Trolling just got a bit tougher
June 4, 2013 12:51 PM Subscribe

Patent trolling is viewed by many as modern scourge on the software industry that arguably does not stimulate innovation, but stifles it. The drumbeat has been getting louder for action against the practice. Today, the White House has announced that it is ready to take action. Previously
posted by Didymium (176 comments total) 14 users marked this as a favorite

Great "please don't sue us!" headline from Bloomberg.
posted by Artw at 12:57 PM on June 4, 2013 [3 favorites]

What's the point of talking about this when it's an empty and cynical political gesture? Does anyone really expect anything to change? This is like commenting on bills that are expected to die in committee. There's absolutely no way that the corporatist interests will allow this to go anywhere.
posted by ishrinkmajeans at 1:00 PM on June 4, 2013

Here's the thing. What makes patents valuable to the inventor is the ability to transfer the rights to the patent. Your average inventor or company does not have the resources to pursue every claim. Just like other types of property, a patent's transferability is its value.

Also, let's look at who we are really fighting for, from the linked Slate article:

such deep-pocketed industrial giants like GE Healthcare, Siemens, and Philips

Giant companies are the ones doing the complaining--that they can't make things because entities who bought patents are suing them. Well, boo-hoo. Life is so difficult for Siemens that they cannot make money stealing other people's intellectual property.

What this stuff really is is a trojan horse for European and Japanese style patents, which have little value. Its no coincidence that large companies want to make a lot of money building things other inventors and companies invented without paying for them.

The problem is that they've enlisted a lot of programmer types (who rarely see the profits from their work anyway, because they are automatically owned by the companies they work for) who are into downloading stuff for free by making the case that their giant need for profits is the same as wanting to download that Audioslave record they've always wanted.

In the end, this is big business not wanting to pay smaller people the value of their inventions. By destroying the secondary market for intellectual property and inventions, they are the ones that are stifling invention.
posted by Ironmouth at 1:02 PM on June 4, 2013 [4 favorites]

There's absolutely no way that the corporatist interests will allow this to go anywhere.

Uh, what makes you think corporatist interests are against this?
posted by naju at 1:03 PM on June 4, 2013 [1 favorite]

Having anything to do with programming certainly is a good way to have it demonstrated to you that the current patent system is mostly garbage, yes.
posted by Artw at 1:04 PM on June 4, 2013 [16 favorites]

This Week's This American Life is pretty damning. Although there is a lot of money in the "buy a maddeningly vague patent and use it to bully businesses into paying out tens of millions of dollars," this actually seems to be the work of a small group of aggressive (and very sketchy) attorneys, many of whom hold multiple shell organizations, and ends up being something that actually is a problem for corporations.
posted by Bunny Ultramod at 1:06 PM on June 4, 2013 [7 favorites]

Because its essentially monopoly-capitalistic. It allows people who already have power and money to increase their holdings by buying up the ability to sell product in the marketplace. Apple loves the ability to buy up any and all patents for anything that is vaguely phone shaped. It's a way of building a wall around all future competitors.
posted by ishrinkmajeans at 1:08 PM on June 4, 2013 [2 favorites]

In the end, this is big business not wanting to pay smaller people the value of their inventions.

You don't have to invent anything to get a software patent. The guy who cost RIM millions in fees got his patent for nothing more than thinking up a few random ideas and then filing the patents. He had no expenses, his ideas were obvious to anyone with knowledge of the field and he had never made any attempt to commercialize his patents.

It's a straight-up shakedown via the courts.
posted by GuyZero at 1:09 PM on June 4, 2013 [26 favorites]

What this stuff really is is a trojan horse for European and Japanese style patents, which have little value.

I can't speak for other industries, but in the software industry it is blatantly obvious to just about everyone that reducing the value of patents would be a very good thing. We are all stuck in a bizarre, inefficient game of mutually-assured destruction where every company has to waste time and money filing patents not because they are actually good for anything, but because someone else might someday come along and threaten us with a worthless patent of their own, which we'd then have to defend against. Thus there is an arms race which causes the multiplication of bullshit patents.

The point of the patent system, originally, was to convince inventors to publish their designs, so that people could benefit from them. Software patents do no such thing; people learn about new techniques and algorithms from academic papers or from open source projects, and most software engineers are in fact forbidden from reading about patents for fear of accidentally triggering "wilful patent infringement" and the accompanying treble damages.
posted by Mars Saxman at 1:16 PM on June 4, 2013 [33 favorites]

This is like commenting on bills that are expected to die in committee.

From EFF, not exactly known to be corporatist:
"Even more important are the five executive actions the President intends to take with or without congressional help."
posted by Lemurrhea at 1:17 PM on June 4, 2013 [3 favorites]

In the end, this is big business not wanting to pay smaller people the value of their inventions. By destroying the secondary market for intellectual property and inventions, they are the ones that are stifling invention.

Are you contending that so-called patent trolls are not a problem?
posted by mr_roboto at 1:17 PM on June 4, 2013 [2 favorites]

If we take power, can I hit Nathan Myhrvold in the side of the head with his stupid poncey cookbook?
posted by sonascope at 1:20 PM on June 4, 2013 [13 favorites]

There's absolutely no way that the corporatist interests will allow this to go anywhere.

Uh, what makes you think corporatist interests are against this?

I think there's plenty of support in the business community, particularly in tech, for addressing the issue of extortionate patent holders. The interests that will be strongly against this are the law firms. They're the ones who make money on the inefficiencies in the current system.
posted by mr_roboto at 1:20 PM on June 4, 2013 [5 favorites]

What's the point of talking about this when it's an empty and cynical political gesture

If the Electronic Frontier Foundation says it's good news, than it's probably good news.

In the end, this is big business not wanting to pay smaller people the value of their inventions.

Listen to the This American Life episode. Large corporations aren't the bad guys here. Or at least they aren't the primary bad guys.
posted by diogenes at 1:22 PM on June 4, 2013

I'll take big corporations over parasitic bullshit corporations.
posted by Artw at 1:23 PM on June 4, 2013 [3 favorites]

Today, the White House has announced that it is ready to take action.

Never a comforting phrase.
posted by IndigoJones at 1:27 PM on June 4, 2013 [1 favorite]

parasitic bullshit corporations

The patent troll shell companies should be forced to include that on the front door of their empty offices.

Innovate ProgressTech Research LLC
(Parasitic Bullshit Corporation)
posted by diogenes at 1:29 PM on June 4, 2013 [2 favorites]

And also, in the case of NTP vs RIM, the US Patent Office eventually found 8 of NTP's patents to be invalid.

Part of the problem is that the US Patent Office is terribly underfunded and understaffed and many (if not all) software patents granted are actually junk patents. But it takes millions of dollars to prove it, so many companies settle rather than fight.
posted by GuyZero at 1:30 PM on June 4, 2013

Apple loves the ability to buy up any and all patents for anything that is vaguely phone shaped.

Apple would probably love to have Lodsys and similar trolls cut at the knees even more.
posted by Blazecock Pileon at 1:31 PM on June 4, 2013

Patent trolls are only a problem when you aren't the patent troll. There's no way to get rid of patent trolls, or really distinguishing what a patent troll even is, without destroying the patent system. Is Apple a patent troll or Google or GE? They buy up a lot of patents and sue a lot of people. There's no way to distinguish between troll and rightful ownership by letter of the law.

This handwringing is just a belated realization that exploiting a system for ones own gain at the expense of the greater good eventually causes the exploiter to become the exploited. Arbitrary property law based on symbolic adversity is a system which is easily exploited and therefore unstable.

If large corporations eventually pass legislation barring Intellectal Ventures et al from holding patents you really haven't solved the problem now have you? You've moved from an arms race to buy patents to an arms race to buy legislators declaring your competitors patent trolls and defending your own claims.

In the meantime the little guy is still screwed because he doesn't have the money to buy legislators or patents. Whatever the end result of this debacle nothing will, effectively, change.
posted by ishrinkmajeans at 1:32 PM on June 4, 2013 [1 favorite]

Google bought up a huge bag of phone-shaped patents and honestly it hasn't done them a ton of good. Patents are an arms race for a war that no one even wants to fight.

But this isn't even a good comparison - "phone" patents include a lot of serious actual patents for things like codecs and hardware implementations of standards-compliant (or standards-defining even) technical things most of which are made available for use by anyone under FRAND (fair, reasonable, and non-discriminatory) terms. These patents have a quantifiable economic value in that they usually generate a predictable royalty stream.

Software patents, on the other hand, generally have nothing to do with standards bodies and are generally used in huge all-or-nothing lawsuits.
posted by GuyZero at 1:35 PM on June 4, 2013 [3 favorites]

There's no way to get rid of patent trolls, or really distinguishing what a patent troll even is, without destroying the patent system.

Eliminate software patents, eliminate business method patents, seriously enforce the obviousness criterion.

That will solve 90% of the problem, and make the system more, not less, robust.
posted by mr_roboto at 1:43 PM on June 4, 2013 [13 favorites]

seriously enforce the obviousness criterion.

Yes. Give the patent office more resources than that rubber stamp they currently use.
posted by GuyZero at 1:44 PM on June 4, 2013 [9 favorites]

Ironmouth: Here's the thing. What makes patents valuable to the inventor is the ability to transfer the rights to the patent.

No. This is a thing that makes patents valuable to the inventor. But you could license the rights without selling them. And even without that, you could start a company that utilizes the patent, or even utilize it (at a small scale) yourself.

Giant companies are the ones doing the complaining--that they can't make things because entities who bought patents are suing them. Well, boo-hoo. Life is so difficult for Siemens that they cannot make money stealing other people's intellectual property.

LOTS of people are complaining about patent abuse. If the system is broken it should be fixed regardless of who benefits from it!

What this stuff really is is a trojan horse for European and Japanese style patents, which have little value.

If that's a trojan horse, I will happily open the gates for it.

The problem is that they've enlisted a lot of programmer types (who rarely see the profits from their work anyway, because they are automatically owned by the companies they work for)

Why treat a perspective accurately when you can stereotype a class of people?
posted by JHarris at 1:46 PM on June 4, 2013 [5 favorites]

I'm in favor of limiting or eliminating patents wherever possible. It would short circuit this Chinese garden of monopoly control and allow small businesses to get off the ground.
posted by ishrinkmajeans at 1:48 PM on June 4, 2013

Eliminate software patents, eliminate business method patents, seriously enforce the obviousness criterion.

Any two of these would be as if Patent Jesus had descended to begin the Second Coming. And seems about as likely in the near future, but it'd be great. I'd settle for just the obviousness thing, though.
posted by Holy Zarquon's Singing Fish at 1:48 PM on June 4, 2013 [2 favorites]

A couple comments removed, cut it out with the bickering.
posted by cortex (staff) at 1:48 PM on June 4, 2013

There's no way to get rid of patent trolls, or really distinguishing what a patent troll even is, without destroying the patent system.

If you read the actual White House statement it has some pretty clear actions that would actually help make software patents less of a useless anchor for the industry:

Tighten functional claiming: requiring patent applicants to explain their inventions better and to limit those inventions to a specific way of accomplishing a task, as opposed to all ways of accomplishing a task.

That's pretty much the core problem with software patents in my view. You can't get a patent for a car engine by just describing in vague terms that it's a box that takes in energy and makes the car move, but you can get ridiculously broad software patents for things like showing a notification when you receive an email on a mobile phone. Software patents should be granted for novel algorithms, which are roughly analogous to real-world machines, rather than for high-level descriptions of something that can be done with software.
posted by burnmp3s at 1:52 PM on June 4, 2013 [18 favorites]

I bet Apple would like to see something done about trolls right about now...
posted by Thorzdad at 1:56 PM on June 4, 2013

Eliminate software patents, eliminate business method patents, seriously enforce the obviousness criterion.

What, exactly, is a software patent? Methods are (and should be) patentable. Why should it matter that a method is performed with/using/by a computer?

The obviousness criterion should be handled better than it is, but there are very good reasons for the problems there.

you can get ridiculously broad software patents for things like showing a notification when you receive an email on a mobile phone

[Citation Needed], but even if you can supply one, I'll bet the filing date is far enough back that it's plausible nobody had ever thought about showing email notifications on a mobile phone before. That's the point of a patent: not just that you explain how to do something, but that you're the first person to have both the idea of doing it and the technical chops to make it happen.

IMHO, software is problematic because many of the things you can do with it, stating the problem is all it really takes to solve it. (E.g., "have one button on a web page that if you click it, no further activity is required to cause the order to be placed.") It's not technically difficult to accomplish that, but somebody had to think of doing that, and realize that it would be a useful thing to do.

Should we disincentivize people thinking of things like that? Make them feel that "eh, it's not worth actually doing it, since BigCo is just going to steal the idea from me."

The idea of a shorter-term patent is occasionally floated, and I think that makes a lot of sense. 5-7 years is an eternity in software, and a term like that would really cut down on the problem mentioned above, where somebody in the 90's (perhaps) thought of notifying email on a cell phone, but it took 10 years for the rest of the world's infrastructure to catch up so that you could actually do it.
posted by spacewrench at 2:16 PM on June 4, 2013 [2 favorites]

Is there any history on how software patents ended up so squishy? I mean, you're right that a patent for "a device that accepts combustible fuel as an input and produces kinetic energy directed to wheels" would get laughed out of the office, while software patents in the form of "any and all methods to produce Onscreen Effect X" are routinely granted. But how in the name of everything even remotely sane did that happen?
posted by Holy Zarquon's Singing Fish at 2:17 PM on June 4, 2013 [2 favorites]

Should we disincentivize people thinking of things like that? Make them feel that "eh, it's not worth actually doing it, since BigCo is just going to steal the idea from me."

In terms of the one-click patent? I can assure you they'd still do it. Because it's completely obvious. And if big company "steals" the idea, you would see no decline in your own profits. Everyone can have a one-click system and everyone wins. It's not a zero-sum game.
posted by Jimbob at 2:30 PM on June 4, 2013 [2 favorites]

Moreover, Amazon registered the one-click patent. Are you really suggesting they just wouldn't have bothered to institute the idea without patent protection? Because that seriously beggars belief.
posted by ssg at 2:37 PM on June 4, 2013 [1 favorite]

I used to work for attorneys in a patent office of a big company and they were the best bosses in the world. I don't think most people can understand the minutia that goes on with patents (and this was mostly hardware). I really wish I had stuck with them instead of moving onto the software engineering marketing group (the patent guys offered me a f-t job after I'd transferred). In fact, most attorneys I have ever worked for were the most pleasant people, even the criminal and divorce attorneys. But the patent attorneys were very pleasant and cheerful. I really, really wish I'd stayed with them.

I am trying to follow this conversation as best as I can with my limited knowledge, and it strikes me that Ironmouth's description might be a good one -- I can certainly understand his language and reasoning. But if there were a more layman's terms response and/or article about the whole trolling and software issue, I would appreciate that as well. I think the back and forth gets really confusing for lay people, but we do the best we can with the resources we have at our disposal. As in, how does this affect the common man and his efforts toward software production... or making apps, etc.?

I'm sorry if I'm showing my ignorance, but there it is. How and why should the common man or woman care about this?
posted by Marie Mon Dieu at 2:39 PM on June 4, 2013

That's the point of a patent: not just that you explain how to do something, but that you're the first person to have both the idea of doing it and the technical chops to make it happen.

Again the recent This American Life episode is instructive. Guys who's great "technical chops" led them to distribute magazines on casette tape are now going after podcasters for patent infringement. Meanwhile, I'm left wondering how magazines-on-tape from the mid-1990s are any more innovative than books-on-record from the 1940s...
posted by Jimbob at 2:41 PM on June 4, 2013 [2 favorites]

I wonder what this will do to the patent troll threatening a lot of the major podcasters. The EFF is working on the case but it will take a while.
posted by fifteen schnitzengruben is my limit at 2:42 PM on June 4, 2013 [1 favorite]

There's no way to get rid of patent trolls, or really distinguishing what a patent troll even is, without destroying the patent system.

Then let's destroy the patent system. It is doing far more harm than good to the software industry. Big businesses will benefit, sure, but small businesses will benefit even more, and open source software will benefit most.
posted by Mars Saxman at 2:42 PM on June 4, 2013 [2 favorites]

How and why should the common man or woman care about this?

They shouldn't? This is an industrial regulatory reform. It's not exactly removing lead from gasoline, but it will remove an unnecessary cost and impediment for companies whose primary product is software.
posted by GuyZero at 2:44 PM on June 4, 2013

Here's how fucked the patent system is:

Until ~2005, I believe, International Gaming Technology held a patent on:

The wheel. Specifically, any mechanical device or representation thereof that rotates through 360 degrees on an electronic gaming device in order to award a prize.

In other words: They patented the idea of slot machines having a wheel that spins to award a prize.

Methods are (and should be) patentable.

I have yet to read a convincing argument for that parenthetical. I look forward to yours.
posted by PMdixon at 2:51 PM on June 4, 2013 [3 favorites]

There's no way to get rid of patent trolls, or really distinguishing what a patent troll even is, without destroying the patent system.

Then let's destroy the patent system. It is doing far more harm than good to the software industry. Big businesses will benefit, sure, but small businesses will benefit even more, and open source software will benefit most.

Big businesses are the ones who are claiming to be "victims" here. As if it is a terrible thing that they have to prove that they came up with something first.
posted by Ironmouth at 2:57 PM on June 4, 2013

idea of doing it and the technical chops

But that's exactly the problem with software patents. You don't have to show the technical chops, or describe how to do 'it'. A software patent doesn't require that you submit a single line of working code. Instead, you can draw some pretty pictures, write some text and, congratulations, you have a patent.

Here's the patent on Dynamic Web Page Generation from the mid=90s. It pretty much covers damn page on the web, including this one. It's a bunch of vague language, and and a few crude drawings. It was 'invented' by the guys who did my.yahoo.com, and Yahoo has sued folks based on it (e.g. Facebook).

It includes no code. If it did, it would be pretty easy to show that whatever method they used back in '97 has little to do with how we generate web pages these days.

It's exactly these kinds of bullshit software patents that Patent Trolls use to extort money from people who actually create stuff. I'm thrilled people are starting to notice this is a problem.
posted by Frayed Knot at 2:57 PM on June 4, 2013 [8 favorites]

Big businesses will benefit, sure, but small businesses will benefit even more, and open source software will benefit most.

There's a great compression algorithm that would be perfect for a genomic data format my software works with, but sadly either Fujitsu or IBM own the patent on implementation. I'd love to see Obama's proposal see the light of day, but it is doubtful anything concrete will come out of this for him to sign before leaving office. This is just pie-in-the-sky from a lame duck administration.
posted by Blazecock Pileon at 3:00 PM on June 4, 2013 [1 favorite]

Big businesses are the ones who are claiming to be "victims" here.

So? I mean, there are a lot of wrongs to right in the world and I suppose this one isn't human trafficking or ethnic cleansing, but it's still not good. Even big companies deserve their little slice of justice. The trolls are indefensible and your argument essentially boils down to companies not deserving of justice because they're "too big".
posted by GuyZero at 3:02 PM on June 4, 2013 [1 favorite]

Big businesses are the ones who are claiming to be "victims" here. As if it is a terrible thing that they have to prove that they came up with something first.

As if it is a terrible thing that they can be sued for millions of dollars for scanning documents and then e-mailing them, as if that was somehow an "obvious" use of a scanner that plugs into an Internet-connected PC. The whiners!
posted by Holy Zarquon's Singing Fish at 3:03 PM on June 4, 2013

Bullshit is bullshit and is not worth championing, even if it hurts people you disapprove of. I can assure you it harms lots of small businesses and individual developers too, if that helps.
posted by Artw at 3:05 PM on June 4, 2013 [6 favorites]

Software patents, on the other hand, generally have nothing to do with standards bodies and are generally used in huge all-or-nothing lawsuits.

Counterpoint: the thousands of patents in the MPEG LA patent pools, covering such standards as MPEG-2, ATSC, H.264, and MPEG-4, and which have generally not been used in huge all-or-nothing lawsuits.
posted by jedicus at 3:06 PM on June 4, 2013

Here's your citation. It's from 2000, granted in 2009. Honestly, poll 100 CS undergrads and probably 10 of them will come up with the exact system described in this patent. It's a bit sad.

Oh and it took 5 seconds of google searching to find that patent. It's the top hit for "email notification patent."

Did you actually read that patent? It's not a general email notification patent. It's specifically about notifying a user of a new email by sending them a voicemail by converting part of the email header using text-to-speech.

So two points: 1) that's vaguely interesting, especially in 2000 (or really in June of 1999, as that's when you'd need to find prior art from and 2) nobody I know uses such a system so who cares about that patent?
posted by jedicus at 3:10 PM on June 4, 2013

Counterpoint: the thousands of patents in the MPEG LA patent pools, covering such standards as MPEG-2, ATSC, H.264, and MPEG-4, and which have generally not been used in huge all-or-nothing lawsuits.

Um, there was that whole back and forth over Google trying ti create an open standard outside of H264 and whether or not it actually Violated those patients.
posted by Artw at 3:11 PM on June 4, 2013 [1 favorite]

Um, there was that whole back and forth over Google trying ti create an open standard outside of H264 and whether or not it actually Violated those patients.

Yes, there was some back and forth. There was not a huge all-or-nothing lawsuit.
posted by jedicus at 3:12 PM on June 4, 2013

A big Mexican standoff where nobody fires a gun is still a Mecucan standoff.
posted by Artw at 3:13 PM on June 4, 2013 [2 favorites]

So I'm not a patent attorney but in my personal taxonomy of patents the MPEG-LA patents are "real" patents since 1) they implemented in hardware sometimes and 2) they actually took work and research to develop. Fundamentally they're not unlike GSM codec patents (well, except they're not expired I guess) in that they're "just math" but they're important math that took work to find and is required for standards compliance.

But clearly patent lawyers consider all patents "real" patents because they're about hating the players and not the game.
posted by GuyZero at 3:13 PM on June 4, 2013 [2 favorites]

Did you actually read that patent?

Yes, but I guess I stopped before the text-to-speech part.
posted by GuyZero at 3:14 PM on June 4, 2013

What happens if you have a set quantity of time to begin development of your patent, and if not it goes into the public domain, or at very least isn't legally actionable? You patent something, you get a year or two to begin the process of commercialization of that patent, or it disappears. You buy a patent, same deal. Doesn't this effectively ruin patent rolling as a business model and reduce the need to file needless patents?

This is called a working requirement. It would be effective if it worked. It seems to get suggested most of the times that patents get discussed on MeFi and elsewhere. Several countries have them, they aren't effective there, and there's no particular to reason to think that the US could do a better job of implementing them fairly and effectively. See this prior comment for why.
posted by jedicus at 3:15 PM on June 4, 2013 [1 favorite]

Big businesses are the ones who are claiming to be "victims" here. As if it is a terrible thing that they have to prove that they came up with something first.

It's everyone who faces the fact that the settlements are cheaper than the court costs. It's just that big companies also have the money to lobby. Are you seriously suggesting that (to pick an example) open source doesn't have a patent problem?

Merely consider the *extremely* poor quality of the majority of software patents. I am someone who specialised in IP (English law), and I think that a great deal of the threatened actions are merely shakedowns that are too expensive to fight. I don't think that the US approach to granting software patents has done any favours to the software industry at all.
posted by jaduncan at 3:16 PM on June 4, 2013 [3 favorites]

The intro makes the e-mail notification patent out to be broader than it is. It says "if" notification will be by voicemail, then you can do text-to-speech. But the actual body of the patent presents text-to-speech as a vital part of the whole arrangement, with a text-only version (sent to pager) as an optional step.
posted by Holy Zarquon's Singing Fish at 3:16 PM on June 4, 2013

But clearly patent lawyers consider all patents "real" patents because they're about hating the players and not the game.

All patents are real patents in the sense that trying to single out particular types of patents for reform is doomed to failure as a political and practical matter. Furthermore, what's good for the goose is good for the gander. There are reforms that would substantially improve patent quality and reduce questionable patent litigation for all patents. Why not focus on those?
posted by jedicus at 3:18 PM on June 4, 2013 [1 favorite]

By destroying the secondary market for intellectual property and inventions, they are the ones that are stifling invention.

Oh, pffft. The "secondary market for intellectual property and inventions" in its current state -- which is primarily a mechanism whereby shameless sleazebags extort money from companies that actually make things, while occasionally paying individual inventors a pittance for a patent here and there -- stifles invention all on its own.

The only people I've ever heard from who seem happy with the current system are lawyers. That right there is a pretty good indicator that it's perverse. The people actually making stuff seem almost universally agreed that it's a broken system that rewards creativity in patent applications, and not actual "invention" in any practical sense.
posted by Kadin2048 at 3:18 PM on June 4, 2013 [4 favorites]

The intro makes the e-mail notification patent out to be broader than it is. It says "if" notification will be by voicemail, then you can do text-to-speech. But the actual body of the patent presents text-to-speech as a vital part of the whole arrangement, with a text-only version (sent to pager) as an optional step.

What matters is the claims. The claims all require the text-to-speech and voicemail stuff.
posted by jedicus at 3:19 PM on June 4, 2013

Why not focus on those?

We are in complete agreement. My comment on "real" patents was intended to mean that actual patent lawyers don't get the benefit of constructing arbitrary groups of "real" and "fake" patents to deal with like I do. Indeed we should raise the bar for granting patents in a non-discriminatory way, mostly through the patent office employing actual experts and doing actual checks into the existence of prior art.
posted by GuyZero at 3:21 PM on June 4, 2013

I don't think that the US approach to granting software patents has done any favours to the software industry at all.

Oh, you mean the US and its world-dominating software industry? The same software industry whose major monopoly (well known for its "embrace, extend, extinguish" business model) first encountered significant competition after changes in the case law in the 90s opened the doors to software patents, leading to the patents upon which Google was founded and upon which Apple's comeback was founded (in the form of the iPhone and iPad)?

Yeah, the US approach to software patents has really killed the US software industry. Too bad we couldn't have been more like Europe and Japan. It's just like biotech, really. The US embraced biotech patents while Europe and Japan resisted them, and now Europe and Japan are where all the biotech companies are. Oh wait.
posted by jedicus at 3:24 PM on June 4, 2013

Student Wins Intel Science Fair; Threatened With Patent Infringement Claims For Patent Not Yet Granted
posted by homunculus at 3:28 PM on June 4, 2013 [2 favorites]

you can get ridiculously broad software patents for things like showing a notification when you receive an email on a mobile phone

[Citation Needed], but even if you can supply one, I'll bet the filing date is far enough back that it's plausible nobody had ever thought about showing email notifications on a mobile phone before. That's the point of a patent: not just that you explain how to do something, but that you're the first person to have both the idea of doing it and the technical chops to make it happen.

The one I was talking about was this one, the European equivalent of which was used to shut down Apple's push email notifications in Germany. And yes, it was originally filed in 1995 for pagers rather than email, but the fact that an old pager patent is stopping software from doing trivially easy tasks using systems that in no way resembled the systems that the patents were originally created for is a symptom of software patents given far too much leeway in terms of what they can cover.
posted by burnmp3s at 3:28 PM on June 4, 2013 [2 favorites]

An example of the difference in approach would be the Amazon one-click patent (US 5960411). The patent was used to force settlement from Apple and Barnes and Noble. After the conclusion of the litigation, the USPTO eventually rejected the vast majority of the claims and Amazon was forced to amend the two major one-click claims to far more limited forms. The patent as granted was poor quality, but even defendants with the resources of Apple settled. What chance do smaller entities have?

The EPO stated that in their view the patent application was too broad and that the technology of the one-click checkout lacked a genuine inventive step. The reader must take their own view of the superior approach.
posted by jaduncan at 3:29 PM on June 4, 2013 [2 favorites]

How Long Before A Patent Kills A Hundred Million People?
posted by homunculus at 3:29 PM on June 4, 2013 [1 favorite]

If your position is that Microsoft's dominance of the OS market through the 90's and early 2000's was due to patents, I would love to see some details.

Generally speaking, patents have been an annoying sideshow in the software world. The tendency of specific companies to dominate specific sections of the software market has more to do with other, non-regulatory factors.

Apple's resurgence with the iPhone is generally held up as an example of where patent protection is irrelevant - the iPhone was a trade secret up until the moment they unveiled it. Some competitors dismissed it immediately to their detriment (RIM) and others immediately started trying to catch up (Android).

But iPhone-related lawsuits haven't stopped competing smartphones from entering the market to any degree that I'm aware of. Perhaps you know stuff I don't.
posted by GuyZero at 3:30 PM on June 4, 2013 [1 favorite]

leading to the patents upon which Google was founded and upon which Apple's comeback was founded (in the form of the iPhone and iPad)?

Lots of innovation went into the iPhone, and yet the patent's Apple actually sues with are for stupid shit like rectangles.
posted by Artw at 3:31 PM on June 4, 2013 [1 favorite]

Sorry, I'm getting snarky. Getting to the substance of the administration's proposals: they're a bit weaksauce, to be honest, but mostly positive or at least unobjectionable I'm not sure the focus on real parties in interest will actually make much of a difference, for example. Intellectual Ventures uses a maze of holding companies to own its patents, but it sues under its own name. And it's well-known who many of the patent assertion entities are really working for. I don't suppose it will necessarily hurt anything, though.

Rolling software patents into the special program for business methods is a mistake, just like the business method program was in the first place. First, it's special treatment for a particular class of technology. Second, it's yet another round of administrative agency review. At some point we have to say "okay, the Patent Office got a first, second, and third look at this, let's be done with it so the courts can have their first, second, and potentially third look at it." Adding more rounds of review only makes the whole process more expensive and time-consuming, not less.
posted by jedicus at 3:31 PM on June 4, 2013

If your position is that Microsoft's dominance of the OS market through the 90's and early 2000's was due to patents, I would love to see some details.

My position is that Microsoft's dominance of several markets was enhanced by its ability to carry out its "embrace, extend, extinguish" strategy, a strategy that is broken if the competitor has a patent and so cannot be unwillingly embraced or extended.

Apple's resurgence with the iPhone is generally held up as an example of where patent protection is irrelevant - the iPhone was a trade secret up until the moment they unveiled it.

Apple certainly invested a lot of money in patent protection, not to mention specifically stating patenting the hell out of it in the introductory keynote address, for something that was irrelevant.

Lots of innovation went into the iPhone, and yet the patent's Apple actually sues with are for stupid shit like rectangles.

That's just the (design) patent that got media coverage. The cases have involved a lot of other patents.
posted by jedicus at 3:35 PM on June 4, 2013 [1 favorite]

Oh, you mean the US and its world-dominating software industry?

I do indeed mean the world-dominating software industry that started before software patents were common or indeed permissible. The industry which flowered without widespread patents, and regarding which Bill Gates stated "If people had understood how patents would be granted when most of today’s ideas were invented, and had taken out patents, the industry would be at a complete standstill today."
posted by jaduncan at 3:37 PM on June 4, 2013 [5 favorites]

That's just the (design) patent that got media coverage. The cases have involved a lot of other patents.

Rubbish ones like "having a lock screen".
posted by Artw at 3:40 PM on June 4, 2013

Rubbish ones like "having a lock screen".

No, you mean:

A method of controlling an electronic device with a touch-sensitive display, comprising:

detecting contact with the touch-sensitive display while the device is in a user-interface lock state;
moving an unlock image along a predefined displayed path on the touch-sensitive display in accordance with the contact, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device;
transitioning the device to a user-interface unlock state if the detected contact corresponds to a predefined gesture; and
maintaining the device in the user-interface lock state if the detected contact does not correspond to the predefined gesture.

The patent covers a particular kind of lock screen, or rather the method for unlocking it. There are many perfectly serviceable alternatives, even for a touchscreen device. For example: a simple passcode or a gesture that does not involve moving an unlock image (e.g. the Google "connect the dots" approach).
posted by jedicus at 3:46 PM on June 4, 2013 [2 favorites]

Yup, having a lock screen. It's a garbage patent.
posted by Artw at 3:48 PM on June 4, 2013 [7 favorites]

I'd argue that if, on the face of it, a game-theoretic analysis of a policy regime meant to promote some particular activity would indicate that under said regime the winning strategy for a deep-pocketed rational actor would be to refrain from engaging in said wished-for activity then it ought to follow that said policy regime should be laughed all the way out of the room and into the recycle bin of dumbass policies of yesteryear unless -- and only unless! -- its advocates can muster copious evidence that despite how it looks at first glance there are higher-order dynamics that result in far more of the wished-for activity than accounted for by the first-order-only analysis.
posted by hoople at 3:48 PM on June 4, 2013 [2 favorites]

I'm not entirely sure why you're dismissing limits on software patents as so unbelievable, jedicus; the fact that Bilski v. Kappos requires the Machine-or-Transformation test but leaves open the definition of other tests for the patent courts suggests that software patents could well be limited in future.
posted by jaduncan at 3:48 PM on June 4, 2013

Also, though I'm sure trying to crush the opposition in the courts is a fine tactic I would contest that it's the source of Apples recent success - having a good product and marketing it well accounts for that, both things they could have achieved without patenting rectangles or lock screens.
posted by Artw at 3:50 PM on June 4, 2013

regarding which Bill Gates stated "If people had understood how patents would be granted when most of today’s ideas were invented, and had taken out patents, the industry would be at a complete standstill today."

The same Bill Gates that missed the significance of the internet as late as 1995? Forgive me if I have relatively little confidence in his prognostications. Gates fell prey to predictions of the always-coming-but-never-arriving software patent apocalypse. His 1991 memo, from which that quote is derived, referenced a League for Programming Freedom paper claiming that "software patents threaten to devastate America's computer industry." Looking back at the history of America's computer industry since 1991, "devastated" is not the word I would use.
posted by jedicus at 3:57 PM on June 4, 2013

Apple's 'slide to unlock' patent invalidated by German court
posted by Artw at 4:00 PM on June 4, 2013 [1 favorite]

Okay, we get it.
posted by Blazecock Pileon at 4:01 PM on June 4, 2013

the fact that Bilski v. Kappos requires the Machine-or-Transformation test but leaves open the definition of other tests for the patent courts suggests that software patents could well be limited in future.

Bilski required the MoT test? It did just the opposite. "This Court's precedents establish that the machine-or-transformation test is a useful and important clue, an investigative tool, for determining whether some claimed inventions are processes under § 101. The machine-or-transformation test is not the sole test for deciding whether an invention is a patent-eligible "process."" Bilski v. Kappos, 130 S. Ct. 3218, 3226-27 (2010). A patent could claim eligible subject matter while failing the MoT test.

I'm sure software patents could be limited in the future. Indeed, I suspect that's the likely outcome of a likely Supreme Court appeal in the CLS Bank case. I just don't think it's a good idea.
posted by jedicus at 4:03 PM on June 4, 2013

Ironmouth: " Giant companies are the ones doing the complaining--that they can't make things because entities who bought patents are suing them. Well, boo-hoo. Life is so difficult for Siemens that they cannot make money stealing other people's intellectual property. "

Are you seriously claiming that patents describe modern inventions sufficiently enough to duplicate and steal designs from?
posted by pwnguin at 4:08 PM on June 4, 2013 [2 favorites]

Yeah, the US approach to software patents has really killed the US software industry.

Back to NTP vs RIM, what's your opinion on NTP's patents, jedicus? It was a huge distraction and drain for RIM and in the end the Patent Office basically admitted the patents should never have been granted in the first place.
posted by GuyZero at 4:09 PM on June 4, 2013

Are you seriously claiming that patents describe modern inventions sufficiently enough to duplicate and steal designs from?

If they don't, then they should be invalid for failing the written description and enablement requirements:

The specification shall contain a written description of the invention, and of the manner and process of making and using it, in such full, clear, concise, and exact terms as to enable any person skilled in the art to which it pertains, or with which it is most nearly connected, to make and use the same, and shall set forth the best mode contemplated by the inventor or joint inventor of carrying out the invention.

posted by jedicus at 4:12 PM on June 4, 2013 [1 favorite]

If they don't, then they should be invalid for failing the written description and enablement requirements:

Frankly I wonder how software patents have ever gotten by on this basis. Most of they are notoriously hand-wavey with a bunch of block diagrams and never specify any specifics at all. Which is of course the objection to them because they become so broad as to encompass any minor variation on the idea.

Imagine a biotech company getting a patent on "A drug that changes your mood". I mean, that's how too many software patents read to me.
posted by GuyZero at 4:17 PM on June 4, 2013 [12 favorites]

If they don't, then they should be invalid for failing the written description and enablement requirements:

And yet companies continue to sue on invalid patents. That's practically a description of the problem.
posted by Artw at 4:17 PM on June 4, 2013 [3 favorites]

IIRC software patents are actually barred from containing implimentation details at a code level - too close to being a mathematical formula, which cannot be patented.

Useless by design.
posted by Artw at 4:19 PM on June 4, 2013 [2 favorites]

Back to NTP vs RIM, what's your opinion on NTP's patents, jedicus? It was a huge distraction and drain for RIM and in the end the Patent Office basically admitted the patents should never have been granted in the first place.

I think it's an excellent demonstration of the foolishness of the presumption of validity that we give patents based on the assumption that the Patent Office is (or even ever could be) good at its job. That presumption should be eliminated, moving the marginal case toward a finding of invalidity or not being filed in the first place.
posted by jedicus at 4:19 PM on June 4, 2013 [1 favorite]

Yes, thanks for correcting my transposition, given that it makes the second half of my comment make any sense at all. My only defence is that it's early in the morning here.

Let's come to the nub of the issue, then. Why, save for your previous statement that the US software industry is larger, do you believe that the permissive approach is better than the EPO technical effect test?
posted by jaduncan at 4:21 PM on June 4, 2013

My position is that Microsoft's dominance of several markets was enhanced by its ability to carry out its "embrace, extend, extinguish" strategy, a strategy that is broken if the competitor has a patent and so cannot be unwillingly embraced or extended.

While Microsoft certain did its fair share of this, their two primary businesses, Windows and Office, didn't really do this is any meaningful way. Microsoft anti-competitive behaviour with preferential pricing for distributors who locked out competitors was far more important to their success, no patents required. Anti-trust was another issue of course.
posted by GuyZero at 4:22 PM on June 4, 2013 [1 favorite]

And yet companies continue to sue on invalid patents.

Well, no, they sue on presumed-valid patents that either arguably should be or do in fact end up being invalidated later. Suing on an invalid patent is a one-way trip to sanctions-town.

IIRC software patents are actually barred from containing implimentation details at a code level - too close to being a mathematical formula, which cannot be patented.

You recall incorrectly. The Patent Office historically discouraged the use of pseudocode and encouraged the use of flowcharts, but it's getting slightly better about this.

Frankly a big part of the problem with software patents is that the Patent Office actively resisted them for years and thus actively resisted hiring anybody who knew anything about software. Result: a bunch of poorly examined patents.

But, as is so often the case, this is not truly unique to software patents. The Patent Office has a godawful retention rate for examiners. 80% of examiners have fewer than three years of experience, and fewer than 10% have more than 10 years. The result is that patents in all technology areas suffer.
posted by jedicus at 4:25 PM on June 4, 2013 [2 favorites]

Plus having looked at the patent examiner salary scale, no one with any competency in computer science at all is likely to take a job there. It's a joke really.
posted by GuyZero at 4:27 PM on June 4, 2013 [1 favorite]

Why, save for your previous statement that the US software industry is larger, do you believe that the permissive approach is better than the EPO technical effect test?

Because historically the US has had an extremely broad approach to patentable subject matter. Because that was the intent of the 1952 Patent Act in particular. Because software is not special in any meaningful way.
posted by jedicus at 4:27 PM on June 4, 2013

The USPTO examiners also used to have an eight hour average review time target, which led to both poor quality patents being granted and a high degree of pressure being placed on examiners to close examinations as early as possible.
posted by jaduncan at 4:29 PM on June 4, 2013

Plus having looked at the patent examiner salary scale, no one with any competency in computer science at all is likely to take a job there. It's a joke really.

Is it? Starting salary typically 60-70k, topping out at 155k, with government benefits and rapidly gaining the option for near full-time telecommuting, for a job that is pretty much strictly 40 hours a week?

I dunno, I can see the appeal, especially compared to jobs that require significant unpaid overtime and living in really expensive or crowded parts of the country. But I guess few other people do.
posted by jedicus at 4:31 PM on June 4, 2013 [3 favorites]

OK. That's really an argument from base principles. Wouldn't you agree that a generous patent office is likely to generate economic inefficiency due to litigation/settlement costs?
posted by jaduncan at 4:32 PM on June 4, 2013

Wouldn't you agree that a generous patent office is likely to generate economic inefficiency due to litigation/settlement costs?

You're assuming an awful lot with the word "generous," as though we could know beforehand that there was a correct or optimal number of patents that should be granted and that a "generous" patent office is one that grants more than that.
posted by jedicus at 4:33 PM on June 4, 2013

I'm assuming it's basically like when Homer decided to work from home and got the nodding bird to do his work for him.
posted by Artw at 4:36 PM on June 4, 2013

My memory may be of Canadian patent examiner salaries which are a lot lower. But who knows, maybe companies don't even bother to file in Canada these days.
posted by GuyZero at 4:41 PM on June 4, 2013

I am; I didn't set out specifics precisely because it's a question of principle. I think that the State Street era patent thickets around software are an example of this though; the number of patents granted and the cost of defence means that virtually all software is liable to patent suits from so many sources that prediction of legal issues/costs is impossible. If the ability to create almost any software essentially rests on lack of patent holder action, that seems problematic (and especially so when the determination of patent quality before settlement is also expensive).

It seems at the least open to a shakedown method of litigation.
posted by jaduncan at 4:53 PM on June 4, 2013

Because historically the US has had an extremely broad approach to patentable subject matter. Because that was the intent of the 1952 Patent Act in particular. Because software is not special in any meaningful way.

Again, I don't know anything about the state of patents in other industries, but if software really is as not-special as you claim, then I think you've just made a very strong case that the historical approach to patents in the US is fatally flawed and ought to be replaced with something far narrower and less generous. Software patents are utterly broken, the incentives wholly perverse. If that is the intent of the 1952 patent act then the 1952 patent act has proven itself to be a bad law.
posted by Mars Saxman at 4:55 PM on June 4, 2013 [1 favorite]

Not to derail, and hopefully the Prenda case has been linked here, but it's another example of using IP (not patents, but copyrights) for extortion. At this point, it's been revealed that the lawyers behind Prenda created their own shell companies and were actively seeding their own IP via BitTorrent. Also Previously. Their "business model" involves accusing people of downloading porn, and then making a settlement offer (or referring the accused to one of their own lawyers for defense!). Popehat has some great coverage of developments.
posted by antonymous at 5:04 PM on June 4, 2013

The transfer of patents isn't the problem. The problem is that the patents that actually get through are of abysmal quality. Take the "slide-to-unlock" patent. Imagine if a person said (in 2005): "Imagine you had a touch screen, and you could hold your finger over a button on a line and slide it all the way along that line to make a device ready to be interacted with after it was put to sleep." The slide-to-unlock patent doesn't say anything more technically interesting than that. Is that simple statement patentable?
posted by demiurge at 5:06 PM on June 4, 2013

> In the end, this is big business not wanting to pay smaller people the value of their inventions.

Spoken as someone who has zero idea as how the industry works. (Background: I have two software patents to my name.)

Nearly all software patents in question are the purest and utterest bullshit. First, they nearly all fail the condition, "There must be than one way to do it" - and separately, nearly all of them fail the condition, "Can't be obvious to a skilled practitioner."

Look at these patents, FFS! "Printing email". "Attaching a database to a webpage." "Click to buy." How is the world helped in the slightest bit by some vampire using the violence of the court system to scam hardworking people of their bread by claiming they invented "attaching a database to a webpage"?!

I am actually a small inventor - I have a couple of commercial applications I sell. If I got a software patent claim against me, no matter how bogus it was (and I've been very very careful to avoid known intellectual property(*)) I would simply have to close up shop - because I won't pay protection money, I probably couldn't afford it anyway, and I cannot possibly afford the $100K it would cost simply to get to the door of the court room. (Actually, I'd probably just move to Europe - my company only owns the binaries of my code, I own all the actual source code... :-D)

I can't imagine anyone who listened to the NPR story possibly coming back with this as a response.

(* - actually, the system is so fucked that if you're in a large company, you're actually advised NOT to do any checking for prior art at all. That's right, actually looking to see if anyone has done anything like you actually makes your patent claim weaker!)
posted by lupus_yonderboy at 5:16 PM on June 4, 2013 [6 favorites]

I think the back and forth gets really confusing for lay people...

I didn't see this addressed in the subsequent back-n-forth, but it's critically important to understand:

A patent doesn't cover what is described in the readily-understandable text portion and the pretty pictures. It covers what is in the claims.

Here is a symptom of the typical misunderstanding:

No, you mean:
A method of controlling an electronic device with a touch-sensitive display, comprising:

detecting contact with the touch-sensitive display while the device is in a user-interface lock state;
[...]
maintaining the device in the user-interface lock state if the detected contact does not correspond to the predefined gesture.

The patent covers a particular kind of lock screen, or rather the method for unlocking it. There are many perfectly serviceable alternatives, even for a touchscreen device. For example: a simple passcode or a gesture that does not involve moving an unlock image (e.g. the Google "connect the dots" approach).

> Yup, having a lock screen. It's a garbage patent.

> but you can't fix any of those problems just by saying "OK, we'll no longer allow software or business-method patents."

Why?

Forget about business method patents - suppose we overnight completely disallowed software patents. Why wouldn't it fix the problems? And what problems would it cause?

Be specific - don't just wave your hands and say, "People wouldn't innovate." I innovate all day long, and I'm scared shitless of some patent troll. Patents provide not the slightest protection for me nor for anyone I know.

In fact - name ONE software patent, just one, that you think deserves upholding. Should be easy, right?
posted by lupus_yonderboy at 5:23 PM on June 4, 2013 [2 favorites]

Well, if we consider GSM codec patents and things like the MPEG-LA pool to be "software" then yeah, I think those should be upheld.
posted by GuyZero at 5:30 PM on June 4, 2013 [1 favorite]

That it's a really wordy description of a lock screen makes it more bullshit, not less.
posted by Artw at 5:32 PM on June 4, 2013 [1 favorite]

It's not possible to give a coherent definition of a software patent. AT&T filed some patents back in the 80's, I think, that were software patents described in terms of mechanical devices. Any competent draftsman can do the same, with claims that would be infringed by a computer running a program.

I suppose you could come at it from the enforcement angle, and say "it is not possible to infringe a patent using a computer and nothing else." But what comprises a computer? Does "computer" include "mouse?" Does it include "power cord?" Does it include "ethernet cable?" Does it include "phone line?"

Point is, no matter where you draw the line, patent attorneys will just add that thing, and you're back where you started.

Re: naming ONE software patent, why not uphold the screen-lock patent Artw was griping about upthread? I don't know of any prior art on those claims, do you? Again, that patent does not cover all screen lockers. It only covers one with a "switch" image that you slide to make it unlock. (Note that the previous sentence paraphrases the claims, and, as I said a minute ago, the paraphrasing must be taken with a grain of salt.)
posted by spacewrench at 5:33 PM on June 4, 2013

Well, if we consider GSM codec patents and things like the MPEG-LA pool to be "software" then yeah, I think those should be upheld.

Interestingly, those are the sorts of software patents that veer very close to being mathematical formulae...
posted by Jimbob at 5:33 PM on June 4, 2013

No, Artw, that "really wordy description" makes the claim cover much LESS than you think it covers.
posted by spacewrench at 5:33 PM on June 4, 2013

GuyZero: Rather than just saying, "I think so" can you give us a reason why? Some facts or reasoning to support you point?

Remember, the people who create the software are protected by copyright laws; they can protect visual and naming aspects with trademark laws; and they can preserve trade secrets.

Why do they need to protect abstract concepts like "a GSM codec"? How does it benefit society as a whole? If they don't want to reveal what they're doing, they aren't required to do so.

Regarding the locking patent - this is is exactly the sort of patent that is useless.

> The patent at issue DOES NOT cover a lock screen, as jedicus points out. It covers a particular type of lock screen with a bunch of details. You can have a lock screen with any other set of operational details; the patent only prevents you from making a lock screen that is exactly described in the claim.

So how do We The People benefit from this? How is this fair, or just? Why does such a right to prevent people from creating things even exist - and why does it accrue, not to the engineer who first actually does the work, but to a lawyer who describes in extremely generic terms how it might work, if it existed?

Why is it at all fair, or useful to society, or reasonable in any way that I can do some legal magic and prevent others from creating some subset of screen lockers - particularly since I can prevent you from taking my actual work in creating such a screen locker using copyright law?
posted by lupus_yonderboy at 5:35 PM on June 4, 2013

So how do We The People benefit from this?

We don't, particularly. The theory of patents is that they provide an incentive for people to explain how to do something that otherwise wouldn't be figured out for a while yet. (If you don't offer such incentive, then people who can keep their methods a secret, will do so. Or they won't put forth the effort to commercialize them at all.)

The screen locker patent protects its owner from people copying the idea. It's more akin to copyright, but it doesn't last as long. In that case, anybody who saw the screen locker could make something like it (so the patentee can't keep his method a secret) but we offer patents to the worthy and the unworthy alike.

The thing is, you're allowed to make almost any screen locker you want, as long as it's different from the patented one. By arguing about that one particular screen locker, saying you should be allowed to make exactly that one, you're proving that that particular version is special, and therefore worth more than anything you could come up with.

So...why, again, should the person who came up with it, let you copy it?
posted by spacewrench at 5:41 PM on June 4, 2013 [1 favorite]

Why do they need to protect abstract concepts like "a GSM codec"? How does it benefit society as a whole? If they don't want to reveal what they're doing, they aren't required to do so?

I didn't mean the abstract concept but specific codecs. I am not a patent lawyer so I will no doubt stumble over some issue that was long-ago resolved by those in the field. But I fall back to the basic definition of non-obviousness. Extremely specific data compression algorithms with time or space constraints are very non-obvious. They're the kind of thing that everyone wants to copy but that inventors should be rewarded for researching, like drugs. The addition of standards bodies requiring fair and non-discriminatory licensing terms makes them pretty palatable and with these types of patents the system works the in a nearly ideal way - many companies make mobile phones, the ones that invested in the basic research to come up with the underlying standards get paid for their part of the system through royalties.

I think this is a specific example of the type of patent I'm thinking of. But it seems like a more recent one vs the ones that are presumably 30 years old at this point.
posted by GuyZero at 5:43 PM on June 4, 2013 [1 favorite]

A touch sensitive lock screen with visual feedback and if you don't correctly unlock it reverts it locked! Genius! That changes everything!
posted by Artw at 5:47 PM on June 4, 2013 [1 favorite]

> > So how do We The People benefit from this?

> We don't, particularly.

So this baffles me. If the people as a whole do not benefit from copyrights, then they should be abolished - end of story.

> The theory of patents is that they provide an incentive for people to explain how to do something that otherwise wouldn't be figured out for a while yet.

i.e. innovation would suffer - but why would it suffer?

And the screen locker is most certainly not in that category. Any bozo could figure out any number of "screen lockers" given an hour. What value are you giving to the world by "explaining" how to make a screen locker that's even vaguely comparable to the tremendous value you get from preventing all these other people from doing something that's completely obvious to everyone?

You act as if these "explanations" have value to anyone other than the "explainer" - but they do not, not for software patents.

> So...why, again, should the person who came up with it, let you copy it?

Wrong question - very wrong. The actual question is, why should the force of the law be allowed to prevent me from building new things?

Without the force of the law, why would they have to "let" me copy anything?

And no one is "copying" anything. No one sees these patents - they aren't good for anything. No one is "copying" their screen locker patent - how could they, when these trolls don't ever build a screen locker? No, their whole ploy is that they hope that NO ONE sees these patents - so they can come and take the money that your hard work made for you.
posted by lupus_yonderboy at 5:49 PM on June 4, 2013 [2 favorites]

So...why, again, should the person who came up with it, let you copy it?

Because imitation is the sincerest form of flattery?

What if the concept of a "platform game" or "first person shooter" had been patented? The world of computer games would be a dreary place. Hell, if the present software patent environment had existed 30 years ago, someone might have patented "a method for entertaining people through computer-controlled games" and we might have nothing, since every kid who came up with some shareware game to send around on floppy disk would have faced million-dollar lawsuits.

The lock-screen should be something to be built-on - designers, developers should be allowed to see it, take inspiration from it, and create something even better, the way someone hears a piece of music in a specific genre, and creates their own to expand on it. Overly-broad patents kill this true innovation, by shutting doors, by creating penalties for anyone who displays the least bit of inspiration at another's work.
posted by Jimbob at 5:50 PM on June 4, 2013 [2 favorites]

Piggybacking on lupus_yonderboy's comments, the copyright clause exists:

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.

We're at a point where we need to reduce the "limited time" in order to account for advances in computer programming and other information-based fields. How long does it take for a new "discovery" to progress science and art, and how can we encourage those who further this goal?
posted by antonymous at 5:50 PM on June 4, 2013

GuyZero: IF I were to concede the validity of any patents, it would be ones like those that were extremely non-obvious - a tiny, tiny fraction of the software patents out there.

But I still don't see why this is needed. Look at the GSM people - they designed these ideas, and then implemented them and sold the code. If they hadn't patented the ideas, they would not have had to reveal how they were done. They still would have had a massive competitive advantage, both having the technology in-house, and having first mover advantage (as well as the copyright and trademark protection too).

So why do they need this additional protection? Telecoms are making money hand over fist. Are you honestly claiming that they would not have invented GSM codecs if there weren't for software parents? Of course they would have - they wanted to make money selling these things, and they did.
posted by lupus_yonderboy at 5:55 PM on June 4, 2013

If patents are operating as copyright then they should be eliminated in favour of copyright.
posted by Artw at 5:57 PM on June 4, 2013 [1 favorite]

Hoo boy, lots to respond to...I'm hoping jedicus tags in soon...

I said: "The theory of patents is that they provide an incentive for people to explain how to do something that otherwise wouldn't be figured out for a while yet."

lupus_yonderboy: "i.e. innovation would suffer - but why would it suffer?"

Because we'd have to wait a while for somebody else to figure it out.

Jimbob: "The lock-screen should be something to be built-on - designers, developers should be allowed to see it, take inspiration from it, and create something even better"

That's the point of patents: if you convince someone to explain how to do X, then other people can build on it and do X + Y. But to convince the first guy to explain X, you have to offer him something, like the exclusive right to do X for a while. And even if you figure out X+Y, you still have to pay the X guy for a while. (Otherwise, everybody will just do X+0.000001Y and say "I don't owe X guy anything -- I'm not using his idea.)

lupus_yonderboy: "Wrong question - very wrong. The actual question is, why should the force of the law be allowed to prevent me from building new things?"

Nothing prevents you from building new things. You're merely required to respect the owners of the things you build upon, as people who build upon your work are required to respect you.

antonymous: "We're at a point where we need to reduce the "limited time" in order to account for advances in computer programming"

That's what I said.
posted by spacewrench at 5:58 PM on June 4, 2013

The patent at issue DOES NOT cover a lock screen, as jedicus points out. It covers a particular type of lock screen with a bunch of details. You can have a lock screen with any other set of operational details; the patent only prevents you from making a lock screen that is exactly described in the claim.

Sure, but the bunch of details are mostly-to-entirely the minimum set of things that anybody would need to produce a lock screen. Things like "detecting contact with the touch-sensitive display while the device is in a user-interface lock state", i.e. enabling you to interact with the lock screen. Good news for people who want to make lock screens that never unlock again - they're in the clear.

Like "while the device is in the user-interface lock state, preventing the device from performing a predefined set of actions in response to detecting any contact with the touch-sensitive display that does not correspond to the predefined gesture.", i.e. the lock screen stops your phone from doing things when it is locked, and can't be unlocked without the gesture. So a lock screen that doesn't lock the phone is still allowed.

Like "A portable electronic device, comprising: a touch-sensitive display; memory; one or more processors; and one or more modules stored in the memory and configured for execution by the one or more processors, the one or more modules including instructions to set the device to a user-interface lock state", i.e. it is limited to devices that have processors, touch screens, memory and software. Boy, that really narrows down the patent.

It's no different than GuyZero's hypothetical drug patent on "a drug that changes your mood" also having a bunch of specific details, like "the pill is taken by swallowing it through the mouth into the stomach", "the pill is smaller than your head", "the pill can be stored at room temperature" and "you take the pill fewer than 50 times a day". Sure, I suppose those details limit the scope and someone can hypothetically create a different version, like they could build a lock screen that you can't interact with, doesn't lock your phone and doesn't run on a smartphone. Oh, wait.
posted by Homeboy Trouble at 6:00 PM on June 4, 2013 [11 favorites]

Patents are required where inventors want rights but where interoperability is required - the "GSM people" are numerous companies that make handsets, base stations, chipsets, management gear, etc. In a world without patents we get oligopolies and a slower rate of change.

The worst software patents inhibit innovation while the good ones help foster it. It's just tough to build a system that knows which is which a priori.
posted by GuyZero at 6:00 PM on June 4, 2013

I agree with the broad strokes of your claims, but nevertheless I think the world of gaming would in fact be a better place if no one could make FPSes. :)
posted by You Can't Tip a Buick at 6:02 PM on June 4, 2013

That's the point of patents: if you convince someone to explain how to do X, then other people can build on it and do X + Y. But to convince the first guy to explain X, you have to offer him something, like the exclusive right to do X for a while. And even if you figure out X+Y, you still have to pay the X guy for a while. (Otherwise, everybody will just do X+0.000001Y and say "I don't owe X guy anything -- I'm not using his idea.)

Since X is obvious to all and sundry they are completly pointless then.
posted by Artw at 6:02 PM on June 4, 2013

. But to convince the first guy to explain X, you have to offer him something, like the exclusive right to do X for a while.

But, in the world of technology, how long should "a while" be? What if there had been a 20-year patent on computer games? What if Tim Berners-Lee had patented the "web server", preventing any open source projects who couldn't afford his licencing fee from developing one? Where would the internet be? Devoid of innovation, that's where. Dead. Given the current rate of progress in computing, I'm starting to feel 3-month patents on software might be pretty fair.
posted by Jimbob at 6:02 PM on June 4, 2013 [1 favorite]

Are you honestly claiming that they would not have invented GSM codecs if there weren't for software parents? Of course they would have - they wanted to make money selling these things, and they did.

No, I'm not saying that. But if there was only one company in the world making mobile phones and keeping all the details as trade secrets we wouldn't have the competition and selection of devices we have today.

As for telcos making "enough" money, that's the operators and not the equipment manufacturers. Besides, "making money" isn't a crime. I'm all for keeping costs down, but there's no moral imperative to minimize profits.
posted by GuyZero at 6:04 PM on June 4, 2013

> Sure, but the bunch of details are mostly-to-entirely the minimum set of things that anybody would need to produce a lock screen.

You are all ignoring the critical element of the claim (I'm assuming jedicus copied it correctly, I didn't go look at the source):

moving an unlock image along a predefined displayed path on the touch-sensitive display in accordance with the contact, wherein the unlock image is a graphical, interactive user-interface object with which a user interacts in order to unlock the device;

That is the red button that you have to slide over to unlock. There has to be an image displayed on the screen, and the user has to use his finger to push it along a particular path. It's also why the Android "draw a pattern" lockscreen doesn't infringe the patent. (probably. IANYPL. TINLA.)
posted by spacewrench at 6:04 PM on June 4, 2013 [1 favorite]

Maybe to be fair it should be the time it would take an outside to come up with X times 5? So in the case of software patents it would be five minutes.

Nothing of value is being preserved by these things that is not better preserved by other means.
posted by Artw at 6:05 PM on June 4, 2013 [1 favorite]

Yeah, while Apple's lockscreen patent is mostly shit, it really doesn't stop lots of other versions that accomplish the same feat. Which is sort of my objection to it - if you've patented hardly anything at all, why are you being granted a patent for it?
posted by GuyZero at 6:06 PM on June 4, 2013

why are you being granted a patent for it?

Because the Constitution says "shall be granted?" It doesn't say "may be granted if GuyZero thinks it's an awesome idea."

(Apologies for the snark, but... if the patent hardly covers anything at all, what harm to you that it's granted? And, OTOH, if it's a really slick idea, then why shouldn't the inventor make some money off it?)
posted by spacewrench at 6:08 PM on June 4, 2013

if you've patented hardly anything at all, why are you being granted a patent for it?

Do you really trust the Patent Office to decide whether or not an invention is "worth" a patent? If it's groundbreaking enough? If it's valuable enough? Does that seem like something a patent examiner (e.g. someone with a bachelor's degree in CS and about two years of experience at the PTO) would be qualified to do? Or something that the government should be deciding in the first place?

Alternatively, look at it from the other direction: if they've patented hardly anything at all, then why do we care whether they've got a patent on it?
posted by jedicus at 6:10 PM on June 4, 2013

Because the Constitution says "shall be granted?"

It doesn't say that, actually. The Patent Act is framed in positive terms (i.e. an inventor is entitled to a patent unless...) but it could be framed in negative terms (i.e. an inventor is entitled to a patent if he or she can prove...).
posted by jedicus at 6:12 PM on June 4, 2013

if the patent hardly covers anything at all, what harm to you that it's granted?

Because it becomes a tool for frivolous litigation.
posted by Artw at 6:12 PM on June 4, 2013

Mea culpa, jedicus is right. Constitution doesn't say "shall be granted," but Patent Act does say "inventor is entitled to a patent unless ..." Either way, your right to a patent is not contingent on somebody else's opinion of the worth of your invention.
posted by spacewrench at 6:16 PM on June 4, 2013

And no one is "copying" anything. No one sees these patents - they aren't good for anything.

Define "these patents." There is evidence that researchers in at least some fields (e.g. biotech) do read patents. And plenty of people read Apple's (and other tech companies') patents and patent applications to try to gain insight into where their research & development is headed.

No one is "copying" their screen locker patent

The argument is not that the lock screen is copied from the patent but that the lock screen is copied from the patented product. The patent merely serves as a public disclosure and description of what, exactly, is claimed as the invention.

- how could they, when these trolls don't ever build a screen locker?

This is confusing. First, Apple definitely sells devices that implement the lock screen patent. Second, if a "troll" doesn't make devices, then that makes people more likely to look at their patents rather than their non-existent devices, not less.

No, their whole ploy is that they hope that NO ONE sees these patents

Patent litigation is expensive and is designed to compensate, not produce a profit. Everyone, whether a company like Apple or a patent troll, would prefer to avoid litigation and instead either exclude infringers from the market or get a license. In both cases they actually want people to be aware of their patents.
posted by jedicus at 6:18 PM on June 4, 2013

actually, the system is so fucked that if you're in a large company, you're actually advised NOT to do any checking for prior art at all. That's right, actually looking to see if anyone has done anything like you actually makes your patent claim weaker!

Going into detail here would be kind of a derail and also start to smack of legal advice, but this is not true for many reasons. Lots of large companies pay lots of money for prior art searches and submit reams of prior art information to the Patent Office along with their patent applications. There are good reasons for doing this in many cases.
posted by jedicus at 6:21 PM on June 4, 2013

Lots of large companies pay lots of money for prior art searches

I think this misses the point. What you say is true, but for engineers at a company, it's often counterproductive for them to look at patents. IOW, engineers are often instructed not to check patents.

Without advising whether that's a good idea, some relevant considerations are:

Companies may be imputed to know what any employee knows.
If an employee reads a patent that is relevant to some other group's product, but the employee doesn't know it, the company may still be deemed to know of the patent (and therefore, to be willfully infringing the patent)
Same issue, but with respect to patent applications: engineer A in group 1 reads a patent that would be relevant to a patent application of engineer B in group 2. Patent doesn't concern engineer A, so he ignores it. Engineer B gets her patent, but it's held invalid for fraud on the patent office, because Company didn't disclose the patent that engineer A looked at.

posted by spacewrench at 6:29 PM on June 4, 2013

Metafilter: "the pill is smaller than your head"

Patent trolls are only a problem when you aren't the patent troll. There's no way to get rid of patent trolls, or really distinguishing what a patent troll even is, without destroying the patent system. -- ishrinkmajeans

As far as software is concerned I'd be all for it, as would most people in the software industry.

Is Apple a patent troll or Google or GE? They buy up a lot of patents and sue a lot of people. -- ishrinkmajeans

Neither apple or google are patent trolls in the strict sense because they actually make things using the patents they acquire. The legal term that's often used is "non-practicing entity"

Also, Google and most companies use their patents defensively. If a company like apple sues them, they can sue back since apple products will probably be using some of the patents that they purchased. If it weren't for the patents, they wouldn't need to have the patents in the first place. Google has actually spent billions of dollars buying the rights to cellphone related technology patents.

If large corporations eventually pass legislation barring Intellectal Ventures et al from holding patents you really haven't solved the problem now have you? -- ishrinkmajeans

There's no way to "solve the problem" of abusive lawsuits, but this will make things better. That's like saying since we'll never fully cure cancer we shouldn't even bother researching it.

You've moved from an arms race to buy patents to an arms race to buy legislators declaring your competitors patent trolls and defending your own claims. -- ishrinkmajeans

You can already buy legislators and use them to screw over your competitors. Look at the laws they're trying to pass in VA to ban Tesla Motor's sales model, for example. Changing patent will neither increase or decrease that problem.

___

Giant companies are the ones doing the complaining--that they can't make things because entities who bought patents are suing them. ... In the end, this is big business not wanting to pay smaller people the value of their inventions. By destroying the secondary market for intellectual property and inventions, they are the ones that are stifling invention. -- Ironmouth

Yeah, basically you're just lying here. There have been numerous cases of patent trolling companies suing app developers and website operators. Small developers are at a much greater risk, since they wouldn't even be able to afford the legal fight it would require to defend themselves (yes, litigation is far more expensive then basic software development)

That doesn't mean it isn't also a huge problem for huge corporations either. It's a problem for everyone except the patent trolls.

And coming up with an "invention" that only exists on paper and then waiting for someone else to come up with it and actually make it so you can sue them isn't "innovation", it creates no value for the rest of the world, which is the only reason we have a patent system in the first place.

___

What, exactly, is a software patent? Methods are (and should be) patentable. Why should it matter that a method is performed with/using/by a computer? -- spacewrench

You know how you can't patent pure math or facts about the world? In the EU the class of un-patentable things includes "programs for computers". Seems like a pretty a pretty effective definition to me.

To be even more specific you can say an object cannot infringe a patent simply due to the electronic data stored on the system ROM or other memory devices (that might also free up FPGA designs which is collateral damage I'm OK with)

[Citation Needed], but even if you can supply one, I'll bet the filing date is far enough back that it's plausible nobody had ever thought about showing email notifications on a mobile phone before. -- spacewrench

How is an email notification different then a text-message notification? How is it different then an email notification on a computer screen? Isn't a smartphone just a computer? And what's "email"? Stuff that comes over SMTP or does it include stuff like Facebook messages? What about twitter private message?

The problem is that you're basically patenting a metaphor, in fact that's literally what you're doing, the icon on the screen is a metaphor. And obviously metaphors can apply to lots of different things. Metaphorically, a smartphone is like a cellphone. metaphorically is like a personal computer. If I stick a 3G modem in a laptop with email notifications am I suddenly violating the patent? It's absurd.

The fact that someone filed paperwork on a certain metaphor in a certain context doesn't mean they should be able to sue anyone who comes up with a similar metaphor in a similar context

stating the problem is all it really takes to solve it. (E.g., "have one button on a web page that if you click it, no further activity is required to cause the order to be placed.") -- spacewrench

No, you also have to actually write the code as well.

Should we disincentivize people thinking of things like that? Make them feel that "eh, it's not worth actually doing it, since BigCo is just going to steal the idea from me." -- spacewrench

Amazon came up with the one-click shopping thing, and even Jeff Bezos thought it was ridiculous it was granted. They mainly filed it as a defensive patent, I think.

There are tons of and tons of software developers out there writing code without worrying about BigCo stealing their 'idea'. If you do something that gets a big userbase then companies will buy you out for the users and the codebase, not the idea. Anyone could re-implement Instagram pretty quickly, but Facebook bought them for a billion dollars because they wanted the users.

Not only that, but small developers actually do have to worry about patent trolls. If you want to implement a complete product, you're probably going to have to use some "metaphor" or other that some random patent troll happened to file. For example, developers were getting sued for making free versions of their apps that included an 'upgrade' button to try to sell people the full version.

(That isn't to say that all software patents are 'metaphorical' but certainly stuff like one-click shopping or email notification would be)

We don't, particularly. The theory of patents is that they provide an incentive for people to explain how to do something that otherwise wouldn't be figured out for a while yet. (If you don't offer such incentive, then people who can keep their methods a secret, will do so. Or they won't put forth the effort to commercialize them at all.) -- spacewrench

It wouldn't be possible to keep the lock screen secret. Duh. And no one was trying to commercialize a 'lock screen' they were simply trying to make it more difficult for other companies to sell phones with obvious features by filing that patent. They would still have made a lock screen if they couldn't get a patent for it.

That's the point of patents: if you convince someone to explain how to do X, then other people can build on it and do X + Y. But to convince the first guy to explain X, you have to offer him something, like the exclusive right to do X for a while. And even if you figure out X+Y, you still have to pay the X guy for a while. (Otherwise, everybody will just do X+0.000001Y and say "I don't owe X guy anything -- I'm not using his idea.) -- spacewrench

I don't know if you're not paying attention to what other people are saying or not. No one needs to look at patents to figure out how to implement software, and software developers that work at big companies aren't even allowed to look at them.

The secrecy argument with respect to software, or the idea that we are somehow "getting something" out of the patents are filed is total nonsense. It's really only an issue with industrial equipment that might be kept secret in factories or something, it's totally irrelevant to consumer technology.

People understand what the rational for patents are in general. The point is that those rationales do not make sense for software.

___

As in, how does this affect the common man and his efforts toward software production... or making apps, etc.? -- Marie Mon Dieu

Why don't you ask them? I mean it's not like there aren't software developers in this thread telling you what they think or anything...

All patents are real patents in the sense that trying to single out particular types of patents for reform is doomed to failure as a political and practical matter. -- jedicus

Change patent law so that neither can electronic data be considered infringing, nor can a device be considered infringing just because of the data stored on it either in ROM or other storage. Simple and efficient. It's not really clear why you think this wouldn't work.

Politically it just depends who spends more on lobbying.

Oh, you mean the US and its world-dominating software industry? The same software industry whose major monopoly (well known for its "embrace, extend, extinguish" business model) first encountered significant competition after changes in the case law in the 90s opened the doors to software patents, leading to the patents upon which Google was founded and upon which Apple's comeback was founded (in the form of the iPhone and iPad)? -- jedicus

Microsoft lost power because the DOJ took deliberate steps to take their power, which succeeded, just as they succeeded with IBM paving the way for Microsoft. You also seem to be implying that if it wasn't for the possibility of a patent being filed on a technology, like pagerank (or one-click shopping) the technology would not have been created, which is obviously false and probably a deliberate distortion as well.

My position is that Microsoft's dominance of several markets was enhanced by its ability to carry out its "embrace, extend, extinguish" strategy, a strategy that is broken if the competitor has a patent and so cannot be unwillingly embraced or extended. -- jedicus

It couldn't happen if the DOJ's consent decree banned them from doing it, which is what actually happened. Can you think of a single example of Microsoft being prevented from embracing and extending due to a patent? The major ones were Java, where they were sued for violating a contract and the browser, which was directly prevented by the government.

Patents are required where inventors want rights but where interoperability is required - the "GSM people" are numerous companies that make handsets, base stations, chipsets, management gear, etc. In a world without patents we get oligopolies and a slower rate of change. -- GuyZero

Most of the internet is built on un-patented yet still standardized technology. GSM standardization was mandated by the EU government. In the US we got a bunch of competing, non-interoperating standards and we still have them.
posted by delmoi at 6:36 PM on June 4, 2013 [2 favorites]

I'd be less worried about this problem if I could get a judge in the Eastern District of Texas to entertain a motion for summary judgment on patentability before Markman, but alas.

Same issue, but with respect to patent applications: engineer A in group 1 reads a patent that would be relevant to a patent application of engineer B in group 2. Patent doesn't concern engineer A, so he ignores it. Engineer B gets her patent, but it's held invalid for fraud on the patent office, because Company didn't disclose the patent that engineer A looked at.

Do you have authority that found inequitable conduct under those circumstances? Because I am not aware of any.
posted by monju_bosatsu at 6:44 PM on June 4, 2013

Also, the copyright clause to the constitution allows congress to pass laws for the purpose of securing "by securing for limited Times to Authors and Inventors the exclusive Right". It doesn't require congress to actually do it.

It's not the first amendment where congress is in fact restricted from passing laws that censor speech or establishing religions.

People in the US do not have a constitutional right to acquire patents on anything they want. Instead, Congress has a constitutional right to pass whatever patent laws it wants too. Including none at all.

Also even if you argue that congress was required to do it, they could change the law so that software patents are only valid for five minutes or something, which isn't any less absurd then copyright lasting for 130 years.

Do you really trust the Patent Office to decide whether or not an invention is "worth" a patent? If it's groundbreaking enough? If it's valuable enough?

No, I want all software patents banned (with software defined as binary data). Like I said, simple, easy to understand and implement. Solve all the problems with one simple change.
posted by delmoi at 6:47 PM on June 4, 2013 [1 favorite]

Do you have authority that found inequitable conduct under those circumstances? Because I am not aware of any.

No authority; that was the rationale I was taught when coming up as a patent associate, for why LargeCo would instruct engineers not to look at patent databases. FWIW, I think it'd be a stretch, but I wouldn't bet my malpractice policy on it.
posted by spacewrench at 6:49 PM on June 4, 2013

No, I want all software patents banned (with software defined as binary data)

I don't think there's any binary data claimed in that lockscreen patent. You may be proposing that "binary data can't infringe a patent," but does that mean I can do anything I want, as long as I have some binary data in my process?

That's the problem that people don't seem to be getting: how exactly do you define these things, and what happens at the boundaries? For all its other problems, that's one of the strengths of the legal system: there's a long history of cases that present questions right up in the grille of one boundary or another, and you can make a pretty good guess as to what will happen in other cases, based on that history.

If you make a clean sweep like you're proposing, you're also erasing all that history and forcing people to pay over again to get to the same level of estimation accuracy. Frankly, it sounds to me like a make-work program for lawyers.
posted by spacewrench at 6:54 PM on June 4, 2013

No authority; that was the rationale I was taught when coming up as a patent associate, for why LargeCo would instruct engineers not to look at patent databases. FWIW, I think it'd be a stretch, but I wouldn't bet my malpractice policy on it.

Given that intent to defraud the PTO is an element of inequitable conduct, I think you'd be okay making that bet. The Therasense opinion sets a pretty high bar; to satisfy the intent requirement, "the accused infringer must prove by clear and convincing evidence that the applicant knew of the reference, knew that it was material, and made a deliberate decision to withhold it."
posted by monju_bosatsu at 7:06 PM on June 4, 2013

The problem with slide-to-unlock is that if it's not a patent on the lockscreen itself, then it's patently (har har) obvious. If you're declaiming at the outset that you didn't invent the idea of a screen that requires some kind of touch input to return to the actual operating system, then the only "invention" there is tracing a line. Which is different from many other kinds of input you could have chosen to use, but there's a difference between choosing and inventing.
posted by Holy Zarquon's Singing Fish at 7:06 PM on June 4, 2013 [1 favorite]

spacewrench: "What you say is true, but for engineers at a company, it's often counterproductive for them to look at patents. IOW, engineers are often instructed not to check patents.

Anecdotally, I had a professor corroborate this. He claims he was talking to Microsoft engineers and they said they don't worry about patents, because "our lawyers are better."

It really seems like the current patent system favors large corporations. IBM, Apple, and Microsoft, et. al. buy up patents as a defensive portfolio and are then (relatively) free to infinge on each other because the possibility of mutually assured patent destruction is enough to deter most lawsuits. OTOH, the threat of a lawsuit is enough to force any small company to fold or settle, and free software can't afford access to the large patent pools like MPEG-LA, so the biggest winners under the current system are billion dollar companies and patent trolls.

It's interesting actually to look at software patents in the light of the Wright brothers' patent fight. After their first successes with planes the Wrights apparently they spent so much time going after patent infringement they fell behind their competitors in aircraft technology.
posted by Wemmick at 7:08 PM on June 4, 2013

IBM, Apple, and Microsoft, et. al. buy up patents as a defensive portfolio and are then (relatively) free to infinge on each other because the possibility of mutually assured patent destruction is enough to deter most lawsuits.

This might be true of relatively small stuff, but when real business is on the line these companies still sue each other. The cell phone patent wars are a great example, as is the recent wind turbine litigation in which GE obtained a $168 million jury verdict against Mitsubishi Heavy Industries.
posted by monju_bosatsu at 7:11 PM on June 4, 2013

No, I want all software patents banned (with software defined as binary data). Like I said, simple, easy to understand and implement. Solve all the problems with one simple change.

So no patents on any invention that makes use of any kind of digital circuit, even tangentially, since those circuits ultimately use at least some binary data? Or do you mean no patents that claim only binary data (e.g. claims that are written in the form "a computer readable medium comprising..." or that whose scope is coextensive with such a claim)?

Which is different from many other kinds of input you could have chosen to use, but there's a difference between choosing and inventing.

You're suggesting that there is no arrangement of user input for a phone lock screen that could possibly be nonobvious, since the putative inventor is always simply choosing from the set of possible inputs.

But what if phone lock screens used passcodes for years and years and nobody did anything different? And then suddenly Apple comes up with a different method, a method that the market reacts very positively to, a method that others try to copy? Is that not evidence that maybe it wasn't so obvious after all?
posted by jedicus at 7:14 PM on June 4, 2013

monju_bosatsu: "This might be true of relatively small stuff, but when real business is on the line these companies still sue each other. The cell phone patent wars are a great example, as is the recent wind turbine litigation in which GE obtained a $168 million jury verdict against Mitsubishi Heavy Industries."

Oh, definitely. But even as the large companies are suing each other, small ones are prevented from entering the market entirely. I doubt it's possible for a company without a multi-million dollar legal budget to enter the Western smartphone market at all currently, even if much of the tech is straightforward.
posted by Wemmick at 7:23 PM on June 4, 2013

HZSF: then the only "invention" there is tracing a line

You're paraphrasing again. The patent covers even less than "tracing a line" -- you have to drag an "unlock image" along with the tracing. You could (probably, IANYPL, TINLA) have a lockscreen exactly like Apple's, but NOT move the button across the screen along with the finger. For example, you could have the button jump across when the finger gets to the other side, or you could simply have a simulated rocker switch, and unlock when the "other" side of the switch is pressed.

I'm sorry to keep harping on this patent, but it's really not a good poster child for how awful the patent system is. There are MUCH WORSE patents out there -- ones that have clear prior art and were only granted because they were cleverly drafted and the patent attorneys were sharper than the USPTO examiners.
posted by spacewrench at 7:23 PM on June 4, 2013 [1 favorite]

You're paraphrasing again. The patent covers even less than "tracing a line" -- you have to drag an "unlock image" along with the tracing.

It's called a slider - its a pretty common UI element.
posted by Artw at 7:26 PM on June 4, 2013 [2 favorites]

The claim doesn't say "slider;" it says "unlock image." The claim probably covers a slider, but it might also cover a sprite that follows your finger, or a slider/sprite/other sort of image that you have to drag through a simple maze.

I suppose it seems nit-picky and lawyerish, but that's how patents work. And conversely, they don't work in the straightforward, "obvious" way that people think they work. It's not magical or inaccessible to laypeople, but arguing about it from a lay perspective is like arguing about whether your car runs on gas or diesel by saying that it can only run on the liquid from the pump with the green rubber hood, and will be damaged by the liquid from the pump with the blue hood. That's just not how it works, and you can imagine a mechanic getting frustrated when you insist "I know I used the green pump!"
posted by spacewrench at 7:41 PM on June 4, 2013

That's kind of the way I think of it, only in my mind it's the gas company, declaring that it didn't come up with the actual fuel, but it did invent the idea of pumping it through a green pump, and if anybody else uses a green pump they're stealing its IP.
posted by Holy Zarquon's Singing Fish at 7:48 PM on June 4, 2013

Do you really trust the Patent Office to decide whether or not an invention is "worth" a patent?

No, it was more of a philosophical question rather than strict examiner guidance. I'll settle for actual examination of prior art.

But there is certainly a cost to issuing more patents. It becomes more complex and costly for companies to assess whether or not they're infringing and in some ways devalues the system overall by reducing the value of individual patents.

Ultimately I'd like a set of rules that doesn't simultaneously get people gaming the system to extract maximum benefit with the public paying the price ultimately (also, tax law) but I do realize this is a little like complaining about thermodynamics.
posted by GuyZero at 7:48 PM on June 4, 2013

The claim doesn't say "slider;" it says "unlock image."

They put an arrow on the slider?

I'm seeing that as more of an innovation in roundabout description than UI design, TBH.
posted by Artw at 7:50 PM on June 4, 2013 [3 favorites]

I am really nonsnarkily curious: why is copying interface elements and design something the patent system should prevent? Are we really claiming that absent patents of this type, there would be less advancement in these areas?

As the much mocked round here Matt Yglesias points out, copying is how advances in consumer goods work. I really think the proper analogy here is to the fashion industry. Minimal IP protection. And yet there's lots of money in it. I claim software should basically enjoy the same status.

I also think someone who wanted to spend more braincells than I do could provide a robustly workable definition of software. But let's be honest here: there's a fair Potter Stewart aspect to that category, no? Or is there actually some disagreement about what we're talking about that I'm not seeing?
posted by PMdixon at 8:10 PM on June 4, 2013

I'm sorry, I don't know Potter Stewart. Perhaps you're alluding to something like the "Is it obscene?" -- "I know it when I see it." conundrum.

Speaking as someone who's been programming for 35 years, and doing patent work for 10, I don't think there's a good way to define software that would allow you to eliminate software patents without also causing a significant amount of collateral damage. (As I mentioned earlier, a clean-sweep approach would mess up a lot of things that people currently rely on, and cost a lot of money as people fought over where the new boundaries are.)

A better approach would be to reduce the term of a "software" patent from 20 years to, say, 7. That keeps the rest of the playing field the same, and just reduces the "tail" value of a patent. It also reduces the amount of time you'd have to wait if you were a brilliant but destitute programmer who absolutely had to use somebody else's patented invention as a building block for your own. (It also reduces the time/harm that a troll can do on the basis of some overly-broad, improvidently-granted patent.)

Reducing the term would cause the people who cared to fight over the definition of software (if they could convince a judge that their invention wasn't software, they'd get another 13 years of patent coverage). The results of those fights would help draw clearer boundaries.

And finally, if we could get more competent programmers working at the Patent Office, we wouldn't have so many crappy software patents granted. (The other side of crappy software patents is patent attorneys who are not programmers -- it's not all the examiners' fault!)
posted by spacewrench at 8:25 PM on June 4, 2013

I don't think there's a good way to define software that would allow you to eliminate software patents without also causing a significant amount of collateral damage

Didn't New Zealand just ban software patents? Don't many regimes effectively exclude software patents?

I own a small software business and I am of the impression that some of the work we do could be patented - but I think the net and gross effect of excluding software patents to me would be: more certainty, less risk, and one less measure of absurdity in the world.
posted by ~ at 9:44 PM on June 4, 2013

I am really nonsnarkily curious: why is copying interface elements and design something the patent system should prevent? Are we really claiming that absent patents of this type, there would be less advancement in these areas?

As the much mocked round here Matt Yglesias points out, copying is how advances in consumer goods work. I really think the proper analogy here is to the fashion industry. Minimal IP protection. And yet there's lots of money in it. I claim software should basically enjoy the same status. -- PMdixon

Yeah, it's essentially just interactive design, in fact it's often called interaction design or user experience design. If we allowed patents for that kind of thing back in the 80s basically all of the common 8-bit game styles could have been patented, just imagine how much that would have stifled innovation in the game industry. Imagine if Xerox had been able to patent the idea of having applications in windows or SRI had patented the hyperlink.

I also think someone who wanted to spend more braincells than I do could provide a robustly workable definition of software. -- PMdixon

As I said, binary data. That's more then just software, of course. Images, video, text and the like can be encoded as binary data, but those are protected by copyright, not patents, so that's not a problem.

If you want to be more precise you can say something like an algorithm that can be implemented on a Turing machine, but that's not as intuitive for people who aren't familiar with CS theory. Plus, you can write a program that outputs any binary string anyway, so I don't think there's any benefit to using that definition over the binary string one.
___

I don't think there's any binary data claimed in that lockscreen patent. You may be proposing that "binary data can't infringe a patent," but does that mean I can do anything I want, as long as I have some binary data in my process? -- spacewrench

That's the problem that people don't seem to be getting: how exactly do you define these things -- spacewrench

No, you don't seem to get that I did exactly, with mathematical precision define what should be considered software for the purpose of a patent. Maybe a few other things like FPGA netlists would also be excluded but I'm OK with that. (It would also exclude plans for 3D printers, which I am also OK with)

If you make a clean sweep like you're proposing, you're also erasing all that history and forcing people to pay over again to get to the same level of estimation accuracy. Frankly, it sounds to me like a make-work program for lawyers. -- spacewrench

No, people involved in pure software won't have to pay lawyers anything, which is exactly the problem I'm proposing a solution for. patent lawyers have with getting rid of software patents. It means they won't be able to continue to extract economic rents from software developers. Which is why all the "solutions" they offer involve changes to the way patents are litigated, which means you still have to litigate, and thus you still risk needing to pay them a shit-ton of money, possibly more money then a small developer even makes off of their work.

Also It wouldn't "erase history", the history books will still say how things used to be and also that they sucked.

Speaking as someone who's been programming for 35 years, and doing patent work for 10, -- spacewrench

You've been programming for 35 years and you don't know what binary data is?
___

So no patents on any invention that makes use of any kind of digital circuit, even tangentially, since those circuits ultimately use at least some binary data? Or do you mean no patents that claim only binary data (e.g. claims that are written in the form "a computer readable medium comprising..." or that whose scope is coextensive with such a claim)? -- jedicus

Are you seriously telling me you don't understand what "binary data" means? It's an unbelievably straightforward concept. A digital circuit that contains some binary data is not itself binary data, just like a car that contains gasoline is not itself made entirely out of gasoline. Therefore, the answer is no, the set of things that are "binary data" does not include digital circuits any more then the set of things made out of gasoline does not contain cars.

(Again as I said, the data for an FPGA layout for that circuit would qualify, though. But an ASIC version might violate the patent)

___

Didn't New Zealand just ban software patents? Don't many regimes effectively exclude software patents? -- ~

looks like it,

There law basically just says:

(1) A computer program is not an invention and not a manner of manufacture for the purposes of this Act.
(2) Subsection (1) prevents anything from being an invention or a manner of manufacture for the purposes of this Act only to the extent that a claim in a patent or an application relates to a computer program as such.
(3) A claim in a patent or an application relates to a computer program as such if the actual contribution made by the alleged invention lies solely in it being a computer program.

That's similar to the EU which excludes: "schemes, rules and methods for performing mental acts, playing games or doing business, and programs for computers; "

So it's not like this is some unworkable incomprehensible thing and no one can figure out what "computer program" or "programs for computers" means. I think in general in these countries you can patent specific devices like washing machines that rely on some software to run, but it wouldn't cover pure software running on general purpose computer, which is what a smartphone is.

___
Anyway my idea of banning software patents by considering anything composed of binary data not to be a patent violation (along with just not accepting software patents at the patent office) seems like it would quickly and easily solve the problem. I haven't heard any criticism besides "That's too complicated to figure out" (When in fact it's mathematically precise and quite straightforward) or "What about X or Y" when it's actually quite easy to tell if X or Y qualify because you just check to see if they fit the mathematical definition of a binary string.
posted by delmoi at 10:09 PM on June 4, 2013 [1 favorite]

spacewrench: do you believe that to a practitioner reasonably competent in the field that this slide-to-unlock patent discloses any information that, had it not been so disclosed, would have left said practitioner faced with material difficulty in replicating said patented mechanism?

If you do, would you perhaps point out which element(s) or combination(s) of element(s) enumerated within this patent you believe the patent elucidates their working above and beyond that which could be easily inferred by a cursory examination of the same mechanism as embodied in any of the products making use of it?
posted by hoople at 10:15 PM on June 4, 2013 [2 favorites]

It also reduces the amount of time you'd have to wait if you were a brilliant but destitute programmer who absolutely had to use somebody else's patented invention as a building block for your own

Yeah, but that's not what happens. Every normal programmer routinely reinvents things which have been patented, as a matter of everyday life, and simply never knows about it. Companies acquire patent portfolios not because they expect to learn anything useful from them - they won't - but because they can go through those portfolios and find things that their competitors are doing and then threaten to sue for patent infringement. The programmers at the competing company who wrote that code never knew about the patent either; they just solved the problem in front of them in a sensible way and moved on. At this point it goes one of two ways: the target company either rolls over and pays the patent-holder to go away, deriving zero benefit from the "license" to the patent, or the target company looks through their own portfolio of bullshit patents, finds something they can plausibly accuse their attacker of infringing, and threatens to countersue. Then the lawyers work it out.

At no point in this process does any programmer ever benefit from the creation of these patents. At no point does any programmer even know about any of them, other than the bullshit patents they themselves are required to waste time helping to create, not to educate other programmers (that happens through completely different channels which have nothing to do with patents) but in order to bolster their own company's portfolio.

Writing your invention into a software patent is a great way to ensure that all the best minds of the industry ignore it as thoroughly as possible. If anyone ever implements your idea, it is because they have independently reinvented it, which demonstrates that it wasn't such a valuable idea in the first place.
posted by Mars Saxman at 10:25 PM on June 4, 2013 [6 favorites]

No, as I said above, (and although delmoi disagreed) for many software inventions, simply stating the problem is adequate to solve it. But "difficulty of implementation" is not a criterion for patentability in the U.S. Instead, it's novelty -- that is, the fact that nobody else has solved the problem in that way before. That makes sense if you think about it: you want people to come up with simple solutions. They're inexpensive, they're reliable, and after they go off patent, they're unquestionably a public good.

Further considering the "difficulty of implementation" idea shows that it's really unworkable. You obviously think (and I agree) that any competent coder could implement slide-to-unlock after seeing it in action once. But you also have to agree that there are plenty of incompetent coders who would struggle mightily with that task. Should we give those guys a patent, but deny it to the people who can do it in an afternoon?

At any rate, as I've said before, the screen-lock patent is a really bad poster child for stupid software patents. As far as I know, it is novel. There are easy, viable workarounds (I don't know whether they're patented, though). But most of the complaints seem to boil down to "I'm offended that I can't use that guy's good idea for free, and instead, I must come up with some other idea of my own."
posted by spacewrench at 10:37 PM on June 4, 2013

But most of the complaints seem to boil down to "I'm offended that I can't use that guy's good idea for free, and instead, I must come up with some other idea of my own."

You seem to think that ideas have value, in the context of software development, and they simply don't. Implementations have value, and are adequately protected by copyright. Patents merely prevent other people from doing their own work to create their own implementations.

But "difficulty of implementation" is not a criterion for patentability in the U.S. Instead, it's novelty -- that is, the fact that nobody else has solved the problem in that way before. That makes sense if you think about it: you want people to come up with simple solutions. They're inexpensive, they're reliable, and after they go off patent, they're unquestionably a public good.

This makes a pretty good case for abolishing software patents, whose chief effect is to encourage people to develop complex, indirect, baroque solutions for straightforward problems not because it creates any value for anyone but merely to avoid other people's useless bullshit patents.
posted by Mars Saxman at 10:41 PM on June 4, 2013 [1 favorite]

Binary code is a mathematically precise concept, it's information that can be encoded as a string of bits. Ones and zeros. What I am saying is that those strings of bits themselves should not be able to be patented, and thus a file you download from the internet, get on a disk or USB stick or type in from a book couldn't violate a patent.

As someone who basically agrees with where you're coming from, delmoi, I have to say that you're way off base here. The problem isn't that people are patenting actual strings of information (and why binary? Why does the base matter?); the problem is that they are patenting abstract concepts.

If we allowed patents for that kind of thing back in the 80s basically all of the common 8-bit game styles could have been patented, just imagine how much that would have stifled innovation in the game industry. Imagine if Xerox had been able to patent the idea of having applications in windows or SRI had patented the hyperlink.

This is so very true.

But "difficulty of implementation" is not a criterion for patentability in the U.S. Instead, it's novelty -- that is, the fact that nobody else has solved the problem in that way before.

And non-obviousness. I think that this criterion is essentially the same as "difficulty of implementation", and it is (or is supposed to be) the status quo. Many obvious ideas are patented, however.

But you also have to agree that there are plenty of incompetent coders who would struggle mightily with that task. Should we give those guys a patent, but deny it to the people who can do it in an afternoon?

The standard is "ordinary skill in the art". I think this implies that the implementation must be non-obvious to a competent practitioner of the art. To have ordinary skill implies at least competence, no?
posted by mr_roboto at 10:47 PM on June 4, 2013

If anyone ever implements your idea, it is because they have independently reinvented it, which demonstrates that it wasn't such a valuable idea in the first place.

Actually, it demonstrates that it was a valuable idea, but arguably obvious (which is a criterion for rejecting a claim). If we had competent programmers at the USPTO, then there's a slightly improved chance that such patents would be rejected. As it is, though, Examiners must find that somebody else described the code before the patent applicant did.

Note that I said described the code, not wrote the code. That's because patent prosecution is all done in English. Even if somebody wrote the code before the patent applicant, we'd need an Examiner who could find that previous implementation, understand that it was the same as the claimed method, and convince a non-programmer lawyer that that was the case.

One thing programmers can do to help prevent bad patents is to describe -- in English -- ideas and ways of doing things. Publish those online somewhere; make sure it's easy to determine the date they went up. An Examiner doesn't begin looking at most patent claims until 3-4 years after they were filed, and his job is to find published information from before the filing that explains all the aspects of the claims. If he can't find anything, there's little basis for rejecting the patent.
posted by spacewrench at 10:48 PM on June 4, 2013

Re: obviousness, the standard in patent law is tricky for competent programmers to understand. It's not actually "what I can figure out," or even "what one of ordinary skill can figure out." It's closer to "what one can put together via rote combination of materials that were published before the patent application was filed." Again, it's years between filing and examination, and the system is very careful to avoid "impermissible hindsight."

Imagine you invented something -- say, some mechanical gadget that is simple, but unquestionably new and very very useful. You file a patent, but by the time it gets examined (3 years later), everybody in the world has one of those gadgets in their pocket. Kids have grown up with them.

It's not fair for the examiner to look at what people know about the invention when it's being examined -- by that time, for a good invention, of course it's obvious. The question is, if you went back in time and looked at what was available then, would "one of ordinary skill" have been able to make the invention? And the fact that for all of recorded history before then, nobody did it, is pretty strong evidence that it wasn't, in fact, obvious.

Well, it's the same thing for software, except that programmers tend to write code instead of English descriptions of what the code would do if somebody wrote it, and that Examiners are not coders to begin with.
posted by spacewrench at 10:56 PM on June 4, 2013

As someone who basically agrees with where you're coming from, delmoi, I have to say that you're way off base here. The problem isn't that people are patenting actual strings of information (and why binary? Why does the base matter?); the problem is that they are patenting abstract concepts.

My idea is that binary strings couldn't considered to be infringing on a patent. So lets say for example, someone creates a hardware chip that implements a lock screen, and gets a patent for it. If I write a program to implement a lock screen, then that program could not be considered infringing. The patent holder couldn't sue.

If I sell my own phone with my own software lock screen the phone itself wouldn't infringe because the lock screen would just be binary data stored on a flash chip. Or the phone could download the code when it was first activated.

Base-2 because that's all anyone uses. Stuff like base-64 encoded files are actually still encoded as binary strings, with a restricted set of 8-bit substrings. I suppose it's a "loophole" if someone builds a real, working tinary or quaternary computer and then patents the software for it, but it wouldn't stop anyone from implementing the same algorithms on a binary machine.

Re: obviousness, the standard in patent law is tricky for competent programmers to understand. It's not actually "what I can figure out," or even "what one of ordinary skill can figure out." It's closer to "what one can put together via rote combination of materials that were published before the patent application was filed." Again, it's years between filing and examination, and the system is very careful to avoid "impermissible hindsight."

One easy way to simplify that would be to ban software patents.
posted by delmoi at 11:07 PM on June 4, 2013 [3 favorites]

Patent litigation is expensive and is designed to compensate, not produce a profit. Everyone, whether a company like Apple or a patent troll, would prefer to avoid litigation and instead either exclude infringers from the market or get a license. In both cases they actually want people to be aware of their patents.

I'm just going to say that this rosy view may not explain the popularity of submarine patents.
posted by jaduncan at 11:30 PM on June 4, 2013

At any rate, as I've said before, the screen-lock patent is a really bad poster child for stupid software patents. As far as I know, it is novel.

It's kind of an obvious solution once you have a touch screen device that needs to be locked.
posted by empath at 12:19 AM on June 5, 2013

But most of the complaints seem to boil down to "I'm offended that I can't use that guy's good idea for free, and instead, I must come up with some other idea of my own."

Here are some alternatives: I'm offended that a social structure that is intended to spread practical knowledge is being used to stifle it instead. Really, this is the main thing. If a body of law is supposed to prevent X and it instead encourages X, the law needs to be changed. And the same is true in reverse.

I'm offended that end users are now being targeted, like podcasters who violate an ancient magazine-on-cassette patent. Wait until next year when someone with a 19 year old patent on entering text through a television remote control starts firing lawyers at everyone who has ever typed a comment into a box on the internet. And I'm offended by the implication that people should be expected to do patent searches or pay license fees merely to use software.

I'm offended that the tech giants are spending huge amounts of money and wasting the talents of smart lawyers and programmers who could be doing something, anything, more productive. Even if it just means buying more foosball tables or evading more taxes in Ireland

I'm offended by lawyers who think the solution to a bad patent system is more litigation, with the exception of the ones who subsequently dedicate their careers to working IP cases pro bono.

I'm offended that if I solve a problem independently, someone who has come up with the same straightforward and obvious solution gets to shake me down with the Hobson's choice of paying huge license fees or paying huge legal fees (and maybe following it up with huge license fees).

And I'[m also offended that people who hate that a broken system is threatening to strangle innovation in an important industry are portrayed as freeloaders.
posted by Homeboy Trouble at 12:44 AM on June 5, 2013 [5 favorites]

Marie Mon Dieu: I used to work for attorneys in a patent office of a big company and they were the best bosses in the world.

Irrelevant. Heck, many crime bosses are personable and even friendly in person, and if they have nothing against you -- it doesn't mean that they can't be ruthless killers if the situation demands it. It is not always the case that evil stinks on first whiff -- it might even require some careful examination and knowledge of the situation.

Ironmouth: Big businesses are the ones who are claiming to be "victims" here.

This is strange framing. You speak as if "big business" were a monolitic category! But no. Many people think the public are collective victims. Just because some corporate wills are on the right side of the issue doesn't mean we should flee from that opinion like Boris Badenov grousing about sunny weather. And there are plenty of big businesses who have taken advantage of the patent system themselves.

Frayed Knot: I'm thrilled people are starting to notice this is a problem.

I'm depressed that it's taken so goddamn long for Important People to realize it, when the internet at large has known how terrible the situation is ever since Unisys decided to enforce its patent on the simple compression used in the GIF file format, which pissed a lot of people right the hell off. There are many of these intractable problems that most people agree are bad yet it seems impossible to align any kind of political will behind fixing them. At least Obama's doing something, finally!

jedicus: What matters is the claims. The claims all require the text-to-speech and voicemail stuff.

Now you see, in my view, claiming something you can do should not be patentable. If software is to be patented at all, I'd say it's precise implementations, that is to say explicit code or at least algorithms, that should be covered.

The same software industry whose major monopoly (well known for its "embrace, extend, extinguish" business model) first encountered significant competition after changes in the case law in the 90s opened the doors to software patents

That's circumstantial. It's the rise of the web that ultimately challenged Microsoft, by switching everyone's primary use for computers from application software to web browsing, and that was carried on the back of open source software and principles of sharing and free communication, not software patents.

My position is that Microsoft's dominance of several markets was enhanced by its ability to carry out its "embrace, extend, extinguish" strategy, a strategy that is broken if the competitor has a patent and so cannot be unwillingly embraced or extended.

A world in which patents are used to avoid outside extensions to technology is not the happy place you imagine it to be. It is one in which the owner of the patent has a means to control all use of their tech, even once it's entered into user hands. To think I live in an age in which people have become so accustomed to freedom that they unironically argue for less of it.

A method of controlling an electronic device with a touch-sensitive display, comprising:
details snipped to save space

Exactly. HAVING A LOCK SCREEN. If it doesn't state precise algorithms the patent should not have been awarded. Those words might seem specific to you; they are not to anyone who knows a damn thing about programming. Words like "wherein" don't belong in patent applications. IMO, of course.

The purpose of the patent system is "To promote the Progress of Science and useful Arts." To publicly register the details of an invention, for everyone to read and benefit from, in exchange for a period of exclusive use. Nothing in that description isn't already obvious to someone who uses the device! If you're going to be given exclusive use of something, you damn well should be giving us something useful in exchange goddammit!
posted by JHarris at 1:30 AM on June 5, 2013 [3 favorites]

Oh from that "apple betrayed by it's own patent lawyers" article:

The following year, FlatWorld got its second deal—and what appears to be its last—doing a large touchscreen installation for the Philadelphia Zoo. Milekic had never written a line of code or taken a programming class. He told a Philly tech publication that his touchscreen software was built with a program called RunTime Revolution, which says it "makes programming learnable by anyone who can use a computer."

“Since [Milekic] is a psychologist, he really designs for children,” McAleese said in the 2009 interview. “There’s no directions or instruction, people just know what to do because it’s simple and intuitive.”

As intuitive as it might have been, the company only finalized two deals in two and a half years. It never launched ShowMe Tools, a program it said was 80 percent done in 2008. So McAleese and Milekic doubled down on that potentially more lucrative project: who would help them sue Apple?

Hahah. but I guess it's totally legit because according to spacewrench "simply stating the problem is adequate to solve it". No actual code needed or anything.
posted by delmoi at 1:36 AM on June 5, 2013

It's the rise of the web that ultimately challenged Microsoft, by switching everyone's primary use for computers from application software to web browsing, and that was carried on the back of open source software and principles of sharing and free communication, not software patents.

This is really the crux of it. You've got two sides shouting past each other.

Side A says innovation is encouraged by the economic incentives of patents for inventors and/or people (aka, corporations, my friend) who manage to buy patents.

Side B points out that, in fact, that economic system to encourage innovation is actually drastically inhibiting innovation, and that in fact what's been driving innovation in tech is a culture of sharing (open source movement) and need to interconnect complex software and devices (API's).

Then Side A covers its ears and screams that the patent system is supposed to encourage innovation by providing economic incentives to inventors and/or the people who manage to buy patents.

Side B believes it has concrete proof that Side A's argument is incorrect, and Side A just isn't listening. I think it's telling in the original 'When Patents Attack!' that they're unable to find a single actual inventor who's actually benefited from software patents. The system is broken, and the lawyers in the discussion - being creatures of rules - seem unable to think beyond the confines of the current legal arrangement to envision something that might work better. It's quite sad.

For my part, I absolutely agree that the software patent situation is a MASSIVE disincentive to start working on tech business ideas. I have piles of ideas that I could turn into working reality, but it looks to me like it's fucking stupid to get involved in building new software without having a raft of lawyers on my side. The system is broken, and the lawyers in the conversation need to start realizing that.
posted by kaibutsu at 5:10 AM on June 5, 2013 [2 favorites]

Lodsys just took Rovio for an undeclared settlement. I'm sure it's not an issue for small app developers if even Rovio won't fight.
posted by jaduncan at 6:38 AM on June 5, 2013

The solution is simple: In addition to novelty, utility, and nonobviousness there just has to be the additional requirement that granting the patent would "promote the progress of science and useful arts" as stated explicitly in Art I s. 8 of the Constitution.

Everyone can agree on that, surely?

{hamburger}
posted by whuppy at 12:09 PM on June 5, 2013

Homeboy Trouble: "the lock screen stops your phone from doing things when it is locked, and can't be unlocked without the gesture. So a lock screen that doesn't lock the phone is still allowed."

Jesus it just occurred to me that the stupidest misfeature of my phone (that the keyboard lock disables everything but the power and volume control buttons) might be a patent avoidance workaround. In some ways I hope that is true rather than being the result of the second most idiotic design decision I can think of right after putting the high beam dip switch on a stalk.
posted by Mitheral at 9:22 PM on June 6, 2013

« Older "But that would be completely wasted on you, so I... | Damn it, Phill, I'm a writer, not a decorator! Newer »

This thread has been archived and is closed to new comments

MetaFilter

Trolling just got a bit tougher
June 4, 2013 12:51 PM Subscribe

Tags

Share

Trolling just got a bit tougher June 4, 2013 12:51 PM Subscribe

Tags

Share

Trolling just got a bit tougher
June 4, 2013 12:51 PM Subscribe