Internet Software Patentsby Philip Greenspun in February 2008
To judge by the number of lawyers who have been contacting me lately to serve as an expert witness in patent lawsuits, it seems that Internet applications have entered the litigation phase.
Though sympathetic to their case, I turned down their generous offer because I preferred to spend my time engineering new things, not arguing about old things.
A couple of months later, I got dragged into the case via subpoena. Now I was an unpaid material witness. Because I was the actual developer of one-click ordering, not just an expert poring over old documents and publications, I would have to testify at least about what I knew.
One-click ordering is as simple as the checkbox that you see on most Web sites to "keep me logged in". The server sends the browser a "persistent cookie" that the browser returns every subsequent time the user requests a page, even if the computer has been restarted in the interim. That way the server (online store) can identify the browser (customer) and re-use the payment and shipping information from the previous order.
I was asked "Why didn't you patent this yourself, if you developed it first?" My reply was "It only took me an hour to build; if I went down to the patent office after every hour of programming, I wouldn't get very much done."
Fortunately, the case was settled and I never had to appear in court. I decided that maybe next time it would be smarter to take the money and get paid for the work.
A good example is U.S. Patent 5367627, filed in 1993 and issued in 1994. Imagine you have a computerized database of spare parts for different products. A customer calls and says "I need some parts for my Canon 1000 copier". The computer system, rather than showing the sales person all million parts in the system, shows the sales person only those parts that will fit the Canon 1000." The idea of not flooding the sales person's screen with a million irrelevant and incompatible parts is patented. The owner of the patent is suing everyone who sells spare parts in this straightforward manner and, even though the patent did not mention the Web (three years old at the time of filing), claiming that it covers automated Web-based sales as well.
Here's how one could rewrite the claims...
Claim 1a) take model number from customer
1b) database of which parts go with which model
1c) picking at least one part from the database (1b)
1d) making a "proposal" to the customer (web page?) with a part to buy
Claim 2) same as claim 1 but explicitly ask for the "manufacturer's part number" for the equipment (how claim 1 was supposed to work without this, I don't know)
Claim 3) receiving "feature information" instead of a mode number
Claim 4) selecting equipment type and then equipment manufacturer, e.g., "copier" and then "Canon"
Claim 5) printing out the proposal
Claim 6) including in the proposal some extra information about the parts ("specifications")
Claim 7) showing the price of a part
Claim 8) showing "graphical information" (picture of the part?)
Claim 9) "text information" in the proposal (why this is different from Claim 6? specifications are usually text)
Claim 10) putting an animation in the proposal
Claim 11) "graphically illustrating features" in the proposal (how different from 8?)
Claim 12) showing a comparison of features between two different parts that are both compatible, e.g., "500-sheet paper tray holds 2X as much paper as 250-sheet paper tray"
Claim 13) using text to do the comparison? How different is this from 12? How else could you do it? Graphically maybe.
Claim 14) using graphics to describe the two different features
Claim 15) using a portable PC for the whole deal (laptop?)
You might ask "How would you do the '627 patent without a computer?" Here is a process:
Answer phone by saying "Canon parts and supplies. Good morning". Ask customer what model copier he owns, e.g., "Canon 1000" and basically what he needs, e.g., "toner cartridge". Go to filing cabinet. Pull out folder marked "Canon 1000". Leaf through to a page describing "Toner Cartridge 1000 for Canon 1000" and "Toner Cartridge 1001, Long Life, for Canon 1000". Suppose that each page has photos of the cartridges, illustrations of how to put them into the copier, and a price. Using paper and pen, write cover letter to customer saying "Thank you for calling Canon. There are two replacement toner cartridges compatible with your Canon 1000 copier. Attached are product sheets on each. Please call back with a purchase order or credit card number if you wish to order one." Fax everything to the customer. This seems to cover claims 1 through 9 and maybe 11. To cover claim 10, mail a VHS video tape to the customer. To cover claims 12 and 13, include an extra sentence in the cover letter: "Note that the long life cartridge can print twice as many pages." To cover claim 14, draw two stacks of paper, one twice as large as the other, with a caption saying "the long life cartridge won't run out until you've used twice as many pages." To cover claim 15, deal with the customer via email, write the cover letter in a word processor, and scan all of the documents from the filing cabinet, creating folders on the desktop with names that are the same as the names from the filing cabinet folders.It is tough to imagine how anyone could run a parts and supplies department without having done something similar, once they had made the observation that not all parts fit all models. But how to use this as prior art in a court case? You'd want to find a publication explaining this process in detail, but why would an employer do that since one would expect to be able to hire a temp worker and walk him or her through the process in 10 minutes. You can't find an academic publication on how to answer a telephone, open a filing cabinet, and put some documents into a fax machine. Cookbooks don't say "make sure to set the table with plates before serving this food; you don't want to place the steak directly on the table or tablecloth". Paradoxically, the more trivial the process the harder it will be to find prior art in a publication.
How come the patent examiner grants patents on stuff that a 12-year-old Visual Basic programmer would probably implement without too much thought? A patent examiner is only permitted to spend about 12 hours looking at a patent. If he or she cannot find prior art within those 12 hours, the patent is issued. Mostly patent examiners look at other patents, but sometimes they search academic journals and popular magazines. Still, if you had 12 hours to search, including time spent writing correspondence to the applicant, would you be able to find a publication on how to use a file cabinet with hanging folders?
Looking at the file history for some of these patents, one sometimes finds the examiners citing "In Re Venner (1958)" that you can't patent the simple replacement of a manual activity with an automated one. But then eventually the patent is allowed with some narrowing of the claims.
If programmers get dumber every year, how come we're smart enough to keep discovering clever new things to patent, things that those pioneers in computer science didn't dream of? We can buy all of our books on amazon.com and the early Internet pioneers couldn't go shopping online because they weren't smart enough to envision online shopping, right?
The answer is that the early Internet pioneers did envision essentially every service available on the present-day Internet. They wrote about it and distributed those writings to tens of thousands of people. They demonstrated prototypes, sometimes to rooms full of more than 1000 people, and distributed films of those demos. The only reason that we believe ourselves to be innovative is that we are too lazy to go to the library and read what was done in the 1960s.
If those old guys were so smart, why didn't they build amazon.com, eBay, and Google? Well, many of them died before the 50,000th person obtained Internet access. There wasn't much point in having an online store when there were only 50 or 100 computers on the Internet.
They couldn't build our modern world for us back in the 1960s because the hardware hadn't caught up. If you'd given them 50 million quad-core 2 GHz Pentium with 4 GB of RAM and 30 Mbps Verizon FiOS connections to every home, they would have built you all of the services of the modern Internet and probably many that would have been better.
What would happen if you gave present-day computer programmers those same powerful hardware gifts? We did that experiment. Our modern day best-and-brightest built Microsoft Windows Vista (TM).
According to Freeman Dyson, Von Neumann didn't envision home computing.
-- Jim B, February 14, 2008
Jim: I did not mean to imply that every old timer independently predicted every development in computing from 1950 through 2007, only that every significant development was predicted by at least one old timer.
-- Philip Greenspun, February 14, 2008
Awesome Phil. Wanna join the End Software Patents team?
-- Brad Feld, February 15, 2008
While the Old-timers accurately foresaw much of todayıs common technology, they missed the boat in two important areas: user interface, and implications of Mooreıs law. Even as late as the early 80ıs Ken Olsen and the brain trust at DEC didnıt think that there would be a computer in every home ı that was certainly fringe thought in the 60ıs. I used the Minitel system when I lived in France ı91-ı94 and while it was useful, the tiny text screen at low speed has no comparison to the utility of say, Google Maps. Minitel can be used a couple of hours a week to accomplish a practical need like ordering a train ticket, not hours per day to read, communicate, interact, and publish. The ability to interact visually makes such a dramatic in improvement in human productivity. Contrast the iPhone practicality to the green screen 3270 user interface. 60ıs computing was about processing large amounts of data, with the rare exception about making computers personal, portable, connected and interactive. Science fiction dreaming is different than engineering how to build todayıs common systems. Finally, one of Bill Gatesıs greatest insights was the economic implication of Mooreıs law ı all processing, memory (and disk and network bandwidth) would eventually be free, so the value is all in the software. In 1975, no one thought you could maintain a software-only company. Today, everyone takes for granted that software companies like Microsoft and Google have among the highest market capitalizations around. At its current size and weight, Microsoft clearly no longer attracts the best and brightest. Vista proves your aforementioned theory of human endeavor ı that the most exciting work is elsewhere than adding another 30 lines per day to a 50,000,000 line behemoth.
-- David Wihl, February 15, 2008
I am similarly underwhelmed by Vista. However, it doesn't seem fair to compare Vista to the work of Doug Engelbart. Engelbart was trying to change the world. Vista was only trying to be a compatibility environment for a particular application suite. Don't be fooled by Microsoft's need to market it as something more. It would be better to evaluate XO running on OLPC. I'm not sure it would completely vindicate modern achievements, but it at least has a few new ideas.
-- Pete Gontier, February 15, 2008
I'm reminded of Ted Hoff, who told my father that Intel never considered patenting the microprocessor because it seemed so obvious. "Computer on 200 chips" becomes "computer on 20 chips", then "computer on 2 chips" and finally "computer on a single chip". Nothing novel, really.
I believe Dr. Hoff has been supporting himself for the last two decades as an expert witness in patent disputes.
-- Curt Hagenlocher, February 15, 2008
We have to remember that the patent system exists for the benefit of the public, not the inventor. The opportunity to profit from exclusive use of something (that can otherwise be duplicated at essentially zero cost) presumably motivates people to innovate. And society benefits. So the question we should be asking is: "Would Amazon (or Phil) have developed one-click ordering without the possibility of a patent?" With this and other obvious-in-20-minutes-to-the-first-capable-person-to-face-the-problem "innovations", the answer is clearly yes. How can we codify such a common-sense test into patent law? That's the real issue.
-- Steve Strickland, February 15, 2008
Patents can also work against the public good. The Wright Brothers patented the first airplane with a patent so broad that the entire technology moved to Europe until the patents expired. Marconi, of radio fame, was so appalled by this that he put together one of the first patent pools and the Radio Corporation of America (RCA).
If I remember, there was an aluminum patent that was so effective as to create a situation that was remedied by a "must license" decision under anti-trust law.
If you actually look at who makes the money from a new technology, it is usually the second round of folks who bring in the resources once all the innovation has been done for them. There were no textile millionaires in the industrial revolution. In fact, most of the innovators made modest livings, and a good number of them died broke. (Check out Farewell to Alms for some good numbers on this).
-- sas @mit-ml, February 15, 2008
"there were no textile millionaires in the industrial revolution"
Actually there were - lots of them. It's just that in those days a few thousand pounds was the equivelant of millions in today's money. The north of England (Bradford, Manchester etc) and parts of Scotland (especially Dundee and Perthshire) were centres of textile production that saw massive exports across the globe. The industrial revolution, don't forget, lasted a long time. But even at the start, people were getting rich out of it.
The 'jute barons' of Dundee built massive factories and communities. I live in a tenement block once occupied by workers in the textile industry, and there are hundreds more nearby. Virtually all the people who lived here worked for very rich men who got their wealth as a result of the industrial revolution and its effect on textile production.
-- Jonathan Baldwin, February 16, 2008
Something that could help would be an "Open Source" inventors pool. It could be an area in a Wiki for great ideas and inventions that people want to share and get credit for, with the realization that they aren't going to patent them. They would be in the public domain in a place that would eventually be included in patent searchs, a repository of prior art. People could also contribute other prior art, with links to references.
-- Bryan Walls, February 16, 2008
Given recent developments... http://news.google.com/news?sourceid=navclient-ff&rls=RNFA%2CRNFA%3A1970--2%2CRNFA%3Aen&um=1&tab=wn&hl=en&q=software+patents&btnG=Search+News
Anyone want to edit the wikipedia pages on software patents? http://en.wikipedia.org/wiki/Software_patents http://en.wikipedia.org/wiki/Software_patent_debate
-- Tom Turner, February 22, 2008
"While the Old-timers accurately foresaw much of today's common technology, they missed the boat in two important areas: user interface, and implications of Moore's law"
Not so. Regarding UI: In As We May Think (1945), Vannevar Bush described an interface that is stunningly recognizable as web browsing. Doug Engelbart's work in the '60s (from which sprang the GUI, as Phil mentioned) was single-mindedly focused on providing the best user interface for what he called "knowledge workers". Some of his UI features, such as those for condensing text passages for quick skimming, are still unmatched. Direct manipulation showed up in Ivan Sutherland's Sketchpad (1963), whose constrained drawing features are also still unmatched. Windowed UI and WYSIWYG editing came with the Xerox Alto (1973). Even today, Photoshop and its clones still use the UI invented by Bill Atkinson for MacPaint (1983). And many believe that Atkinson's HyperCard (1987) is what the web should have been.
The two decades between Englebart's 1962 opus and the release of the Macintosh in 1984 were far and away the most fertile period of UI innovation. Most progress since then has been simply the wide-scale adoption of those ideas.
Regarding Moore's law: Look at the section "A Simple Vision of the Future" in Alan Kay's Early History Of Smalltalk. You will see a man who intimately understood Moore's Law, even in the late '70s.
-- Bret Victor, February 24, 2008
Most of the responders show their age in the focus of their quite accurate comments.
However, on the origins of computer possibilities, one can go back to the late 1940s and all of 1950s and find most of the things being discussed in various science fiction stories from that era. The silly one that comes to mind is Dick Tracy's communicator watch. That was really the first mention, I can recall, outside of science fiction of the cell phone and related technologies. Not so silly any more?
Much of what the latter day "innovators" have discovered were found in that era, especially in the more hard core science magazines such as "Analog". Those stories range from the silly to the sublime. It is a shame that so much of that literature has been lost or worse cooped without attribution. It would be interesting to see the effect of bringing those stories into the patent wars.
I find the ideas that "futurists" are offering to be old stale ideas from that era and earlier. Much of the credit for inspiring new technologies that is given, justly since the older literature is forgotten, to "Star Trek" and other newer SciFi productions is a reprise of the ideas from that era.
Heinlein's old short story, "The roads must roll" predicted the, probably patented, people movers we find in airports and other places.
I forget the author but "Satellite E-One" predicted private ventures to move into space, I only remembered that due to the title popping up as I typed. There are many others.
One final observation. When we finally first landed on the moon, many older science fiction readers were not so terribly excited. They just asked, "What took us so long?"
-- Josh Bernstein, October 30, 2008
I'm convinced that it's time to eliminate the software patent process, and simply use the existing copyright and trademark protection for software. See my article on blog.startupprofessionals.com titled "Software Patents: Time for a Change" for specifics.
Marty Zwilling, Founder & CEO, Startup Professionals, Inc.
-- Martin Zwilling, January 16, 2009