CNET también está disponible en español.

Ir a español

Don't show this again

HolidayBuyer's Guide

Proprietary code--what's really at suit

In the case of "open source" versus "proprietary" versus "patented" versus "copyrighted" code, there are some other aspects that have not been discussed much in the press.

In response to the June 27 Perspectives column by Charles Cooper, "Putting open source on trial":

Your points on intellectual property are well-taken and pithy, but in the case of "open source" versus "proprietary" versus "patented" versus "copyrighted" code, there are some other aspects that have not been discussed much in the press.

The first is the nature of a patent as an agreement between society and an inventor: In exchange for the inventor disclosing the invention, society grants him the exclusive right to profit from the invention for a limited period of time.

Copyrights are used by an author--or programmer--to register a particular work. Although their purpose is to prevent the copying of the work or the code, they do not protect the invention.

Proprietary code is not disclosed--in order to protect it. There is no legal protection for proprietary code except to the degree that theft is illegal. In other words, the code is not protected, but its theft is prohibited. If you show proprietary code to a second party, you must have it sign a non-disclosure agreement. You must protect it through civil agreements.

This is a bit of an over-simplification, but if SCO is claiming patent infringement, I have seen no claims of any particular patent that has been violated. If it is claiming copyright infringement, then I have seen no evidence that there is a public record of the code it has copyrighted.

I believe that it is claiming the theft of proprietary code. In order to do this, SCO must show that:

• The code originated with SCO, which would mean that it did not copy the code from a public source. Simply showing that the code is the same as some other code is not sufficient.

• SCO took great care to protect the code.

• The code was then stolen and used elsewhere. It may be very difficult to truly establish when the code was written or by whom, without some public record of it existing at the patents and copyright office or as a public source on the Web. How does SCO intend to show that the code was not copied backwards from a public source?

Without the protection of a patent or a copyright, the long-term effects of the suit may be quite negligible. IBM may be forced to pay SCO large sums of money, and the open-source community may be forced to rewrite the portions that are claimed to be stolen. But, since the algorithms are not protected--only the explicit code is--this job may be quite easy.

It is big mistake to penalize the open-source community for making code freely available and to reward the "commercial" sector for secrecy. By disclosing source code, the open-source community opens itself up to both legitimate and frivolous attacks. This is an area where legal rights are still being forged, but it seems clear to me that statutes of limitation should be short, and that the requirements on those who claim proprietary code has been stolen should be quite strict.

Society should not invest a lot of money and time in protecting code, writings, inventions--or anything else that has value only as long as it is kept secret and for which secrecy is an impediment to commerce and progress. All laws, like the patent law, should encourage disclosure in exchange for limited exclusive rights.

Otto Smith
Port Townsend, Wash.