It depends. (Doesn’t it always?)
IANA lawyer, but I’m an engineer who’s had to deal with patent issues a few times. Here’s my experience in the matter:
If a product (or process) is patented, another company may not be able to reproduce verbatim without paying license fees. If the originator decides not to patent the product or process, then nobody else gets to make it until the patent expires. For utility patents filed in the US, expiration is 20 years from the initial filing date, although some companies, notably pharmeucitical manufacturers, have managed to extend this by making a new product that is a functional extention of the original and causing all formulations to fall under the new patent. Other types of patents (ornamental, biological, utility, software, et cetera) have different limitations and functions. See your local patent lawyer for more detail.
Processes and products that are identical to patented ones still fall under the control of the patent-holder, even if they were developed independantly. You can choose to challenge a patent or the applicability and, short of a legal injunction, continue producing the product in question, but you might be liable for retroactive license fees (which could be in excess of profits) in the case that your challenge fails. Typical challenges include that the patent is too general (doesn’t cover the specifics of the challenged product), too specific (product is manifestly different in some way from the patent description), that the product or process is already in the public domain (was “generally known” prior to issuance of the patent) or that the challenger invented the process prior to the patent-holder (in the US), that the patent results in gross monopolization of an entire industry, and so forth. It’s exceedling rare that patents are overturned, but patent-holders are not infrequently required to license their patents out to other manufacturers.
There are also trade secrets, which is “inside information” on a process that isn’t patented (generally to prevent a competitor from adopting or improving upon a process) and proprietary information (specifications or data that are held by employees of a company who are legally bound not to disclose marked information). While it is not generally illegal to utilize trade secrets or proprietary information, the methods for acquiring said information are often unethical if not downright illegal. For instance, here’s an example of the use of proprietary information of Lockheed Martin Corporation by The Boeing Company for use in a proposal in which the two companies were competing against each other. Here’s another example of Boeing using proprietary Raytheon information. (I’m not deliberately trying to pick on the defense sector here; it’s just that with all of the cooperation with potential competitors and exchange of engineering and executive talent between major players this sort of thing has the potential to occur, and it’s actually surprising how rarely it does, which is a tribute to how much the industry has cleaned up at least this part of its act since the Eighties.) In this case, the case being contested isn’t that the proposed designs are too similar, but that information was acquired through processes that are illegal.
In general there is nothing preventing you from starting with a product and reproducing its qualities without specific knowledge of how it functions; and it’s not illegal to buy a product, tear it down, and figure out how it works. If you can make a competitive product that works on the same fundamental process but does it with sufficient distintion to the original product then you can evade a patent restriction, though this is a very grey area that depends upon how general or expansive the patent is. (Often, companies or individuals will attempt to patent a number of similar processes, even ones they can’t make work or minor variations on the original, in order to prevent a competitor from making a similar product. In short, it’s a lot of legal manipulation and tiptoeing to outwit your competitors. I find the whole field to be rather tedious and irritating–having to sort through patents, argue with other people what falls under a patent and what doesn’t, figure out how you can somehow twist the language to make your imitation seem unique–but on the plus side it does offer the ability of the small patent-holder to sell or license a design to an industry giant who can best implement the product without having it stolen out from under them. Well, sometimes, anyway.
Richard Feynman allegdly sold his patent rights for a nuclear-powered airplane, rocket, and electrical powerplant to the government (Department of Energy? Department of War? He doesn’t specify) for one dollar each, and made a stink about getting his dollar, which he spent on candy. (“Surely You’re Joking, Mr. Feynman”, pgs 162-5.)
So…reverse engineering itself isn’t illegal, but it might be illegal, or at least unethical, to produce a product or base a proposal upon the information you’ve obtained. There may be exceptions to this, such as breaking or distributing the method for an encryption algorithm for the apparent purpose of defeating copyprotection (DeCSS) but in general, knowledge by itself isn’t illegal.
Stranger