Talk:Diamond v. Diehr

What was patented here?
Did this case rule that the machine was patentable or the program patentable? It seems ambiguous. This link is Broken 15:16, 10 September 2005 (UTC) "Diamond v. Diehr, 450 U.S. 175 (1981)[1], was a U.S. Supreme Court decision which held that a machine controlled by a computer program was patentable. This decision did not make a computer program, by itself, patentable." I'm really not sure how much clearer the point can be made. Mmmbeer 17:00, 10 September 2005 (UTC)
 * It's answered in the VERY first two sentences. I'm really not sure what else you want it to say:
 * it says "did not make a computer program, by itself, patentable" but was the program patented when you patented it with the machine. i.e. Could I have the same machine as someone else but with a different program (so the program isn't by itself) patented. Or does the machine also need to be novel. This link is Broken 21:48, 10 September 2005 (UTC)
 * The patent was for a machine controlled by a computer program. The question answered by the court is not whether the machine by itself or the program by itself was patentable, but rather, whether the combination was patentable.  So the rule from the case is exactly as it is in the intro, and as I pasted. Law is very particular about language. Mmmbeer 19:36, 11 September 2005 (UTC)
 * The case is significant because, before this ruling, the mere presence of a computer in an otherwise patentable invention could make the invention unpatentable. Even though this decision didn't specifically allow for software patents, it established the general concept that processes involving computers can be patented as long as they produce a useful result (and meet the other criteria for patentable subject matter). It was subsequently argued that, since "non-mathematical algorithms" are step-by-step processes, they should be subject to patent protection if they produce a useful result. Bryan 15:39, 27 February 2006 (UTC)


 * Can you point to some case in which presence of a computer caused an otherwise patentable thing to be unpatentable? Looking at the text of the decision, it seems to me that the patent examiner objected to patenting the automation of a well-known formula.  The Supreme Court disagreed, stating that the object of the patent wasn't the formula, but a new process for molding rubber.  It didn't explicitly allow software patents, but allowed a semantic end-run to achieve the same result.  Am I right?  Frotz 09:13, 11 March 2007 (UTC)


 * The majority in this case did not see the patent as a semantic end-run. They saw claimed novelty in non-software elements and in the way software/non-software elements are combined.  This is what distinguished the case from Flook, where the only claimed novelty/non-obviousness was in software.


 * I think it depends upon what you mean by "the same result". As I read it, the Supreme Court said that, in a process that is otherwise patentable subject matter, you can't disqualify it simply because you are using a computer to carry out a portion of the process.  The Supreme Court also said that the novelty of the process could reside in the method steps carried out on the computer.  In that sense it does allow for software patents subject to the constraint that the claimed process produces a useful result.  But also in that sense, "software patents" have always been allowed under US patent law in the sense that method patents have always been allowed.  That includes method patents were at least some of the steps are performed by a machine.  Does that address the question?--Nowa 18:50, 11 March 2007 (UTC)


 * What has me confused is that the formula used to determine when to open the press has been known to chemists for more than a hundred years. The decision states that the main problem in using the formula was getting an accurate temperate reading.  It further states that the solution was to constantly check the temperature, which in the previous sentence was said to be impossible.  It seems to me that the only new thing was to figure out how to accurately check the temperature, but the means for this are not mentioned.  Given this, the patent should not have mentioned computers at all, but instead focused on the method used for determining the temperature inside the mold.  Frotz 22:35, 11 March 2007 (UTC)

The means are not mentioned in the case, but they are in the patent:


 * The actual mold temperature is fed to the computer on a substantially continuous basis, for example, every ten seconds. Thermocouples, or other temperature-detecting devices, located directly within the mold cavity may read the temperature at the surface where the molding compound touches the mold, so that it actually gets the temperature of the material at that surface. The computer then performs series of integrations to calculate from the series of temperature readings and from the time-temperature cure data a proper cure time and to compare that cure time with the elapsed time. Recalculation continues until the time that has elapsed since mold closure corresponds with the calculated time. Then, the computer actuates the mold-opening device and the mold is automatically opened.--Nowa 01:30, 12 March 2007 (UTC)

That is probably what the invention really was. But it should have been recited in the claim. Because it was not recited, we have Breyer in Mayo saying there "must" have been some invention in there that the Diehr Court perceived and we have Thomas saying in Alice that the claim was patent eligible because it was directed to a technological procces--molding rubber. PraeceptorIP (talk) 01:50, 18 July 2015 (UTC)

Analytic dissection
The article editorializes when it assumes criticisms of analytic dissection in Diehr are criticisms of the Flook opinion. Diehr does not say this. Further, as I read Flook, it explicitly does not use analytic dissection.


 * Reply to unsigned comment: I am sorry but that misreads Diehr and Flook. Flook makes a point of saying that it is conceded that everything is old or conventional in the claim except the use of the smoothing equation. Then it says that under Morse and Nielson you must presume that the Equation, the only novel element, is in the prior art. There is a flaw in that statement that is picked up and remedied, I think, by Mayo and Alice. Stevens should have added that there is no suggestion that the (novel) equation element cooperates with the other claim elements in a novel and unobvious way. Recent decisions of the Fed. Cir. usually add that. PraeceptorIP (talk) 01:39, 18 July 2015 (UTC)

Bilski test?
The article right included a discussion of whether the patent would pass the Bilski test. Per the commentary, since THE DRAWINGS only show an algorithm, this would fail the Bilski test. Never mind that the abstract talks about transformation of rubber products and the independent claims all relate to rubber curing. Am I missing something? If this doesn't pass the Bilski test then what does? —Preceding unsigned comment added by 75.34.19.80 (talk) 15:56, 18 June 2009 (UTC)


 * Good point. It probably does not pass Bilski, Mayo, or Alice, because of the way it was claimed. BTW, the abstract and spec should not be read into the claim. PraeceptorIP (talk) 01:50, 18 July 2015 (UTC)

Reference to Microsoft case in See also
Research Corp. Technologies v. Microsoft Corp., No. 2010-1037 (Fed. Cir., Dec. 8, 2010). This case is not concerned with Diehr issues. It should be deleted. I will delete it after a reasonable time if there is no objection. PraeceptorIP (talk) 01:30, 18 July 2015 (UTC)
 * I think it's OK to delete it now since it doesn't link to an article.--Nowa (talk) 05:52, 18 July 2015 (UTC)

Not a "computer program"
Often, Diehr is believed to have introduced software patents. But it actually only relates to a conventional system that performs a calculation on an observed quantity: to calculate a time from temperature measurements. The formula of the calculation is not new (it stems from the 19th century), neither is the method to perform the calculation (the algorithm). Rbakels (talk) 19:27, 12 March 2024 (UTC)