Should Software be Protected by Patent, or by Copyright?

I came across a column in the Manilla Standard Today (http://www.manilastandardtoday.com) discussing the relationship between Software and Patents vs. Copyrights. It's focused on encouraging resistance to allowing software patents in the Philippines, but I found it intriguing.

The writer, a fellow named Chin Wong, has some interesting and intelligent (a rare combination) things to say about that relationship, things I hadn't considered before. So I thought I'd drag it out in the middle here, and see what the rest of you think of the matter. You can find the full article here (http://www.chinwong.com/index.php/site/article/patently_wrong/), as your interest leads you, but here are a few of the points I found most interesting.

Under a patent system, inventors could enjoy the fruits of their discoveries exclusively for a limited number of years in exchange for sharing details of their inventions with the rest of the world. This system, so the argument goes, encouraged inventors by giving them a time-limited monopoly but also guaranteed that others could eventually be able to build on and improve those inventions.I found this interesting because a very similar point often crops up in our various discussions of copyright. Except here, the idea was to encourage technological innovation rather than cultural and ideological innovation.

Mr. Wong points out that this approach works well when the thing being patented has a lot of R&D time and money behind it, allowing the innovator to recoup his investment, during a period where he's protected from those who can copy the design far more inexpensively then actually developing it ever was. And while it occurs to me that something like an OS, or an involved application or game does have that sort of long, costly R&D and testing period (with the possible exceptions of new Microsoft OS's, of course), he points out that this isn't what software patents are being actually granted for:Yet today, patents are being granted for commonplace notions such as selling goods online; electronic shopping carts; video distribution over the Web; credit-card payments on the Internet; and video streaming. None of these qualify as major innovations requiring state protection, so why give any one party monopoly rights over them?

Software has traditionally been protected by copyrights, which prevent others from copying a program’s underlying code. But software patents are much more restrictive and prevent all similar work from taking place, even if it is developed independently. Common sense tells us this does not encourage but inhibits innovation.The article is relatively brief, and the bulk of it consists of a rather amusingly done supporting example Microsoft trying to enforce patent rights on a look and feel that was essentially copied (not to say 'pirated'), in turn, from both Xerox and Apple (who failed at suing Microsoft over more or less the same claims some years back). It got me thinking, more than anything, to what extent is software technology? To what extent is appearance function?

In some cases, form and function are largely the same thing, a scroll bar, for instance, is a scroll bar. There are only so many ways a scroll bar can be made to look, and still perform the function. Yes, it's really clever, and terribly useful, but is it really something that warrants a full-blown patent?

Is there a separate patent on the scroll wheel? That would be an interesting thing to know.

Anyway, I've found I learn far more around here from listening than I do from talking, so I'll close by asking again: what do the rest of you think on the subject?

One notion that occurs to me, as I read this, is that refusing software patents might discourage monolithic, isolationist development in favor of smaller modular products. I like this idea, but some people may prefer being able to buy a big chunk of software from one house, even if it limits what they can add on to it.

Being from the academic community, I tend to look at software as knowledge/tools. Just as research would get nowhere if we were forced to re-invent everything for every new development, I believe software patents simply serve to hold back progress.

Software needs to be reclassified as knowledge and subjected only to copyrights.

One reason MS may not go after the "violations" is that they are afraid of losing. Many patents have been granted, as the writer pointed out, for self evident things. (One wag once submitted an application for the ice cube.)

In many ways MS is like the old line book publishers fighting to keep their old business model.

Patents do serve a valid purpose, they were never designed to stop development or keep innovation from the market. Yet, that is exactly what has happened in software patents. I understand Intel getting a process patent on a way to manufacture a new chip. I do not understand someone getting a patent on putting an auction online. (I wrote an auction program in dBase II under CP/M back in 1982 for a dial-in 300 baud modem. Wonder if I can challenge the patent?)

In today's world, everyone is a software developer, it seems. Without patents, how would anyone build a company on a good software idea. The examples given (online shopping carts, credit card processing, etc) may seem silly now, but imagine the first one. Today, if you launch a good idea, it would not last a month before another group of developers had taken your idea and created an "inproved" version. The only retribution would be that their version would also not last a month. This is the reason for patents. Maybe there are some aribitrary grants out there, but it is the idea that must protected, not just the specific code. It is too easy to write different code that steals another person's idea.
If we think competition is ugly now, just imagine a world without patents. Huge software companies can rest on them now and try to gain the advantage through the protection of the patent. Without that protection, the methods for obtaining a strong foothold on the market with any "new" ideas could only get worse.

I believe the problem here is that software is too unique an entity to the Patent and Copyright offices, and they've never taken the appropriate time to work out what is what.

For instance, maybe an entire package of software can be patented or copywritten, but does that mean someone else's software can't do the same task... or that snippets of code within the patented software can't be used by others? And that is where the real problem lies: Software companies claiming that they have rights to pieces of code, not just entire applications, or that someone else's piece of code is too similar to theirs in function, if not in fact.

In patents, you can patent a device, even if it has other patented devices within it... but you can't patent a piece of a device (say, a screw) if it is not markedly dissimilar to other screws. In copyright law, you can patent a written collection of words, but not individual words. It seems to me, based on legal arguments I've read about, as if software and code challenges are based on the idea that pieces can be patented and/or copywritten, which seems to violate the accepted guidelines of patenting and copyright law.

Perhaps what is needed (and even as I write this, I'm biting my tongue) is a separate entity devoted to software protection, and applying a system that isn't patent, and isn't copyright... but something tailored to the unique issues of software.

The examples given (online shopping carts, credit card processing, etc) may seem silly now, but imagine the first one.Imagining the first one does not make them less silly. They are all straightforward applications of preexisting technology.

Today, if you launch a good idea, it would not last a month before another group of developers had taken your idea and created an "inproved" version. The only retribution would be that their version would also not last a month. This is the reason for patents. Maybe there are some aribitrary grants out there, but it is the idea that must protected, not just the specific code.This is not the reason for patents. You cannot patent an idea - that is a fundamental (but common) misunderstanding of patents. You can patent, for instance, a new type of engine you've developed, but you cannot patent the idea of a device that provides power to move a vehicle.

It is too easy to write different code that steals another person's idea.The patent is for how to write code to implement the idea. If it is easy, then a patent should not be granted for it.

If we think competition is ugly now, just imagine a world without patents. Huge software companies can rest on them now and try to gain the advantage through the protection of the patent. Without that protection, the methods for obtaining a strong foothold on the market with any "new" ideas could only get worse.Patents serve a useful purpose, and I don't think we want to do away with them. However, most software patents are for things that would not have been considered patentable in the past. When it comes to judging whether an "invention" is obvious and therefore not patentable, the patent office seems to have thrown the book out the window.

The purpose of the patent system is exactly what is stated in the article - to encourage inventors to disclose how their invention works. In exchange, they get exclusive rights to the invention for a limited time. If disclosing the invention doesn't tell anyone anything that they didn't already know, then there is no value to society in granting the patent. To take an example you mentioned, online shopping carts: you can't patent the idea of online shopping carts, you can only patent how to make them. But how to make them was obvious - that's why its a silly patent.

I don't, in fact, think competition is ugly. I don't think most of the software patents we're talking about involved that much R&D, either, which is what patents are supposed to protect. If someone comes up with a bright idea and someone else finds another way to do it better, faster, cheaper, etc., by using their own brain (i.e., their own R&D), more power to them. The first idea wasn't worth so much after all, apparently.

It does seem to me that patents are really intended to protect a specific design rather than the concept. I.e. a specific tire tread pattern or construction method can be patent protected, but just because Goodyear has a patent on a tire, doesn't mean that Michellen is infringing that patent by making a tire of their own, even if it's very similar in look, feel, function, etc. In fact, any lawsuit that tried to claim such a thing would be laughed out of the clerk's office before it even got filed. Why then do patent applications go through for things like shopping carts? I expect it's because the patent clerks don't have any grasp of 'software stuff' :shrug:

I spent a number of years as the core of the R&D department of a small software firm. When copyright was the only available protection for software (in the U.S.A, that is), our management chose to hold everything as trade secrets. This essentially barred employees from disclosing our inventions to anyone outside the firm. Throughout this period we published almost nothing. We also grew quite cynical about seeing papers at PLDI (and other conferences) that explained innovations we'd made 3-7 years earlier. Of course, as trade secrets they remained unpublished until someone academic had the same idea and got a publication for it.

Then software patents arrived, and the whole game changed. All of a sudden we could ask our management "Is this a big enough deal to file for a patent?" If yes, we'd file -- thus disclosing the idea to the world. If no, we'd immediately suggest that we should publish. That way, no one else would be able to patent the idea out from under us.* We suddenly began publishing quite a bit -- all stuff that would previously have been kept secret.

My experience is that software patents led to an increase in disclosure of inventions.

Xenophon

* Yes, I know. The patent office really s*cks at finding prior art. Nevertheless, our lawyers told us that if we routinely submitted things for publication we'd be in a much stronger position if someone else got the patent.

I can well believe this, although I wouldn't assume every SW developer works as your office did.

But your comment about the patent office's inability to do a proper job at it reinforces my belief that a separate entity is needed, one that provides a patent/copyright of sorts, but specializes in SW.

I realize we're talking about more bureaucracy, but for such a major thing as computers and software, I think well worth it.

I'm increasingly thinking that a large part of the problem is that the folks at the Patent Office just aren't well enough versed in the software side of things to grasp what's going on with it. So another agency, or perhaps a specialized branch of the Patent Office might be the best way to approach it. I detest bureaucracies as much as the next fellow, but they exist in the first place because sometimes they're the best way to approach things. The drawback, of course, is that you need to constantly stand by with a machete and a flame-thrower to prune them back. :evil:

Patents also nurtures startups. If it weren't for them, we'd be bullied by the likes of MS.

On the other hand if a generic concept, as an example the 'chair', was patented, evolution would not be possible.

The only fair thing is a time limit like we all have...

But your comment about the patent office's inability to do a proper job at it reinforces my belief that a separate entity is needed, one that provides a patent/copyright of sorts, but specializes in SW.

I realize we're talking about more bureaucracy, but for such a major thing as computers and software, I think well worth it.

If you read about the history of patents, this inability has historically been an issue for every significant new technology that comes under patent protection. Some examples:

The British Patent Office granted a patent on "the crank" (back in the early days of steam). This was not a patent on "the crank used in combination with a steam engine to..." but rather on the crank -- no limitations. I think that there were a few thousand years of prior art on this one. :happy2: Some bright fellow had to invent the sun-and-planet gear to work around this one.
The Wright Brothers succeeded in getting a U.S patent on "the wing." In spite of substantial prior art that was well known to the public.
See the many, many fights about the RCA radio patents, later found by the courts to have been granted to the wrong inventors.

There're tons more examples in fields ranging from early electronics, to automobiles, to steam engines, to railroads, to chemistry, to VLSI, to you-name-it. During the course I took on the subject, a number of students tried to find even one example where a new technology DIDN'T experience this sort of problem. We couldn't find any. What we did find was a ton of arguments that sounded just like today's no-software-patents folks. "It's a fast-moving field..." "Patents last too long..." "Interfering with progress..." You'll find them all. For airplanes. For steam power. For... Oh, you get the idea.

The common threads that I see are these:

Patent offices s*ck at finding prior art that isn't from their own (or some other) patent office. Even when it's been widely published in easily available journals.
Every field suffers from 15-35 years of ridiculous, never-should-have-been-granted patents.
Every field goes through the "But we're different..." phase of argument.
When the dust clears, the patent system does a pretty good job of promoting progress, and helping inventors benefit from worthwhile inventions.

None of which is to say that things couldn't be better. Only that the current mess with computers and software is nothing new.

I can well believe this, although I wouldn't assume every SW developer works as your office did.

Sure. For one thing, we actually had and R&D group. It was small (3/4 of my time, plus smaller fractions of a few other folks), but it did exist. Most SW developers don't even have that. As an optimizing compiler developer with heavy roots in academia, we had real incentive to stay on top of recent technology. And to invent things of our own. It was our key differentiator from other players in our market.

I recognize that things may be different at other companies. Most of my reason for telling that story is to remind people that the incentives to publish really are there with patents, and really aren't there with copyright and trade secret protection. Assuming, that is, that you are doing anything that was publishable (or patentable) in the first place.

If you read about the history of patents, this inability has historically been an issue for every significant new technology that comes under patent protection. Some examples: ....Thanks, Xenophon, this is exactly the sort of information/perspective/insight I was hoping might come out of the discussion. Not precisely that, of course, since I didn't know what the insight would be, but I was hoping for something meaty like that. :scholar:

That bit of information makes it seem that this may just be a 'growing pains' stage (the growth related phenomenon, not the TV show). If that's so, and it's relatively short term, then that's well and good. The idea that it might be (likely is) so, makes it easier to think about (and deal with) the situation without spiking the ol' blood pressure. http://www.mobileread.com/forums/images/smiliesadd1/pleased.gif