Received: from SOUTH-STATION-ANNEX.MIT.EDU by po7.MIT.EDU (5.61/4.7) id AA25140; Fri, 19 May 95 10:47:52 EDT Received: from zork.tiac.net by MIT.EDU with SMTP id AA29940; Fri, 19 May 95 10:47:50 EDT Received: from arctos.com (arctos.com [199.3.128.65]) by zork.tiac.net (8.6.9/8.6.6.Beta9) with SMTP id KAA04917 for ; Fri, 19 May 1995 10:47:47 -0400 Message-Id: <199505191447.KAA04917@zork.tiac.net> X-Sender: oldbear@tiac.net X-Mailer: Windows Eudora Version 1.4.4 Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Date: Fri, 19 May 1995 10:44:36 +0000 To: rdshydur@MIT.EDU From: rah@shipwright.com (Robert Hettinga) (by way of oldbear@arctos.com (The Old Bear)) Subject: SOF_pat Richard: This was forwarded to me by a colleague and I thought that it would be of particular interest to you, the e-club, any other folks within your usual orbit. Cheers, Will oldbear@arctos.com ----------begin forwarded text---------- >From: John Young >Date: Sun, 14 May 1995 13:38:21 -0400 >To: rah@shipwright.com >Subject: SOF_pat > > >Responding to msg by rah@shipwright.com (Robert Hettinga) on >Sun, 14 May 7:23 AM > > > The New York Times > May 8, 1995, p. D2. > > > Patents [Column] > > > By Sabra Chartrand > > > Guidelines are being prepared to evaluate software, but > only if it is part of a floppy disk or a computer. > > > Washington. The Patent and Trademark Office is looking anew > at the way it judges software inventions. > > > The agency does not patent mathematical formulas, > scientific principals, laws of nature or mental processes. > And for a long time, computer software had been considered > too close to mental processes. Software was treated like > printed material that happened to be on a disk and not > paper -- and, as such, not patentable. > > > But now the agency says that rulings last year in the Court > of Appeals for the Federal Circuit, the court that hears > appeals of rejected patents, made it plain that software > should be patented if it is part of a product like a floppy > disk or computer. So for about a month the agency has been > working on new guidelines for evaluating software > inventions that it expects its examiners to be following by > this summer. > > > "Software is such an important aspect of the country's > economy today that we're taking the court's admonition," > Nancy Linck, the agency's senior attorney, said. "The real > change the guidelines will make is to provide protection > for software programming as an article of manufacture, as > in a floppy disk." > > > In other words, software may be patented, but not alone. > Computer code, by itself, is still too abstract to win a > patent. But software that causes computer hardware to > perform a specific function is not. > > > "The software program itself has to be embodied in a > tangible medium, like a floppy, that is useful, new and > non-obvious," Ms. Linck said. "The real hard question is > whether to limit it in some way to use. A mathematical > algorithm in the abstract, even if it's embedded in a disk, > may not be patentable." > > > The move to adopt new guidelines appears to fit neatly into > Commissioner Bruce Lehman's plans to reshape patent rules > and regulations to give American businesses a competitive > edge in worldwide markets for high-tech products. And Ms. > Linck said the agency's ideas had won support from the > software industry. > > > But as late as December, the agency was still resisting > software patents. It was in court then, defending its > decision to deny a patent for data structure in memory -- > a floppy disk, for example. The agency did not think the > invention qualified. > > > "The Patent and Trademark Office attempted to rqect the > claims as not patentable," recalled Barry N. Young, > an intellectual property lawyer from California who > represented the inventor's appeal. "They said 'all you have > is a memory with a data structure stored on it.' " > > > "They would not," he continued, "give patentable > significance to that because that would give weight to the > content, which would be analogous to patenting a book. We > argued that the claims did not cover the content of the > information, but specifically covered the data structure in > terms of the hierarchy of elements." > > > The court agreed, and overturned the patent office's > rejection of the patent. Ms. Linck said the ruling was a > benchmark for the agency. > > > The ruling came after another case last fall "reaffirmed > the proposal that when you put new software programming on > an old computer, you make it into a new machine and > therefore a patentable machine," Ms. Linck added. > > > The two rulings occurred as other court findings seemed to > restrict the kinds of copyright protection available to > software inventions. So just as it got harder to get a > copyright, winning a patent became easier. > > > Some question whether any software is eligible for a patent > just because it is embedded in a disk or a computer. > > > "Suppose you have information stored, something like an > encyclopedia-type data base," Mr. Young said. "It has a > flat structure; it's not hierarchical; it's only > information in categories. That kind of arrangement of data > is clearly not patentable." > > > The re-evaluation follows a couple of years of tumultuous > software debate at the Patent and Trademark Office. A > patent issued in 1993 to Compton's New Media created an > uproar from software developers who argued that the > technology was commonplace and complained that patent > examiners did not know enough about software. The patent > office agreed there was truth to that -- at the time of the > Compton application, a degree in computer programming did > not meet the agency's education requirements for examiners. > > > The firestorm compelled Commissioner Lehman to hold public > hearings on software-related inventions and to order a > re-examination of Compton's patent. The agency was deluged > with advice and research material from people in the > computer industry pointing the examiners toward proof that > Compton's technology was in the public domain. As a result, > the patent office voided Compton's patent. > > > The agency also started hiring computer programmers as > examiners, and began to expand its library of software > manuals, textbooks and reference materials. Then came the > court rulings that software could be patented. > > > Specifics of the new guidelines have yet to be decided, but > Ms. Linck said she expected they would at least make it > clear that software was patentable when part of a floppy > disk, a computer or as an element of a process. > > > Those are fairly broad requirements, and some skeptics have > pointed out that such software could include music and > literature on disk. But Ms. Linck said that at the very > least the guidelines would make it impossible to patent > songs or novels. > > > "Just how we will reject those claims isn't clear yet," she > added. Ms. Linck said that when the guidelines were ready, > they would be published in the Federal Register and the > Official Gazette of the Patent and Trademark Office so that > the public could comment on them. > > > [End]