of vendors at http://www.piug.org/vendor.html. There is a good presentation of patent tools, Software Tools for Analyzing Patents by Anthony Trippe. The most popular commercial sites may be:

• qpat

• micropatent

• IBM patent server

• DerWent patent explorer

• http://www.ip.com

Corporate subscriptions can be expensive but these sites frequently offer pay-per-use service (vouchers, one day subscription) at a much lower price.

You can be in three situations:

1. The current design or a feature of the product literally infringes the claims of a patent or a patent application, which is likely to be granted with its current scope. In this case the project management has to (1) either modify the product to no longer infringe the patent claims (2) cancel the project (3) or negotiate a license agreement with the patentee.

2. You identified patents and patent applications belonging to the product domain and filed by companies that compete or may compete with your company. You found that none of these patents and patent application claims the product design and features. Then the project development and releasing can go on.

3. You are in the twilight zone. The Federal Circuit explained that "'willfulness' in infringement, as in life, is not an all-or-nothing trait, but one of degree. The law recognizes that infringement may range from unknowing, or accidental, to deliberate, or reckless, disregard of a patentee's legal rights. [...] The term 'willfulness' thus reflects a threshold of culpability in the act of infringement. [...] The law of willful infringement does not search for minimally tolerable behavior, but requires prudent, and ethical, legal and commercial actions. Thus precedent displays the consistent theme of whether a prudent person would have had sound reason to believe that the patent was not infringed or was invalid or unenforceable, and would be so held if litigated." You may find that the product implements neither the limitations of the patent claims nor any reasonable equivalent to these claims’ limitations. You may also find prior art that make the patent not novel or obvious and therefore invalid. But you recognize that the patentee and a court can find that the product infringes the patent. In this case ask for an opinion of counsel as explained in Software and business method patents. If the opinion is positive go on with the project development and releasing as in 2. Otherwise proceed as exposed in 1.

You have to focus on competitors’ patents and patent applications because:

1. Checking all relevant patents and patent applications would require an unreasonable amount of work.

2. A patentee who is not a competitor does not sell less because of the infringement and a court is unlikely to issue an injunction forbidding you to sell or use the product. The Federal Circuit has "repeatedly endorsed the conceptual framework of a hypothetical negotiation between patentee and infringer as a mean for determining a reasonable royalty" to determine what the infringer has to pay to the patent holder. So you are less exposed and the damage awarded to the patentee is related (though usually much higher) to the license fee you would have paid if you had negotiated with the patentee.

3. You are supposed to know what your competitors are doing and the patents they own. This due diligence does not extend to other businesses.

The net result is that you will be sometimes notified by a patentee about a patent infringement because:

1. The patent is in the twilight zone.

2. You gave a relevant patent a miss because this patent was filed by a company or individual not known in your business.

3. You simply did not find a relevant patent. The search process is not 100% reliable.

There are several reasons to not go to court:

1. litigations are expensive and time-consuming; if you are in a profitable and growing business it is usually better to spend your time on your business rather than on litigations;

2. courts presume that patents are valid and the burden of proof is on the defendant (you);

3. in USA patent infringement cases are generally ruled by juries (choice of the plaintiff - the patentee) and the experience shows that juries’ decisions and awarded damages are hard to predict.

Uncertainty and litigation costs are also issues for the patentee. And if he is awarded damages these damages will be awarded after a couple of years. So the interest of the patentee is also to negotiate a license agreement.

There are two factors to consider when you prepare the negotiation:

1. The value of the patent process or method for your business. I discuss this issue in the Damage background section of Software and business method patents. You have to answer three questions (1) can you design around the patent and at which cost? (2) what is the cost for you and your customers if you disable the functions that allegedly infringe the patent? (3) how much money are you making thank to the patent process or method?

2. The likeliness of patentee success in court and the likeliness of patent invalidation in a patent re-examination. Here you have to identify prior art that was not considered or improperly presented in the patent examination.

A reader informed me about a Settlement help site containing a "large Directory of Settlements Help Sites, financial websites and general sites of interest". An article of this site entitled "How $50 Million Settlement Was Reached – Hours Before Trial Began- Esquire Litigation Solutions Seen As Key In Forcing Settlement" shows the way you should present your data. In the negotiation you must demonstrate that:

1. the patent could be found not novel or obvious and invalidated;

2. you can design-around the patent or live without using the protected system or method.

Patents Presentation Search Issues Strategies Business Methods Patentability Analysis MercExchange eBay Trial Reexamination Business view Granted patents Examination USPTO EPO PanIP Eolas 1-click family 1-click analysis 1-click prior art Trademark Copyright

Contact:support@pagebox.net
©2001-2005 Alexis Grandemange   Last modified