# Intellectual-Property Patents Are Evil Should you attempt to patent "intellectual property," such as a computer algorithm? Hell no. Obtaining a patent is easy. I obtained a few before realizing the consequences. You write up any idea you are moderately proud of and send it to your company's patent lawyer. One or two pages are enough. The lawyer translates into ugly and inaccurate syntax that few of your peers could understand. Don't bother to proofread. It won't matter to anyone. The lawyer will handle the submission. The patent office rejects the application once to discourage amateurs. They make random objections that demonstrate they didn't understand a word. They ask how your work is different from other patents with similar keywords in the title. You point out that the cited patents have nothing to do with your topic. The lawyer fixes the rest and resubmits. The revised patent is approved without a hitch. At no time during the application does anyone oblige you to demonstrate that your method is not common practice or obvious to someone in the field. At no time is your patent read by someone who understands it. The office performs no literature search: they simply scan for keywords in previous patents. Your employer, who actually owns the patent, now has the right to drag smaller companies into court and force them to demonstrate that their practices do not fall under the patent. The burden of proof is on the defender, although the owner of the patent proved nothing to obtain the patent. If you leave the company, you'll have to license your own idea. Why would you participate in such injustice? For a negligible bonus? Some say that patents must be sought to defend against other patents. Extra patents won't stop lawsuits, though. To defend yourself, why not simply publish the idea? Publishing requires more work, but at least you can continue to use your own ideas. If you need an easier way to protect yourself, keep a dated lab notebook, and document your debt to obvious and well-known ideas. A patent application must be signed by "the inventor." If you sign, you swear that the idea is your original work and is not obvious to someone practiced in the art. You swear that all prior art has been referenced. If you know those statements to be false, then you should decline to commit fraud. A lawyer or manager cannot write and sign the patent application for you. Patents are one way corporations attempt to own ideas. Once, "knowledge workers" could keep their mental tools when they changed jobs. Employers do not patent the skills of manual workers because those workers depend on expensive mechanical tools. New rules were invented to keep the rest of us equally dependent. "Intellectual property" was invented to own what is in your brain -- to own your skills and your education. Don't make it easier for them. Don't be exploited by your own vanity. Intellectual property cannot be both. Bill Harlan, 1998 (Footnote: patents on which I was primary author are 5,062,086; 5,067,113; and 5,089,994) Postscript: Intellectual property patents are increasingly impossible to avoid, as a condition of maintaining employment. We know how easy it is to create unethical patents, fraudulently taking credit for obvious applications of general domain knowledge. Here are some guidelines for writing a patent and avoiding the contempt of professional peers. __ o The patent must be non-obvious. You can seriously argue that no one else is likely to have thought of this idea. o The idea might have remained a trade secret if not for the patent. This is the original purpose of patents: new technology is disclosed for a temporary monopoly, for the greater good. o Ensure that the claims are not unreasonably broad in the hands of your patent attorney. If you are patenting something specific, keep it specific. o Still meets the basic requirements of a patent: novel (no known prior art); useful (not speculative); and non-obvious (to someone else practiced in the art). _ Remember, if you sign a patent, you must be able to defend it as the inventor. Writing a patent that meets this standard is a lot more work than the unethical kind, but that is the point. Bill Harlan, 2014