IBM's Supreme Court Brief Says That Patents Drive Free Software

H4x0r Jim Duggan writes "For the Supreme Court's upcoming review of the Bilski decision, IBM has submitted an amicus brief claiming that software patents 'fueled the explosive growth of open source software development' (!) (p38 of linked PDF). EndSoftwarePatents, for its own amicus brief, is looking for help building a list of free software harmed by software patents, and a list of companies that distribute free software and are taxed by patent royalties."

Comments

WTF IBM

[Microlith]

Good to see that IBM has no clue what they're talking about. Patents most certainly have not fueled the explosive growth of open source software, the open nature of the licenses and community have. But go ahead and misrepresent the open source community IBM, for your own sake.

Patents sit as an ever present threat that threatens to push development outside of software patent permitting countries, and makes software that is known to violate them into seriously gray territory. I also don't see how a patent, something with the sole purpose of denying use of the described mechanism to others, could possibly aid open source.

Re:WTF IBM

[camperdave]

Patents force people to work around patents.

So a patent produces explosive growth in open source by encouraging the development of alternatives to what the patent covers? Nice. I think I'll use that line as a sig.

Re:WTF IBM

[SanityInAnarchy]

We wouldn't have Vorbis if it weren't for the MP3 patents.

No, but we'd have AAC, which is arguably just as good, maybe even better.

And from what everyone is saying, Theora is far inferior to h.264. If patents weren't an issue, we'd all just declare mp4/m4v with h.264 and AAC as the new standard for the video tag, and there'd actually be cross-browser support.

At the moment, because of real patents, Opera and Firefox won't support h.264 (and thus, youtube.com/html5), and because of imagined patents, Safari won't support vorbis. Thus, it's not just open source projects, but open standards, which are neutered by software patents.

You may have a point with libpng, but then again, gif wasn't that bad. Indeed, gif supports things png doesn't -- animations, for one (there are two competing implementations, one of which has growing browser support (but nowhere near png), and one of which has practically no browser support.) I do prefer png, even with the gif patent expired, but at the end of the day, how big of an improvement was it?

Patents force people to work around patents. It's economically inefficient (just as hurricanes fuel the construction industry) and therefore probably not desirable, but it really does happen.

In other words, it's a broken window model.

Re:WTF IBM

[__aayurq3262]

Saying patents fuel software development (both free and proprietary, since both types are actually harmed by patents) may be a distortion, because it (misleadingly) implies that the patents help the overall situation, but on its face, the statement is literally true. Patents force people to work around patents. It's economically inefficient (just as hurricanes fuel the construction industry) and therefore probably not desirable, but it really does happen.

I regularly go to meetings where management asks the engineers for ideas to get around some competitor's patent. The company would just copy the competing product if possible, but the legal suits say it's too risky. The engineers absolutely love these meetings. Half the time, we come out with some fresh idea on how to make it better or cheaper. The next question the boss asks the suits is whether WE can get a patent. You can't convince me that management would ever have taken the risk to improve the product or do something new without the driving necessity of a competitor's patent.

Re:WTF IBM

[clodney]

Talking about PNG replacing GIF is the wrong angle. You should be talking about PNG and TIFF. AFAIK, the TIFF format can do everything PNG can do (and then some), but was also encumbered by the LZW patent in a significant way at the time. PNG has no real reason to exist, and was a colossal waste of developer effort to create yet another image file format, all because of software patents.

Hmmm, so many things I disagree with in that paragraph. TIFF files can use LZW, but it is one of many supported compression methods, so it is easy to write TIFF codes that doesn't need LZW. TIFF files are limited to 4GB since positions in a TIFF files are 32 bit offsets from the start of the file. That also means that one piece of a multi image TIFF file can not be used without at a minimum knowing where it came from in the file, making it hard to split/join TIFF frames in a TIFF file. There is no restriction that the tags and data portion of a TIFF file be arranged in any particular way, so in general it is not possible to render a TIFF until you have the entire file in hand, making streaming difficult.

In contrast a PNG file is a block of chunks, each of which describes itself, and makes no reference to other locations within the file. It can be streamed easily, and the interpretation of the chunk does not depend on where in the file you find it. PNG files can be more than 4GB. The fact that PNG uses LZ77 rather than LZW is a footnote at best. In case you can't tell, having written software to work with both, I find PNG much more programmer friendly than TIFF.

Re:WTF IBM

[SanityInAnarchy]

Yes, GIF really was "that bad", and PNG was a major improvement on it.

Which does raise the question, would people really have never added these to GIF, or developed PNG, if it weren't for the GIF patent?

And GIF's only "advantage" is mostly used for distracting annoyances,

Yet is still used for, to take an obvious example, AJAX spinners.

it would be quite reasonable to consider it a disadvantage.

Having a feature vs not having a feature?

Here's how it works:

[Tablizer]

1. Stupid patents piss off techies
2. Techies grow to despise corporate-produced software
3. Techies motivated to make open-source variants to take sales away from evil corporations
4. Profit! (Well, okay, I added this one out of habit.)

Re:Here's how it works:

[Verdatum]

I love this idea. I propose a new slogan: "IBM, we're dicks so the good-guys can one-up us."

Re:Here's how it works:

[Rolgar]

Already posted this in another patent story today.

Instead of exclusively considering prior art, we should give the public a chance to respond to every patent application by being given a description of the device, and have an opportunity to develop an invention to the device. If the same or a very similar invention is developed by somebody within a year, then the patent is clearly obvious. If two people submit patent applications simultaneously (and it can be proven that they didn't copy each others' work), then neither person gets the patent. In these cases, we don't need the invention patented, because the point of the patent is give the knowledge of the patent to the public, after a period of time. If multiple people can develop the solution easily, then the device should be considered trivial. The patent examiners should be instructed that patents should only be given out in cases of clear innovation, if in doubt, the device should not be patented.

I really think we should only be handing out a few dozen (if that!!!) patents a year, and I think the above adjustments would move us a long way towards this.

Eric

Re:proper use of the word "patent"

[Anonymous Coward]

"I use the reflections off girls' patent leather shoes to look up their skirts."

Slick Argument

[Anonymous Coward]

Without even reading the brief, you can tell how slick of an argument that is.

Basically, if software "inventions" were only ever covered by Copyright, then IBM and others would never release source code, because it would be too hard to police infringement when rivals companies release compiled products. Plus, Copyright protections would just be too narrow.

But Patents are more broad (though that's too simple terminology), and ostensibly allow you to uncover infringement without first gaining access to the private source code of the competitor, nor do Patents allow you to simply read the source code, then re-write from scratch in a modified manner, as mere Copyright protections would allow.

Of course, IBM is making many assumptions. But you could see how appealing such an argument would to someone already predisposed to accepting Software Patents.

How?

[Errol backfiring]

Forgive for being stupid, but exactly how could a patent help free software? A patent is by its very definition an "unfreedom": a restriction imposed by the holder. If I patent (part of) my software, I cannot call it free software without "disabling" that patent. And then again, I am only putting an unfreedom for somebody else to patent the same idea.

Re:Junk patents

[nine-times]

I'm not against *all* patents. Some algorithms have a serious amount of R&D and ingenuity behind them.

I actually find it a bit tough to come up with a good answer to the problem of patents. The first thing that popped into my head when I read about "free software harmed by software patents" is the whole thing about H264/Theora and the HTML 5 "video" tag.

Now I don't really know what's patented in H264, but I could imagine that it may well be some algorithms with a serious amount of R&D behind them. On the other hand, free software can't legally implement those codecs, and so Firefox can't support it. Free software also can't technically (AFAIK) implement MP3 or AAC encoders/decoders without paying a patent fee.

Now I suppose you could say, "so what?" Fair enough. Still, I have it stuck in my head somewhere that these are important standards that we need to be able to use freely in order to promote the arts and sciences (which is the whole point of patents, isn't it?).

How Free Software is 'Weaponized'

[mpapet]

Classic example of how economic interests take an inherently good thing (Free software) and weaponize it.

IBM couldn't beat Microsoft, so they regrouped around Free software. Everyone still benefits. So far so good.

IBM is still evil though. Anyone old enough to remember when IBM PC *was* a personal computer can back me up on this.

I would argue that IBM is setting themselves up to be able to litigate competitors using Free software on the basis of patented processes inside the code. Sure, the software can be freely distributed, but if you eat into IBM's business, they will litigate the process patents.

Hence the need to conflate Patents and Free software.

Someone please provide some contrary arguments.

over-the-top exaggeration

[prgrmr]

Software is the means by which we use ourcomputers to do word processing, send email and surf the Web; it enables our cellphones to connect to wireless networks; it allows air traffic controllers to safely schedule the arrival and departure of flights; and it permits physicians to diagnose and treat illnesses. Software is, in short, a fundamental, and increasingly indispensable, technological innovation.

Not quite. Software may assist, expedite, or allow, but it certainly does not "permit" a physician to do his job, as he already had that permission prior to the use of any software-driven medical device. Also, if my cell phone bricks, it may be an inconvenience, but it's not indispensable, as I still have other means and methods of communication available to me.

And that's just from the introduction. The rest of it is just a slanted and over-blown, and ultimately, misleading.

In the months since the Federal Circuit issued its opinion, and to IBMâ(TM)s great concern, a number of administrative and judicial decisions have rigidly applied the âoemachine or transformationâ test to questionâ"in some cases explicitlyâ"the patentability of software per se. Software technology is vital in addressing societyâ(TM)s most pressing challenges. IBM is committed to ensuring that such technology is and remains patentable.

(emphasis mine)

This is IBM's only real agenda here.

Too many lawyers

[Anonymous Coward]

Just make a camp, a "fun camp" and "happy camp" (citing Southpark movie) for all of these lawyers and America will be happy place again.

"Open Source" is well-defined by USG.

[dwheeler]

Sure, some lawyers twist the meaning of words... so let's call them on it. But the U.S. government (USG) already has an official definition of "open source software", and it is NOT "you can read it". Office of Management and Budget (OMB) M-04-16 defines the term "open source software", saying that "Open Source Softwareâ(TM)s source code is widely available so it may be used, copied, modified, and redistributed". It's really the "Free Software Definition", but the OSI definition and the Free Software Definition are very, very, very close in practice. And that OMB memo is an official document.

IBM makes piles of money from patents, so no one should be surprised that IBM is for getting more money. But that does not mean it is good for the country. What's more, the Supreme Court has NEVER held that software algorithms are patentable, and the U.S. experiment into software patents has shown that the Supremes were wiser than the patent lawyers. Whether they're willing to make that stick now or not is the big question.

It's not clear that the odds are great, but it would be great if someday the U.S. eliminated the madness of software patents.

Re:explosive

[TaoPhoenix]

More like the anger from patents fuels the manufacture and use of explosives.

Re: poster above you, I REALLY have to get around to my logical fallacy studies project, because this is another one.
Call it 25 technologies produced in anger working around patents, vs 2500 technologies if there was no patent in the way. I don't know the name for that one yet.

Patents are like Go stones. It only takes about 5 brilliantly spaced items to sink 360 squares of attempted growth.