That patents comment

Sunday, December 11, 2011

Comments: 4   (latest 2 days later)

Tagged: amazon, patents, compression, gif, not games

(This is an experimental foray into Zarf Not Talking About Games At All. It may be a confusing new direction here on the Gameshelf, or it may just be a mistake. Be assured that we will continue talking about games too, particularly since Jmac has finished his semester of assistant-teaching and will soon have enough live brain cells for blogging again.)

Several months ago I posted about software patents as they relate to my life. I continue to keep an eye on the subject -- not that there's anything I can do about it -- and so I see the same depressing patent news articles that you do.

A couple of days ago I condensed the following tweet out of the usual whirling filamentous thought matter:

The patent system is based on the premise that ideas are rare and precious, but implementation is easy. No wonder it's failing for software! (-- me, 9-Dec-2011)

This has unexpectedly turned into a minor twitterstorm, with a large handful of retweets, several approving comments, several disagreeing or disdainful comments, quite a bit of threaded discussion in my @mentions, more going on in G+, and at least one person who's already sick of the quote. (Sorry! Nobody plans these things!)

I'm happy with that tweet as a tweet -- but 140 characters is not a legal argument. It's an oversimplification! Of course it is. So, here's a blog post, which will hopefully provide some firmer ground for discussion.

(Note: I am not a lawyer of any sort, nor have I any real-life experience with the patent system. This post will not break new ground in the intellectual-property wars. I just want to connect my little slogan with the real world, and say where I'm coming from -- Twitter can't handle that.)

The first reply I saw was, roughly, "The basis of our patent system is ensuring that innovators get paid." I came back with "That's the goal, not the basis", which was probably overcondensed. I was trying to point out that goals are not results. If the existing system isn't doing its job, the system is broken.

In theory, you can't patent a naked idea; you have to put it in some kind of physical form. In practice, the boundaries between "idea" and "invention" have gotten very fuzzy; the physical form of software (or process, or a procedure) is irrelevant. Okay, that makes my quote fuzzy as well. The question, as several tweeters have pointed out, is obviousness -- or, the difference between the easy part and the hard part.

When Amazon began pushing its "one-click" patent in 1999, people were outraged (I was outraged) because the notion seemed obvious. You keep a user's payment information cued up, and run a sale when he/she pushes a "buy" button. Thinking that up was easy. Building a fully-functioning, easy-to-use shopping web-app was hard. Amazon had done the hard work, and were reaping the benefits, but they didn't want anybody else (Barnes&Noble, for a start) to independently do similar work and reap benefit. Their leverage for doing this was grabbing hold of the easy part -- the idea -- and that felt like an abuse of the patent system.

The same thing is going on with Lodsys. The idea of buying new software features from inside the software is obvious (and, of course, not new at all). Lodsys has contributed no work to Apple's (or Google's) mobile app platform. Nor has the hapless guy they bought the patent from in 2004 or whenever it was. They're coming at this with the attitude of "Well, we (or that guy) did the hard part of thinking up the idea -- the rare and precious idea. Now developers are freeloading -- all they're doing is implementation. We deserve a cut." That's blatantly dishonest; it bears no connection to the real world of software development. It's not even 0.575% true. (The size of the cut they demand, last I checked.)

So that's why I phrased my slogan the way I did. It's a summary, but it gets the idea across at tweet-scale.

(I realize that, no pun intended, writing slogans is the easy part. Rallying political support is the hard part. If people can crystallize some public attention around the issue using my line, that's great! Please use it that way.)

But there is more to this discussion than "easy part", "hard part".

Go back to that first reply: The goal of our patent system is ensuring that innovators get paid. No. That's a fallacy. In my country, the legal kernel says:

(The Congress shall have power...) To promote the progress of science and useful arts, by securing for limited times to authors and inventors the exclusive right to their respective writings and discoveries; -- US Constitution, Article 1, Section 8, clause 8

The goal is promoting progress. The strategy is ensuring that inventors get paid. The means is the US Patent Office and legal system, as laid out in subsequent legislation. You have to ask whether the system is broken, but you also have to ask whether the strategy serves the goal.

The first time software patents infringed my life was the case of Unisys and their "GIF patent". If you were around the Internet in the late 1990s, you can probably say the same.

Unisys did not patent the GIF format. Rather, they (Sperry Corp at the time, where Terry Welch worked) had patented the LZW compression algorithm. The idea of compressing pixel data for an image format is pretty obvious, but LZW compression itself is not obvious. I don't know how it works myself! I could read up on it, but I would never claim that I could have invented it in the first place.

Okay, so LZW was hard to invent and thus deserved patent protection. Open and shut? Not entirely.

What's the public interest in protecting this work with a patent? We want companies to support pure research (industry computer scientists like Welch publishing algorithms). Companies only understand money; patents give them a way to make money off research. Fine so far.

There's also a public interest in letting people use this sort of work. That's why patents have a limited lifetime. The plan is not to grant the inventor power, but to strike a balance between the inventor and the public.

Software patents have some serious problems with this balance -- because in software, we stack inventions higher and faster than in practically any other field. GIF was built on LZW compression. The Web, in large part, was built on GIFs (Mosaic took off because it was a graphical browser). Internet shopping was built on the Web. One-click shopping and in-app purchase were built on Internet shopping... All of that happened in less than twenty years! Most of it happened in less than ten years, 1993 onward. The rate of these things continues to accelerate.

Now, what's the public interest in inventing the World Wide Web as compared to Unisys's desire to make some money off GIFs? How does that balance out?

Look at it this way. Imagine algorithms were unpatentable. Now imagine that Terry Welch was entirely motivated by the prospect of patent income. (Surely a wild exaggeration, but for the sake of argument.) So it's 1983, and Welch scoffs at the idea of inventing a new fast compression algorithm -- or his Sperry bosses do. No profit in it. Welch goes and invents a patentable memory chip instead. LZW is never invented.

What's the result? We wind up with an Internet infrastructure built on a different compression algorithm. But that's exactly what we have! The universal assumption for compression these days is DEFLATE, aka "Zip compression", which is based on the older LZ77 algorithm. It's universal because it was never patented. (Lempel and Ziv are academics; they were at the Technion, not at Sperry.) Web browsers use DEFLATE for fast transfer; Zip, GZip, and PNG files use it; therefore, so do Java's JAR files; and so on. When I needed simple compression for an iPhone project, I looked around and there was zlib, which uses the same algorithm. Bang, problem solved.

LZW fell out of patent in 2003, but that was just a little too late to catch the wave. As far as I know, its most common remaining use is... the legacy of GIFs, which have not been completely displaced by PNGs. But if GIFs were impossible, PNGs would serve the purpose.

If the "worst-case outcome" from discarding software patents is the situation we're actually in, that's a really strong argument that software patents serve no valid public interest.

In fact I don't think software patents need to be discarded. (I also don't think they can be discarded, given the current immense investments today's software giants have made in patent portfolios.) But when you look at this things, you have to look at how they play out in real life. Internet innovation these days is incredibly patent-averse. Look at the mess around video codecs. Look at the idea of "defensive patents". This situation does not promote the progress of science and the useful arts; it is a series of grabs to fence off areas of science from progress.

My usual line is that software patents would be okay if they cycled at the speed of Internet innovation. An 12-month patent lifespan would make sense, but you could talk me up to 18 months.

I mean, imagine Amazon got a monopoly on the one-click "buy" button -- from mid-1999 through Christmas 2000. Wouldn't that have been enough to give them a nice foothold in the online stuff-selling business? After that, their work on an attractive and streamlined shopping site -- you know, the hard part -- would still be in their favor! It's not like they still need the income from B&N and Apple and whoever else has licensed the patent. Let 'em compete on what they've built, or be driven under by someone doing it better.

In a short-software-patent world, I could still spout my line about "idea" and "implementation" -- but at least the public interest would be better balanced.

(But, you say, it's impossible to even file a patent in 18 months, much less litigate a violation before it expires!)

(Yeah, well. The system is broken in other ways too.)

I could go on like this, but I've probably already bored you. I will close with a quote that oughta be repeated more often. From the 19th century, people:

It was never the object of those laws to grant a monopoly for every trifling device, every shadow of a shade of an idea, which would naturally and spontaneously occur to any skilled mechanic or operator in the ordinary progress of manufacturers. Such an indiscriminate creation of exclusive privileges tends rather to obstruct than to stimulate invention. It creates a class of speculative schemers who make it their business to watch the advancing wave of improvement, and gather its foam in the form of patented monopolies, which enable them to lay a heavy tax upon the industry of the country, without contributing anything to the real advancement of the arts. It embarrasses the honest pursuit of business with fears and apprehensions of concealed liens and unknown liabilities to lawsuits and vexatious accountings for profits made in good faith. -- Justice Joseph Bradley, writing for the court in 1883

Comments imported from Gameshelf