What does it take to be "first to invent," and what does it take to prove it?
Subject: What's the meaning of "invent" in "first to invent" ? Date: Thu, 28 Jan 1999 22:09:36 GMT From: skubi@solsoft.fr Message-ID: <78qn6p$12r$1@nnrp1.dejanews.com> Newsgroups: misc.int-property Lines: 30 What's the meaning of "invent" in the expression "first to invent" ? Is it enough to have the idea, or I must have something more (like a description of the invented thing sufficient for an average man skilled in the art to build the thing) ? In other words: if I claim "a gizmo comprising a shpuntz and a frogifier", and I want to say (for the purpose of the American patent law, that is) that I invented this at time T, is it enough to prove that at time T I had a document containing - the sentence "it would be a good idea to build a gizmo comprising a shpuntz and a frogifier" - or the above sentence plus a description of the gizmo (assuming that there are many ways of arranging the shpuntz and the frogifier, and it is not obvious which arrangement gives useful results) - or the above sentence plus a description of why the gizmo is useful/how it should be used - or all of these Thank you in advance. BTW, I like this newsgroup very much. So far, I got very interesting and friendly advice. Thanks to all. -----------== Posted via Deja News, The Discussion Network ==---------- http://www.dejanews.com/ Search, Read, Discuss, or Start Your Own
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Thu, 28 Jan 1999 19:56:34 -0500 From: David Kiewit <dak@patent-faq.com> Message-ID: <36B10742.733E5D69@patent-faq.com> References: <78qn6p$12r$1@nnrp1.dejanews.com> Newsgroups: misc.int-property Lines: 54 skubi@solsoft.fr wrote: > > What's the meaning of "invent" in the expression "first to invent" ? I'm glad you asked. Most of the folks who hear the phrase make a totally wrong assumption about what the answer is. > > Is it enough to have the idea, or I must have something more > (like a description of the invented thing sufficient for an > average man skilled in the art to build the thing) ? Someone once said "Success has many fathers, failure is an orphan". Finding someone who "thought of that a long time and must have been ripped off by someone he talked to" is easy. The law is set up to keep casual poachers of that sort from using the courts to steal rights from the people who really did the work to make them a success. > In other words: if I claim "a gizmo comprising a shpuntz and a frogifier", > and I want to say (for the purpose of the American patent law, that is) > that I invented this at time T, is it enough to prove that at time T > I had a document containing > > - the sentence "it would be a good idea to build a gizmo comprising > a shpuntz and a frogifier" No. The law requires that one be continually diligent during the development period. If the idea is put on the shelf for a while and then resurrected, the "earliest date" moves up to the time when the project started again. Many individual inventors document conception. Few document progress. Companies with reasonably well managed engineering staffs document progress up one side and down the other ("How many hours did you charge to the shpuntz project during the last week of September 1996?" is a pretty easy question for an employed engineer to answer. Individual US inventors almost always play the game according to "first to file" rules in a "first to invent" system. Net, in a priority dispute, the little guy is the designated loser. But most of the inventors' groups are the most virulent defenders of first to invent -- a droll situation. Best regards Dave -- David A. Kiewit (727) 866-0669 5901 3rd St.South, St. Petersburg, FL 33705 Registered Patent Agent http://patent-faq.com
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Fri, 29 Jan 1999 03:46:55 GMT From: jptucson@azstarnet.com (John Pederson) Message-ID: <36b12f0c.1915462@news.azstarnet.com> References: <36B10742.733E5D69@patent-faq.com> Newsgroups: misc.int-property Lines: 53 On Thu, 28 Jan 1999 19:56:34 -0500, David Kiewit <dak@patent-faq.com> wrote: > > >skubi@solsoft.fr wrote: >> >> What's the meaning of "invent" in the expression "first to invent" ? > >I'm glad you asked. Most of the folks who hear the phrase make a totally >wrong assumption about what the answer is. >> >> Is it enough to have the idea, or I must have something more >> (like a description of the invented thing sufficient for an >> average man skilled in the art to build the thing) ? > >Someone once said "Success has many fathers, failure is an orphan". >Finding someone who "thought of that a long time and must have been >ripped off by someone he talked to" is easy. The law is set up to keep >casual poachers of that sort from using the courts to steal rights from >the people who really did the work to make them a success. > >> In other words: if I claim "a gizmo comprising a shpuntz and a frogifier", >> and I want to say (for the purpose of the American patent law, that is) >> that I invented this at time T, is it enough to prove that at time T >> I had a document containing >> >> - the sentence "it would be a good idea to build a gizmo comprising >> a shpuntz and a frogifier" > >No. The law requires that one be continually diligent during the >development period. If the idea is put on the shelf for a while and then >resurrected, the "earliest date" moves up to the time when the project >started again. > >Many individual inventors document conception. Few document progress. >Companies with reasonably well managed engineering staffs document >progress up one side and down the other ("How many hours did you charge >to the shpuntz project during the last week of September 1996?" is a >pretty easy question for an employed engineer to answer. > >Individual US inventors almost always play the game according to "first >to file" rules in a "first to invent" system. Net, in a priority >dispute, the little guy is the designated loser. But most of the >inventors' groups are the most virulent defenders of first to invent -- >a droll situation. Amen. Beautifully put. And so true. John Pederson, Retired Former Director of Patents Major Manufacturing Company
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Fri, 29 Jan 99 15:29:52 GMT From: taquilla@erols.com (Tracy Aquilla) Message-ID: <taquilla.1268011432B@news.erols.com> References: <36B10742.733E5D69@patent-faq.com> Newsgroups: misc.int-property Lines: 89 In Article <36B10742.733E5D69@patent-faq.com>, David Kiewit <dak@patent-faq.com> wrote: >skubi@solsoft.fr wrote: >> >> What's the meaning of "invent" in the expression "first to invent" ? Since this is a legal term of art, it is not something that is susceptible to a one or two line definition. Generally, as a starting point, "invent" in this context means first to conceive, however the first to conceive can lose a priority contest if she is not first to reduce to practice and is not diligent (see below). Conception, reduction to practice, and diligence are also terms of art that have a specific legal meaning. >I'm glad you asked. Most of the folks who hear the phrase make a totally >wrong assumption about what the answer is. Perhaps this is because the priority rules in the US system are a bit complicated? >> Is it enough to have the idea, or I must have something more >> (like a description of the invented thing sufficient for an >> average man skilled in the art to build the thing) ? > >Someone once said "Success has many fathers, failure is an orphan". >Finding someone who "thought of that a long time and must have been >ripped off by someone he talked to" is easy. The law is set up to keep >casual poachers of that sort from using the courts to steal rights from >the people who really did the work to make them a success. You mean there are actually people who would do things like that? ;-) >> In other words: if I claim "a gizmo comprising a shpuntz and a frogifier", >> and I want to say (for the purpose of the American patent law, that is) >> that I invented this at time T, is it enough to prove that at time T >> I had a document containing >> >> - the sentence "it would be a good idea to build a gizmo comprising >> a shpuntz and a frogifier" > >No. The law requires that one be continually diligent during the >development period. If the idea is put on the shelf for a while and then >resurrected, the "earliest date" moves up to the time when the project >started again. Basically true, however, diligence is relevant only when an inventor is first to conceive but second to reduce to practice (assuming only two-party interference). The inventor who is first to conceive and last to reduce to practice is the "first to invent" only if she has been diligent prior to the time the second inventor enters the field. Thus, the first to conceive may 'put it on the shelf', as long as her diligence is resumed before the second inventor's conception, and is continuous thereafter until the invention is reduced to practice. Since filing a US patent application constitutes constructive reduction to practice, in some cases, merely recording conception and filing an application may be enough to establish that one was the "first to invent". However, if the second to conceive can prove an earlier reduction to practice, diligence of the first to conceive becomes critical. Proving diligence can be very difficult without good records, and this has caused serious problems for many independent inventors (eg., Gordon Gould). >Many individual inventors document conception. Few document progress. Yes, this is a common problem. A second common problem for small/individual inventors is that every event must be independently corroborated by another person, and all records should be signed, dated and notarized. Small inventors and those who are new to the game sometimes keep poor records in this regard, even with respect to conception. >Companies with reasonably well managed engineering staffs document >progress up one side and down the other ("How many hours did you charge >to the shpuntz project during the last week of September 1996?" is a >pretty easy question for an employed engineer to answer. Yes, companies that are active in the patent field are generally aware of the rules and know how to preserve their priority. They also know how to copy claims and provoke interferences, and can generally afford highly competent counsel. >Individual US inventors almost always play the game according to "first >to file" rules in a "first to invent" system. Net, in a priority >dispute, the little guy is the designated loser. But most of the >inventors' groups are the most virulent defenders of first to invent -- >a droll situation. Yes, this is somewhat ironic. Tracy (Note that although it might sound like I think I know what I am talking about here, this post does not constitute legal advice.)
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Fri, 29 Jan 1999 14:14:35 -0500 From: David Kiewit <dak@patent-faq.com> Message-ID: <36B2089A.6699AD4@patent-faq.com> References: <taquilla.1268011432B@news.erols.com> Newsgroups: misc.int-property Lines: 41 > > In Article <36B10742.733E5D69@patent-faq.com>, David Kiewit > <dak@patent-faq.com> wrote: <big snip> > >Many individual inventors document conception. Few document progress. > and Tracy Aquilla answered > Yes, this is a common problem. A second common problem for small/individual > inventors is that every event must be independently corroborated by another > person, and all records should be signed, dated and notarized. Small > inventors and those who are new to the game sometimes keep poor records in > this regard, even with respect to conception. Another game plan is to forget about these nearly ritualistic documentation practices and get on with completing the invention. File first, and let your prospective opponent get further behind by chasing autographs. Having to live with your filing date as your date of invention can be unpleasant -- if you're not first to file, you lose. But interferences are rare, so most of the time (98+%??) documenting diligence is a waste of time. Additionally, the cost of seeing an interference through to the conclusion is commonly $100K+, with the initial preliminary motion stage commonly hitting $10K or so. Someplace out there, there's a very disaffected inventor who spent a lot of time documenting diligence and then threw all the paper out rather than sustain interference costs. The alternative isn't for everyone. I, for one, have used it on my own inventions, where documenting diligence would have been a hassle. Best regards Dave -- David A. Kiewit (727) 866-0669 5901 3rd St.South, St. Petersburg, FL 33705 Registered Patent Agent http://patent-faq.com
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Fri, 29 Jan 1999 21:38:15 GMT From: jptucson@azstarnet.com (John Pederson) Message-ID: <36b225da.1210867@news.azstarnet.com> References: <36B2089A.6699AD4@patent-faq.com> Newsgroups: misc.int-property Lines: 59 On Fri, 29 Jan 1999 14:14:35 -0500, David Kiewit <dak@patent-faq.com> wrote: > > > >> >> In Article <36B10742.733E5D69@patent-faq.com>, David Kiewit >> <dak@patent-faq.com> wrote: ><big snip> >> >Many individual inventors document conception. Few document progress. >> >and Tracy Aquilla answered > >> Yes, this is a common problem. A second common problem for small/individual >> inventors is that every event must be independently corroborated by another >> person, and all records should be signed, dated and notarized. Small >> inventors and those who are new to the game sometimes keep poor records in >> this regard, even with respect to conception. > >Another game plan is to forget about these nearly ritualistic >documentation practices and get on with completing the invention. File >first, and let your prospective opponent get further behind by chasing >autographs. Yes!! > >Having to live with your filing date as your date of invention can be >unpleasant -- if you're not first to file, you lose. But interferences >are rare, so most of the time (98+%??) documenting diligence is a waste >of time. Additionally, the cost of seeing an interference through to the >conclusion is commonly $100K+, with the initial preliminary motion stage >commonly hitting $10K or so. Someplace out there, there's a very >disaffected inventor who spent a lot of time documenting diligence and >then threw all the paper out rather than sustain interference costs. Yes!! To the best of my recollection, in the 43 years I spent as a patent professional and, ultimately, as Director of Patents with a major US manufacturer in a patent active industry, diligence or lack of diligence never made a difference in any interference or other case in which we were involved. Actually, during the last 20 years of that time, we decided to adopt exactly this game plan. And the time saved by our technical community by not documenting diligence was used instead to make additional inventions and/or otherwise improve productivity. Now under the new provisional application rules, I would file immediate automatic PPAs on all invention disclosures as received by my patent department, thus nailing down the earliest possible filing date for those on which we decided to file complete applications. This would give the best of all worlds, for only $150 per PPA. John Pederson, Retired Former Director of Patents Major Manufacturing Company
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Sun, 31 Jan 1999 08:54:24 -0700 From: Bruce Hayden <bhayden@ieee.org> Message-ID: <36B47CB0.DDB3434D@ieee.org> References: <36b225da.1210867@news.azstarnet.com> Newsgroups: misc.int-property Lines: 33 John Pederson wrote: > > > >Having to live with your filing date as your date of invention can be > >unpleasant -- if you're not first to file, you lose. But interferences > >are rare, so most of the time (98+%??) documenting diligence is a waste > >of time. Additionally, the cost of seeing an interference through to the > >conclusion is commonly $100K+, with the initial preliminary motion stage > >commonly hitting $10K or so. Someplace out there, there's a very > >disaffected inventor who spent a lot of time documenting diligence and > >then threw all the paper out rather than sustain interference costs. > > Yes!! > > To the best of my recollection, in the 43 years I spent as a patent > professional and, ultimately, as Director of Patents with a major US > manufacturer in a patent active industry, diligence or lack of > diligence never made a difference in any interference or other case in > which we were involved. The times I have seen diligence make any difference whatsoever is as a junior party in an interference - which actually reinforces your point of filing early, and hopefully becoming the senior party if an interference is called. -- -------------------------------------------------------------------- The preceding was not a legal opinion, and is not my employer's. Original portions Copyright 1999 Bruce E. Hayden,all rights reserved My work may be copied in whole or part, with proper attribution, as long as the copying is not for commercial gain. -------------------------------------------------------------------- Bruce E. Hayden bhayden@acm.org Phoenix, Arizona bhayden@ieee.org bhayden@copatlaw.com
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Sun, 31 Jan 1999 17:07:08 GMT From: jptucson@azstarnet.com (John Pederson) Message-ID: <36b5887c.4677942@news.azstarnet.com> References: <36B47CB0.DDB3434D@ieee.org> Newsgroups: misc.int-property Lines: 50 On Sun, 31 Jan 1999 08:54:24 -0700, Bruce Hayden <bhayden@ieee.org> wrote: >John Pederson wrote: >> > >> >Having to live with your filing date as your date of invention can be >> >unpleasant -- if you're not first to file, you lose. But interferences >> >are rare, so most of the time (98+%??) documenting diligence is a waste >> >of time. Additionally, the cost of seeing an interference through to the >> >conclusion is commonly $100K+, with the initial preliminary motion stage >> >commonly hitting $10K or so. Someplace out there, there's a very >> >disaffected inventor who spent a lot of time documenting diligence and >> >then threw all the paper out rather than sustain interference costs. >> >> Yes!! >> >> To the best of my recollection, in the 43 years I spent as a patent >> professional and, ultimately, as Director of Patents with a major US >> manufacturer in a patent active industry, diligence or lack of >> diligence never made a difference in any interference or other case in >> which we were involved. > >The times I have seen diligence make any difference whatsoever is >as a junior party in an interference - which actually reinforces >your point of filing early, and hopefully becoming the senior >party if an interference is called. Sure. My point was that on the few occasions when we were junior party, we could not have been saved by proof of diligence (no matter how complete) because it turned out that the senior party was first to conceive as well as being first to reduce to practice by virtue of being first to file. So even when we turned out to be junior party, our efforts at documenting diligence went for naught. Our batting average for all our efforts to maintain diligence records was .000 BTW, at least we saved the pain and expense of finding this out the hard way, because we worked out settlement agreements resolving the priority issue by informal exchange of proofs. Moral of the story -- being senior party gets you 99% of the way home in an interference. And, at least IMO, the remaining 1% chance is just not worth the time, pain and sweat of constant documentation by the technical community and policing of it by the patent professionals. John Pederson, Retired Former Director of Patents Major Manufacturing Company
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Sun, 31 Jan 1999 14:12:54 -0700 From: Bruce Hayden <bhayden@ieee.org> Message-ID: <36B4C756.A316C1F3@ieee.org> References: <36b5887c.4677942@news.azstarnet.com> Newsgroups: misc.int-property Lines: 47 John Pederson wrote: > > > >The times I have seen diligence make any difference whatsoever is > >as a junior party in an interference - which actually reinforces > >your point of filing early, and hopefully becoming the senior > >party if an interference is called. > > Sure. My point was that on the few occasions when we were junior > party, we could not have been saved by proof of diligence (no matter > how complete) because it turned out that the senior party was first to > conceive as well as being first to reduce to practice by virtue of > being first to file. So even when we turned out to be junior > party, our efforts at documenting diligence went for naught. Our > batting average for all our efforts to maintain diligence records was > .000 In my last employment, we had a couple where we appeared to have the earlier conception date, but destruction of records precluded us proving diligence (or probably more likely there was a gap). So, as junior party, we were forced to cave. > BTW, at least we saved the pain and expense of finding this out the > hard way, because we worked out settlement agreements resolving the > priority issue by informal exchange of proofs. Similar. > Moral of the story -- being senior party gets you 99% of the way home > in an interference. And, at least IMO, the remaining 1% chance is > just not worth the time, pain and sweat of constant documentation by > the technical community and policing of it by the patent > professionals. Obviously my "exception" proves your rule since if we had been senior party in any of those situations, it would have been up to the junior party to show diligence instead of us. -- -------------------------------------------------------------------- The preceding was not a legal opinion, and is not my employer's. Original portions Copyright 1999 Bruce E. Hayden,all rights reserved My work may be copied in whole or part, with proper attribution, as long as the copying is not for commercial gain. -------------------------------------------------------------------- Bruce E. Hayden bhayden@acm.org Phoenix, Arizona bhayden@ieee.org bhayden@copatlaw.com
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Sun, 31 Jan 1999 21:30:47 -0700 From: Ed Suominen <see-my@web-site.com> Message-ID: <36B52DF7.28DD59A@web-site.com> References: <36b225da.1210867@news.azstarnet.com> Newsgroups: misc.int-property Lines: 36 John Pederson wrote: > Now under the new provisional application rules, I would file > immediate automatic PPAs on all invention disclosures as received by > my patent department, thus nailing down the earliest possible filing > date for those on which we decided to file complete applications. > This would give the best of all worlds, for only $150 per PPA. > John, I respect your experience and history of well-written posts. I just want to make one cautionary note. A PPA is the world's most easily discoverable draft patent application. Inventors can write narrowing and unguarded statements in a PPA such as, "the other piece of code in Appendix B is probably obvious - everyone here thinks so," and "process A only works with blue widgets. It isn't suitable at all for use with red widgets." Do you really want to put such statements in the file wrapper with an "automatic" PPA? If you go through the disclosure to "clean it up," you probably will be spending enough time to justify just getting started on the claims and specification. Let's not forget the value of priority records for Rule 131 declarations, either. I suspect more patents are issued under Rule 131 than after interferences. Even in the unlikely event of an interference, you might wind up being more likely to win good patent protection by using priority records than by relying on a narrowly written provisional application with lots of legal holes in it. -- Nothing in this message is to be construed as legal advice, or the opinion of my firm or its clients. --------------------------------------------------- Ed Suominen, Patent Agent Squire, Sanders & Dempsey L.L.P. Web Site: http://eepatents.com
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Mon, 01 Feb 1999 06:21:04 -0500 From: David Kiewit <dak@patent-faq.com> Message-ID: <36B58E20.9396A5AF@patent-faq.com> References: <36B52DF7.28DD59A@web-site.com> Newsgroups: misc.int-property Lines: 21 Ed Suominen wrote: > > Let's not forget the value of priority records for Rule 131 declarations, either. > I suspect more patents are issued under Rule 131 than after interferences. Showing that you were working on the problem before an unpleasant reference was published is certainly one use for priority records. But Rule 131 diligence is a much softer, fuzzier animal than interference diligence. Best regards Dave -- David A. Kiewit (727) 866-0669 5901 3rd St.South, St. Petersburg, FL 33705 Registered Patent Agent http://patent-faq.com
Subject: Re: What's the meaning of "invent" in "first to invent" ? Date: Mon, 01 Feb 1999 16:35:00 GMT From: jptucson@azstarnet.com (John Pederson) Message-ID: <36b5c7ae.891160@news.azstarnet.com> References: <36B52DF7.28DD59A@web-site.com> Newsgroups: misc.int-property Lines: 105 On Sun, 31 Jan 1999 21:30:47 -0700, Ed Suominen <see-my@web-site.com> wrote: >John Pederson wrote: > >> Now under the new provisional application rules, I would file >> immediate automatic PPAs on all invention disclosures as received by >> my patent department, thus nailing down the earliest possible filing >> date for those on which we decided to file complete applications. >> This would give the best of all worlds, for only $150 per PPA. >> > >John, I respect your experience and history of well-written posts. I just want to >make one cautionary note. > >A PPA is the world's most easily discoverable draft patent application. True. Good point. >Inventors >can write narrowing and unguarded statements in a PPA such as, "the other piece of >code in Appendix B is probably obvious - everyone here thinks so," and "process A >only works with blue widgets. It isn't suitable at all for use with red widgets." Also true, and lamentable. Of course, we worked closely with our technical community and tried hard to teach them not to do such things, sometimes with success and sometimes not. In that regard, it's worth noting that we were a multi-product company, with all engineering, advanced development and research, and patent department, centralized in one metro area. Also our patent department reported to top management in the early years and to the law department in later years, so we had more credibility and I suspect more success in such efforts than if we had been subservient to Engineering management. >Do you really want to put such statements in the file wrapper with an "automatic" >PPA? If you go through the disclosure to "clean it up," you probably will be >spending enough time to justify just getting started on the claims and >specification. Going through to "clean it up" would be a good idea, and would not be so time consuming in this day and age. Just scan it in, bring it up on a word processor, read and edit as you go, and print out. Shouldn't take more than a few minutes a page. > >Let's not forget the value of priority records for Rule 131 declarations, either. >I suspect more patents are issued under Rule 131 than after interferences. Even in >the unlikely event of an interference, you might wind up being more likely to win >good patent protection by using priority records than by relying on a narrowly >written provisional application with lots of legal holes in it. Again true. And another good point. But I know of nothing that says we can't use the priority records for Rule 131 even if we have already filed the (possibly "cleaned up") invention disclosure as a PPA. And my experience was that we used Rule 131 in very few cases also -- the frequency was more often than interference situations, but still definitely under a couple of percent of the cases. A couple of additional comments: Nothing says that we have to claim the PPA filing date even in a complete application filed within a year. So there is an opportunity to "abort" the PPA (by not referencing it and not claiming its priority date) at the time of filing the complete, if it is an important case and it is thought that it would be better to rely on priority documents only. BTW, we followed the practice of destroying all drafts at the time of filing a complete application, and that practice could be extended to destruction also of our copies of an "aborted" PPA. Sure the PPA could still be discoverable in contested litigation -- so are the corresponding "uncleaned up" priority documents. And sure we might wind up once a decade or so losing a case that we might otherwise have won. It's a risk/reward evaluation, and you can't win 'em all. BTW, we haven't been winning 'em all by doing things the old way. Moreover the risk/reward weightings may vary from one situation to another. Our industry was a patent active industry, and there were many blanket cross licenses between the major players. As pointed out above, we were a multi-product centralized company. And we had enough numbers that the statistical odds were meaningful. For those without the numbers, or in single-product companies, or companies without patent departments, or companies with widely decentralized technical and patent operations, or for private inventors, the outlook might be different. Then again, it might not. I continue to feel that as a rule, all too much time and effort is devoted to record-keeping for the amount of benefit at stake, and that a PPA is a new tool that can be used not only to buy time for finding out if the invention is a winner before spending the much bigger bucks to go after a patent, but also to reduce or eliminate the chances of being victimized by the starry-eyed and widely held misconception that we have a true first-to-invent patent system. Thanks for the observations and comments. Any reactions to mine above? - John Pederson, Retired Former Director of Patents Major Manufacturing Company
Terry Ritter, his current address, and his top page.
Last updated: 1999-02-20