Hacker News Reader

Comment by albertsun | Hacker News Reader

Link to a deeper analysis http://fosspatents.blogspot.com/2011/07/these-tables-show-ho...

Claims using regexes to match patterns in a piece of text is infringing.

    By way of example, the functionality within the Nexus
    One includes Android's "Linkify" functionality, which
    "take[s] a piece of text and a regular expression and
    turns all of the regex matches in the text into
    clickable links.

View on HN

That's not what the patent claims. As usual, we are going to burn many tens of messages here debating patents without truly understanding how patent claim matching works.

In this case, Florian Mueller is using examples of Android functionality to illustrate individual elements of the body of a single independent claim. You've excerpted his example without context, leading a reader here to believe that Apple is patenting random Perl features.

I read the patent and it seems that using regex to automate links on a mobile device is exactly what they patented. Can you elaborate why this is not so?

Patent: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sec...

The word "comprising†" in a patent claim means "including but not limited to". In this case, the initial independent claim of the patent --- the broadest in the whole document, the one least likely to be at issue --- comprises:

* an input device

* an output device

* an associated memory

* a server

* a UI

* an "action processor", and

* a processing unit coupled to the input device, the output device, and the memory.

Did you build that? No? Then what are you talking about?

And please, please, please: before we open up Pandora's shipping crate full of geeky arguments about how an end table, a sofa cushion, and an abacus could also be construed as this particular claims construct, can we also acknowledge that we don't know exactly which claim in the document was found binding on HTC? The ITC ruling isn't, as far as I can tell, published on the web. The reason patents have broad independent claims followed by strings of specific dependent claims is that the dependent claims are the ones usually found to be enforceable.

† thanks for the correction

a processing unit coupled to the input device, the output device, and the memory.

So in other words... a computer?

A huge amount of modern software patent language is simply generic, copy-paste babble to say, in long and drawn-out terms, "on a computer". This doesn't make the patent any less stupid.

I repeat myself here for purposes of emphasis:

We don't know which of the claims of this patent bound on HTC.

The initial independent claims of patents are deliberately written to be as broad as possible, to insulate the patent from technological changes. The IP lawyers who draft Apple's patents go way out of their way to make sure that first anchoring independent claim is as general as possible.

A reasonable expectation is that it's not simply the initial independent claim --- which, by design, is the claim mostly likely to be challenged as over-broad --- but a combination of independent and dependent claims that actually bound HTC.

Given that, I'm super not interested in discussing whether a court is likely to find that any computer ever built running Linux violates this patent. I'm strongly inclined to believe that no, your infringing product needs to embody a purpose-built device.

I think your logic about "huge amounts of modern software patent language" pertains to the patent description, not the claims.

Do you believe that Google and/or HTC used this particular patent to get the idea or work out how to implement it on their device(s)?

Does the answer to the above question affect how you feel about the patent (and/or patents in general)?

used this particular patent to get the idea or work out how to implement it

A patent protects the invention, not the text of the patent. The more realistic question is whether HTC's implementation was worked out by looking at how the feature works on iOS.

I'd bet it wasn't. I saw this sort of thing shown off as one of the advantages of Android's Intent system (e.g. highlighting a URL or phone number in a note-taking app) at a developer event months before the G1's release.

So far as I understand, iOS doesn't have the same sort of Intent architecture, so it seems unlikely that that Google's implementation drew any inspiration from iOS (especially as the same sort of thing has been done in web browsers for years to the point where a browser might be prior art).

No, I do not believe that Google benefited in any way from Apple's disclosure of their '96 invention.

Generally, I think software patents are as bad as we tend to say they are.

However, caricaturing their badness with statements like "Apple patented regular expressions on URLs" or whatever sucks power away from the argument that they should be abolished, by allowing the "grown-ups" to point and say that opponents of patents think that e.g. this one says you can't regex URLs.

An HTC Android device is a computer running Linux, Android devices are as much multipurpose computing devices as computers.

But regardless of that, having read the 21 claims of the '647 patent (I don't fully understand the other one, but this is the one that covers what this thread is discussing), there's nothing about it being tied to purpose built hardware.

The example implementation described in the patent itself uses a Power Macintosh, the only part that refers to purpose built computers is "For example, program 165 in FIG. 1 can be stored in ROM, disk, or in dedicated hardware. In fact, it may be realized as a separate electronic circuit."

The term "a processing unit coupled to the input device, the output device" is only being used as a generic way of describing a computing device (hence the Macintosh being used in the example and the reference to dedicated hardware is only made as an alternative method of implementation)

If this patent is valid, it's likely covers a wide range of computer application including those found on desktop computers (for example Skype's phone number highlighting plugin).

Once again I read the patent and it's safe to say that any mobile device (input/output) with a CPU ("action processor") that uses regex to parse text and automatically turn it into actionable links is infringing this patent.

Completely agree. This is exactly what the patent covers: the automated hyperlink-ization, using regex, of discrete pieces of data ranging from URLs to phone numbers, on mobile devices. Which is an obvious improvement any competent programmer could come up with, and in no way the sort of "invention" patents were intended to protect.

The thing that gets me, though is that the language used is so general. It says things like 'input device', 'output device'. My desktop has an 'input device': the keyboard and an 'output device': the monitor.

Also it talks about detecting structures in a grammar. That has been done before. So putting that on a server and accessing it with input and output devices makes it new and original? Surely an old-style unix terminal running awk on a PL/1 could fit in this broad terminology.

Could you please explain what this patent entails then?

Too many times discussions on here have been shortcircuited by claims that the complainant just doesn't understand patents. The discussion fizzles and the absurdist situation with software patents continues.

I've read this patent in detail, as gruelling a task as that is, and it does seem to be as trivial as others are suggesting (as is the case for most trivial patents that get criticized on here). The claims that detail, essentially, a computer are boilerplate in 99% of software patents. They are not particularly relevant and add little specificity.

EDIT: Another submitter has posted a listing of Apple's specific complaints, and it is exactly as others are suggesting -- Regex'ing input and identifying types of text.

If only geeks could debate patents more effectively on HN, your thinking goes, we wouldn't be stricken with software patents?

To a degree, yes absolutely. The overwhelming majority of software developers, I suspect (invented stat), are against software patents, and the few who aren't suffer the illusion that deep in the obtuse language of patents there really is something novel...because they've been told such in so many internet discussions (that usually follow exactly the course of this one). I don't think that is true, and the illusion that there is something deeper is purely because we seldom talk about the specifics because patents are so horrendously worded, with so much boilerplate.

Thankfully Apple itself has provided such an interpretation, and their interpretation, worded to best be in their favor, is exactly as trivial as many have suspected.

This is a critically important discussion for this industry.

Do you really think that the opinion of people on HN has a measurable impact on our patent system?

Getting geeks to dislike patents is easy. We have a lot of geeks who hate patents and are passionate about it. Yet the patent system hasn't been changing as a result. Why not?

"Getting geeks to dislike patents is easy."

Isn't that kind of strange? Aren't we the ones who most benefit from patents?

"Do you really think that the opinion of people on HN has a measurable impact on our patent system?"

Yes I absolutely do. In fact I think the next 24 months will see a lot of debate and eventually significant change to software patents in the US. All I care about are defusing those who attempt to make everyone feel too ignorant to have an opinion about patents (such attempts appear in every discussion about patents. Lots of hand waving about how we all just don't get it, leave it for the big boys, etc)

Aren't we the ones who most benefit from patents?

No. The two groups that benefit most directly from patents are established companies which wish to block competitors, and patent trolls which wish to sue lots of people.

For the kind of startups that HN people are interested in, patents are an exercise of interest mostly because of their potential value to investors and purchasers. But in the meantime they cost money, don't affect sales one bit, and you likely won't get them until years after it is obvious whether you've succeeded or failed as a business.

Until the big boys decide that the cost of patent trolls exceeds the benefit of their patent moats, nothing will change about patent law. In particular I see no sign that the next 24 months is critical. In fact I'm willing to bet you money on that. The next 24 months is mostly going to be spent with a Republican Congress (who see patents as "business friendly" and have bigger political fish to fry) fighting with Obama. After that we'll have a new administration, which is going to focus on top political stuff for their first 100 days.

My personal suspicion is that the economy is going to go south again, credit problems are coming back in a big way, and financial/economic issues are going to keep software patents off of the list of priorities for a long time.

Do you really think that the opinion of people on HN has a measurable impact on our patent system?

Getting geeks to dislike patents is easy. We have a lot of geeks who hate patents and are passionate about it. Yet the patent system hasn't been changing as a result. Why not?

If you can't get geeks, who should all have at least some distaste for software patents, to agree on a solid argument against them, then how can you expect to convince congress? One could view these geeks-only debates as individual geeks refining their arguments and accumulating additional information, in preparation for attacking the problem at its legislative source.

In this case, it looks like Apple is looking to fire the cheapest missiles that have a chance of hurting. The patent does indeed seem to cover the packaging of (what used to be called in OS9 days) "apple data detectors" with a computer. IIRC, Steve Jobs, in his iPhone announcement keynote speech, said that they've filed about 200 patents. This shows how serious they are about not only building these path breaking things, but also about protecting their inventions. I haven't checked, but it seems a bit unlikely that those 200 applications (or the late part of them) haven't yet been granted.

The situation, in other words, looks like a ticking time bomb for the other phones looking to copy the iphone - android or whatever.

Is there anyone who's analyzed apple potential patent portfolio in the mobile space?

From Florian's article (linked from this HN post) "1, 8, 15 and 19"

http://fosspatents.blogspot.com/2011/07/these-tables-show-ho...

The patent: http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sec...

It's worth pointing out that the term "Analyzer server" isn't used in the sense most of us would understand the term server now, but rather referring to an internal software component that's an inherent part of the software rather than an external application.

I understand that it's not single claims that lead to a violation but a match against many of the claims. Still, if the component claims are things like "using regular expressions to create links" and "using an output device to present data" then matching becomes a lot easier.

This isn't even a single claim! It's one element of the body of a claim extracted without context. Geeks have lacerating ignorance of how patents actually work. And fine, why should they understand them? Except that now we get to have our BiMonPatentFreakoutCon on HN about them.

At the end of the day though all software patents are bogus, trivial, and stupid. (And I'm listed as inventor on a dozen).

We should spend energy abolishing them rather than debating how they work.

just my 2c

I agree with this, but, as a nerd, I have a hard time letting go of people who construct arguments like the one that roots this comment thread.

You can start by just saying "No, thanks," to those $1000 patent bonuses at your company.

That's nice. You'll be out the $1000, and with some likelihood you'll still be forced to help file a patent; the boilerplate IP clause in many (most?) California employment contracts requires patent support, as I found out the hard way 11 years ago.

I feel for you dude. Keep fighting the good fight, but you're probably wasting your breath.

This is a continuing source of confusion for me, how a group of people who generally can't rest until they understand something can be so willfully ignorant about patents.

Good fight? I'm fairly sure at this point Ptacek realizes that he was wrong. So wrong that Apple's own self-serving explanation of the patent is identical to the root explanation that caused so much of this discussion.

You insult this community while offering absolutely nothing to counter any complaints about this patent except for, as others have tried, vagaries. Patents are public, so if you've actually got something, let's see it. Otherwise it's just noise.

Wait, where did he say anything that was wrong? He's the only one in this clusterfuck that seems to know what he's talking about.

Where he suggested that people who were accurately describing the patent were wrong because they were...you know....missing a claim or something. Yet those interpretations were entirely right, and are mirrored by Apple's own interpretation of their patent.

Simply responding to every patent comment with vagaries and innuendo is not enlightened or helpful. It's unhelpful noise meant to short-circuit discussions about the very, very broken patent system.

Do you have a link to a specific comment? Because all I remember is people saying "this patent covers x", where "x" is only a single element of a claim. He was right to tell them that they were missing something. If you don't have every element of a claim, you don't infringe the claim.

His very first message in this discussion!

http://news.ycombinator.com/item?id=2777338

That _IS_ essentially what the patent claims. Ptacek tried to add specificity by pointing out that Apple essentially added the disclaimer "on a computer", which is of marginal usefulness.

His original response was wrong, and was the classic alluding to some deeper (but never stated) meaning to the patent. There is no deeper meaning. That is it.

No, he was right. The comment he replied to was discussing a single element of the claim, and implying that anything that does just that element is enough to infringe.

What the patent claims is a device that has every element of a claim. Not just one element, or some elements. If it doesn't have every element, it's not an infringement.

No he was not right. The comment he replied to quoted an example of an infringing implementation based up all claims and dependencies in the patent. It was 100% correct, and remains 100% correct. Someone didn't cherry pick one elemtn of a claim and that was it, they found a specific infrginging activity, and having read the patent in detail, they are dead on.

More handwaving, however. This is why we can't have nice things.

You are wrong. Ptacek is wrong. AND APPLE AGREES. How is this nonsensical side discussion still occurring? The specific example is one that Apple used, so...what is going on here?

Repeating yourself doesn't make you right. Actually go read that fosspatents article. Then read the comment that links to it. The article discusses entire claims, but the comment focuses on a single element that's discussed in the article. Apple is not claiming that matching a regex infringes their patent.

You understand incorrectly. Infringing a single claim is an infringement. However, you must have all the elements of that claim to infringe. The claims are not things like "using an output device to present data".

I love the "an output device for presenting the data" item.

Your comments on this issue strike me as disingenuous.

By my understanding, the patent claims the (obvious, with mountains of prior art) behavior of auto-generating hyperlinks from text using regular expressions, so that (for example) when you receive a text message with a phone number in it, that phone number can be detected with regex and displayed as a hyperlink. When you click it, it can then say "Call, Save To Contacts" etc.

That's just one of the supposed infringements. Your comments here make it sound like this is a legitimate patent, and that HN geeks are simply too stupid to understand the issues. Well, we may not understand the obscure details of every patent claim, written as they are in fearsome legalese, but we understand very well when a company like Apple is trying to make an end-run around the Patent Office. We grok it. Don't pretend we don't, or that our understanding is somehow skewed.

This is exactly the impression I took from Ptacek's post. It is condescending and remarkably offers no counterpoint but instead deals in vagaries and innuendo.

If someone wants to discount an opinion about a patent, they have the patent at the ready for their arsenal. Copy pasting the long-form version of "on a computer" is not a response.

> Well, we may not understand the obscure details of every patent claim . . . but we understand very well when a company like Apple is trying to make an end-run around the Patent Office.

If you don't understand the patent claims, you do not understand what the dispute is. End of story.

The problem is once enough geeks accept that software patents can only hurt them, then the days of software patents are numbered. Instead, geeks tend to fall into the Ayn Rand/Joe The Plumber school of thought and will defend the most asinine patent in the world. This benefits the capital owning class as the peons are doing all their propaganda work for them.

Obviously it's not just patenting regexes. It's patenting using regexes to highlight phone numbers and email addresses in plain(ish) text presented in a UI.

This sounds exactly like the iPhone feature that allows you to tap a phone number in text and call it. What else could it possibly describe?

If the patent explicitly mentions its use in a touch screen phone (as opposed to simply highlighting it and adding it to a menu option, like dumbphones did before the iPhone) then this is a fine example of why US patent law needs reform.

Except the patent doesn't. Why do you comment on something that you don't know about, and have not bothered to check?

The patent was written in 1996. The description above is correct. The question is, was it obvious or not. Was there prior art (some people mentioned some agenda software from 1992, that's worth looking into).

In my book, it's quite simple: either people were doing it before the patent was written, in which case it's obvious and deserves to die, or it was never done before, in which case it doesn't matter how obvious it seems in retrospect.

"either people were doing it before the patent was written, in which case it's obvious and deserves to die, or it was never done before, in which case it doesn't matter how obvious it seems in retrospect."

This isn't necessarily true.

There is also a general concept that if something is obvious to a general practictioner, then it should not be patentable. Does anyone believe that minus the iPhone or that 1996 patent submission that someone else in the computer field wouldn't have "invented" the concept of linking distinct types of text? Now I know that there is plenty of prior art, and this patent will be invalidated, however just discussing the concept that it really was the very first implementation.

Remember the idea of a patent is that you are granted a government monopoly in return for revealing, essentially, your secrets. Who thinks this is a secret? Or the vast majority of patents, for that sake. The current situation is not one where people are actively stealing techniques, but instead it's one where many independent teams of developers inevitably end up at exactly the same end result. That is clear evidence of the farcical, not-as-intended patent system.

Stop and brainstorm for a couple of days about every nuance and variation of wearable or implanted computer. File patents in the hundreds. In a few years you will be a trillionaire.

So the only reason you aren't a trillionaire is scruples?

Of course - it's so easy that only someone of high moral character wouldn't do it.

No, the reason is that the time to file all those thousands of patents, and the filings fees, makes it infeasible for normal people.

But very feasible for IBM, Microsoft, Apple, etc. - companies with many laywers and engineers, and lots of money.

Aside from what azakai mentioned (it tends to be large companies that are the biggest culrpits for laying patent minefields), I have no doubt that a lot of people are doing exactly that. The now obvious value of software patents -- artifacts that for decades were literally little more than harmless curiousity -- has led to an explosion in filings.

The technical term is not 'a general practitioner' but 'one ordinarily skilled in the art.' And that ordinary skill is in the directly relevant art. So the proper question is 'was the invention novel and non-obvious to one ordinarily skilled in the art?' And here we run into endless problems because one of the best tested psychological phenomena is the 'hindsight effect.' Once things are invented and become part of popular culture or of the culture and lore of a technical sub-group they are (remarkably quickly) considered to be 'obvious.' Hordes of people who have never invented anything in their lives moan about how 'obvious' this or that patent is. If it is so easy, they should just stop moaning and go make those billions you so mistakenly believe are so easily to be earned by inventing things that are blindingly obvious. This is an absurd argument. It really is time to put it to bed. If something is blindingly obvious at the time, it probably won't get a patent. The USPTO and other such agencies are not perfect but they are not total imbeciles. Will some bad patents get through - yes. But if you want to spend your whole life focusing on edge cases why spend it on patents? Why not do something useful and campaign for prisoners' rights? After all, we know that some prisoners are innocent? So the whole penal system must be abolished, right? If we are going to have sensible discussions about patent reform we have to get some of the basic facts straight. And one of those basic facts is that what is 'obvious' now wasn't necessarily 'obvious' then and no amount of whinging from today will make that act any different. It has been 'obvious' now for some time that if you have HIV you should take antiretrovirals but it wasn't in 1982 when we started to identify AIDS. That's just 28 years ago. 'Obvious' is a slippery term - get used to it. There are no useful discussions about patents without getting over it.

There is no hindsight effect about the notion of identifying classes of text in a string of text. I fully and completely anticipate that patent being invalidated.

"If something is blindingly obvious at the time, it probably won't get a patent. The USPTO and other such agencies are not perfect but they are not total imbeciles."

The USPTO has shown, time and time again, to be willing imebeciles. They have habitually approved virtually all patents, opening them for review only when contested. I have zero expectation that they are doing any serious review of these patents, and they most certainly are not consulting any industry experts to gauge the true novelty of a patent.

You may or may not be correct about the fact that there was prior art in the case of the Apple suit. That however does not refute my key point which is that time and time again discussions include endless statements about how obvious something is, not a discussion about how obvious it was then.

As for your statements about the USPTO that is just a rant. It doesn't contribute anything to a real discussion.

html was done in 1990, and way before that hyper link systems were done. Things resembling html were done in 1986.

Many a perl script has been written pre 1996 to create linked things from plain english text. Automatically turning email addresses in plain text into html links is very common. So have many text adventure games been made which parse english text to present links. Wikis came out pre 1996, and those presented links based on the patterns of english text.

Not to mention text adventure games.

http://www.podgoretsky.com/ftp/docs/Internet/Web%20Publishin...

It is easier for me to believe that you haven't looked at the patent description at all than it is to believe that you actually think HTML or text adventures are prior art for it.

A wiki seems to be a pretty good example. A wiki processor parses data (text) looking for "structures in the data" (camel-cased words), which it then converts to a UI element (hyperlink) that a user can take action on (click it to cause the browser to transition to a new page). At first glance to me it appears wikis violate this patent, so if you can show a wiki as prior art, the patent can't hold.

> Except the patent doesn't. Why do you comment on something that you don't know about, and have not bothered to check?

Why do you comment on a comment you haven't bothered to read properly? Your attitude stinks.

For one, I said 'if' - I simply don't have time to give a patent the full reading it deserves to make an firm assertion, unlike most people in every patent-related post who read the abstract and make BS claims about it.

In addition, I mentioned a previous examples which makes you wonder (1) why this should be an issue now and not a decade ago e.g. with HTC's previous phones (2) how patentable such a thing should be.

I would not be surprised if this kind of thing appeared in the Psion PDAs, either - but it has been a very long time since I used one.

It could describe features in a CMS that do the same thing with things like ticker symbols or how the NYTimes auto-links Times Topics the first time they are mentioned in an article.

They would not infringe this stupid patent.

Why not? If you look above, you'll notice the initial implementation was on a Power Macintosh, which suggests that the claims are made very widely.

My Sony Ericsson from 2005 does this, why have Apple patented it?

They patented it in 1996.

I thought they filed in 1996.

You're probably right, but that's irrelevant here: a 2005 phone isn't invalidating prior art for a patent filed in 1996. In fact, for some forms of disclosure, there's a cushion of 12 months prior to filing, meaning that conceivably a phone could have been disclosed in 1995 that wouldn't invalidate this one.

No, but you made it seem silly somebody would even create the 2005 phone when the patent had been granted in 1995.

Actually the author isn't using those examples, he claims those are Apple's own examples:

"Deep down in the ITC archives I've dug up the so-called infringement claim charts Apple presented to the ITC in its dispute with HTC. Infringement claim charts, or often just called "claim charts", typically show the language of a patent claim on the left side and a description of the infringing activity on the right side. Any given patent claim is only infringed if all of the characterizations on the left side are matched by (accurate) information on the right side."

If that's truly Apple's own claim-matching chart, it looks pretty deserving of geek scorn. No?

If Apple invented that stuff it might have invented the computer, the internet, and the 21st century too.

This is insane. I just recently implemented this exact same functionality for one of my company's internal apps. I can't believe that such a simple little feature, something that I'd never think of as "invention" can be patented. I wonder what other patents I infringe upon during my normal every day work.

To add just one additional bit of context to the excerpt we are now freaking out about, did you also build "a processing unit coupled to the input device, the output device, and the memory for controlling the execution of the program routines."? No? Then you aren't even nominally infringing the broadest independent claim of the patent.

The patent we're talking about is here:

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sec...

All that lawyer jargon means is that it's running on a device with input, output, and "processing unit" (CPU) with a program that can scan text and automate the linking to different actions. While he may not infringe this patent exactly, it's silly that he's ALMOST infringing it because he parses text and turns it into actionable items.

Edit: True, I realize he can't get in trouble for "almost" infringing. I was just trying to point out that the core part of the patent is not a input/output/cpu but the processing and converting of text into links. This is trivial and it's embarrassing that our system allows such time and money to be wasted.

He's not almost infringing the patent. There is no such thing as "almost infringing" a patent. And unlike a lot of legal language, patent claims language isn't simply "legal jargon"; it is a programming language of its own, interpreted in a specific and strict fashion by the legal system.

A language which the overwhelming majority of people are (understandably) completely unfamiliar.

So, don't use regex on an arduino board.

You don't have to build the processing unit to infringe. You can buy it from a supplier, such as Dell.

A patent conveys the exclusive rights to make, use, offer for sale, sell, and import the patented invention inside the US. To be deemed to "make" the patented invention, it is not necessary that you separately make each of the elements of a claim; it is sufficient to buy them each from a supplier and combine them yourself.

But, in any case, HTC is not being accused of making the patented invention inside the US. They are being accused of importing it (and presumably selling it and offering it for sale).

This is very basic patent law. It is difficult for me to imagine that your ignorance of patent law is so deep that you posted your incorrect assertion out of incompetence, which leaves me with the alternative of malice.

Thomas, why are you attempting to derail discussions of the dangers of software patents by maliciously posting false information?

Naturally, you were expecting to attack me with Capo Ferra, with your orthogonal claim coupled to a caricature of a political argument to which I must adhere if I question you. Fitting, considering the rocky terrain... if your enemy hasn't studied his Agrippa. Which I have! Your point is orthogonal to mine, and questioning your dubious interpretation of this patent doesn't mean I support software patents at all.

And while I appreciate your transparency, we are discussing an actual software patent; we are not here to provide a soapbox to rail about software patents from.

I have no idea what you're talking about. You claimed that someone who doesn't build the "processing unit coupled with" etc. is facially non-infringing. This is false as a basic matter of patent law. I pointed out two independent reasons why what you said is false. That's as far from "orthogonal" as you can get. Furthermore, I haven't attempted to interpret this patent, dubiously or otherwise, or caricature any political arguments, and I haven't made any assertions about how you feel about software patents.

Misspelled references to 17th-century fencers, while amusing, are merely a distraction.

So, I ask again, why are you trying to derail the discussion with falsehoods and amusing distractions? What's your motivation for making rational discussion of patents more difficult? If the continuing discussion annoys you, I for one would appreciate it if you simply didn't post instead of trying to disrupt it with misinformation.

Discussing actual software patent cases is a much more reliable way of understanding the effect of software patentability on the world than attempting to reason from abstract principles and hypothetical situations, and the issues at stake here are much bigger than this particular patent. Mere railing is of course unproductive, but rational debate is essential.

Are they actual fencers? How cool is that? I was just referencing the Princess Bride. You learn something new every day.

My point is that the invention we're talking about must include all these elements. Dell computers do not infringe the patent simply because you could use them to create something that potentially infringe the patent. Obviously, Apple doesn't even build all the parts of their "invention". Why would you think I would be making that argument?

Moreover, to restate a point that you are militantly evading: you don't even know which claim(s) in this patent were found binding. It is more than possible, in fact likely, that it isn't simply the independent claim at the start of the patent but rather a combination of dependent and independent claims that were found to be meaningful.

Finally, I'm just going to keep saying this: I am not defending the patent we're talking about. I'm reacting to the guy who creates an "internal app" for his company and believes he is now infringing this patent because it regexes phone numbers.

> I was just referencing the Princess Bride.

I see; that makes more sense.

> My point is that the invention we're talking about must include all these elements. Dell computers do not infringe the patent simply because you could use them to create something that potentially infringe the patent.

Yes, of course; that's the "staple article of commerce" doctrine. But we're not discussing whether Dell computers themselves infringe the patent, but whether an infringing system built with purpose-built processors would still be infringing if you used off-the-shelf computers from Dell instead of purpose-built processors.

The answer is that it would. Buying some of the elements of the claim off the shelf is not a defense against infringement, and even if it were, it would be irrelevant when you're being charged with importing or selling rather than making an infringing article.

(It's possible that under the machine-or-transformation doctrine, a process might be unpatentable, and patents on it might be invalid, if it could be implemented by software on a general-purpose computer, because a general-purpose computer is not a "particular machine". Unfortunately In re Bilski and Bilski v. Kappos have not resolved this question, but only left it muddier than before.)

> Obviously, Apple doesn't even build all the parts of their "invention". Why would you think I would be making that argument?

Because you said:

> did you also build "a processing unit coupled to the input device, the output device, and the memory for controlling the execution of the program routines."? No? Then you aren't even nominally infringing the broadest independent claim of the patent.

The key word here is "build". The person you were responding to had said:

> I just recently implemented this exact same functionality for one of my company's internal apps.

It's patently obvious that he was combining his program routines ("internal apps") with a processing unit coupled to the input device, the output device, and the memory for controlling the execution of the program routines. You brought up the fact that he probably didn't build the processing unit, claiming that he couldn't be infringing if he didn't build it. But the presumed fact that he didn't build it is irrelevant.

In another thread, you wrote:

> I'm strongly inclined to believe that no, your infringing product needs to embody a purpose-built device.

These assertions of yours are false and have no foundation in patent law. However, I retract my earlier claim that you already knew that.

> Moreover, to restate a point that you are militantly evading: you don't even know which claim(s) in this patent were found binding.

I wasn't evading it; you just hadn't stated it. Now that you've stated it, it's trivial to demonstrate that it is false. The comment at the root of this thread links to Florian Müller's blog post at http://fosspatents.blogspot.com/2011/07/these-tables-show-ho..., which shows exactly which claims Apple is asserting against HTC, and which ones the ITC judge found that HTC is infringing: claims 1, 8, 15, and 19 of the '647 patent, and claims 1, 2, 24, and 29 of the '263 patent.

> I am not defending the patent we're talking about. I'm reacting to the guy who creates an "internal app" for his company and believes he is now infringing this patent because it regexes phone numbers.

Rather than just reacting to him by posting the smokescreen of nonsense and bullshit that you did, it would be more useful to collaborate with him by honestly and accurately explaining the actual patent law and accurately analyzing what would be necessary for his "internal app" to infringe it. Whether or not he is manufacturing his own processors is irrelevant to that.

It's not patently obvious to me that his internal app matches claim 1 of the patent; if it was, I would have commented on it. It is not an infringement of this patent to sell software that regexes phone numbers. Do you believe otherwise?

Is that a pun?

I wasn't arguing that it did or didn't infringe the claim, because I hadn't read the claim thoroughly yet; I was arguing that running it on a purpose-built processor was irrelevant to whether it infringed the claim.

However, yes, I believe otherwise, if you're selling the software to US customers. (The "internal app" guy isn't selling the patented invention, but he is making and using it.)

It is an infringement of this patent to sell software that regexes phone numbers (in a "server") and makes them into clickable links to take some action, if you're selling it to US customers.

Installing the software on a computer and running it necessarily produces an infringing system, according to Apple's assertion, which the ITC judge has upheld. The staple article of commerce doctrine does not protect the sale of articles "which are only adapted to be used in a patented combination".

Microsoft v. AT&T would seem to protect you if you're exporting the software rather than selling it to consumers within the US; "The master disk or electronic transmission Microsoft sends from the United States is never installed on any of the foreign-made computers in question. Instead, copies made abroad are used for installation." However, it would be very surprising indeed if courts construed that to protect sellers of software for use inside the US, in particular because Microsoft was still paying AT&T revenues for US copies of Windows in that case.

First, you spent a lot of words arguing with me before reading the patent. Don't do that.

Second, you've misread the patent. The "server" needs to be in the same machine as the input and output device. The patent description defines the terms used in the claims language (you will easily find a basis for that assertion in patent law if you look); here, the server is a subroutine coupled to the system via an application programming interface.

Further, I believe if you take the time to poke around, you'll find software extracting patterns from text in the prior art references. Claims cannot be interpreted to substantially duplicate the descriptions (not simply the claims) of cited prior art.

Finally, your logic doesn't work. In your worldview, if I sell frobnitz flanges, and Apple invents a phone that uses frobnitz flanges to legibly display the otherwise invisible color blurple, and then files a blurple-display patent referencing frobnitz flanges, I can no longer sell frobnitz flanges. Any elementary description of "how to read patent claims" will show this logic to be false. The infringer must violate all the elements of the claim.

(Late edit: I mean to say, the direct infringer).

Your objection to me not having read the patent would hold more water if you could show something I was mistaken about as a result.

I'll take your word on the definition of "server". I'm not sure why you think I misread the patent, though. If I thought the "server" had to be on a different machine, I'd be claiming Android didn't infringe.

You are correct about claim construal, but no narrower construal suggests itself to me for this claim. So perhaps you are mistaken and there is no such cited prior art, or perhaps the examiner should have stricken this claim, or perhaps there is a narrower construal that isn't obvious to me.

You are mistaken about my worldview. My worldview comes from patent law. In your example, if people have been buying frobnitz flanges before Apple's new invention, they are clearly adapted to some other use than displaying blurple. So, in my worldview, they are a "staple article of commerce", a term I have used several times in this discussion and which you would do well to look up, so their sale is non-infringing. However, a piece of software whose installation and use on a computer necessarily infringes the patent is not suitable for substantial noninfringing use (unless perhaps you can argue that its use to paper your wall or as an illustration in a discussion about software patents is substantial), so selling it separately is clearly just an attempt to infringe the patent in a sneaky way. (Presuming, of course, that the patent is valid at all.)

This is not some hare-brained theory I just thought up on the spur of the moment. You'll find it explained in any patent law textbook, in quite a number of online introductions to patent law, and in §271 (c) of the Patent Code. You are welcome to attempt to explain to Congress and the Supreme Court that their logic doesn't work because it forbids the sale of frobnitz flanges or whatever, but don't blame it on my worldview.

"a processing unit coupled to the input device, the output device, and the memory for controlling the execution of the program routines."

That sounds like you have just described pretty much any user facing computer ever built. Can you explain how this narrows the claims of the patent in any significant way?

Every computer ever build has a processing unit coupled to input, output, and memory that includes an analyzer server for detecting structures and an action processor that controls the execution of program routines on that computer?

Is this trivial to build with COTS parts? Of course it is. Does it describe a web app? No. Web apps do not comprise input, output, memory used to store program routines, a processor coupling them, a server running on that very processor for that uses patterns to recognize structures such as phone numbers, and an action processor that executes specific code stored in that memory based on the structure the analyzer server found. In a web app, half that stuff happens on a server on a different computer, and half happens in the web browser.

If they wanted their patent to describe web apps, they would have said the server was communicatively coupled to the input/output processor, resided on a computer that may be different from the computer with the input/output processor, and that the subroutines run in response to user selections might run either on one computer or the other, &c.

This is not an awesome patent. That's not the point I'm making.

I'm a bit confused about how a web app doesn't infringe. Your claim is it doesn't because it's split between two machines, but that's merely a function of one use. A web app can be used from a single machine, with the server portion and the client portion executing on the same processor. You could even jerry-rig it so the browser and the server communicated via a shared memory map, if using the same memory locations is an issue.

From the link, the system:

"...identifies structures in computer data, associates candidate actions with each detected structure, enables the selection of an action, and automatically performs the selected action on the identified structure... the system may operate on recognizable patterns for text, pictures, tables, graphs, voice, etc. So long as a pattern is recognizable, the system will operate on it. ... The present invention provides a computer system with a central processing unit (CPU), input/output (I/O) means, and a memory that includes a program to identify structures in a document and perform selected computer-based actions on the identified structures. The program includes program subroutines that include an analyzer server, an application program interface, a user interface and an action processor. The analyzer server receives data from a document having recognizable structures, and uses patterns to detect the structures. Upon detection of a structure, the analyzer server links actions to the detected structure."

So they "invented" a device that has embedded pattern feature-extraction, communicates with a server to perform recognition on the extracted data, and return the results, with which the device can do anything it wants with.

Yeah, I don't buy that this is an invention in the 21st century, with no prior art.*

This is a patent for integration of prior inventions, and not really an invention itself (hence, not a real patent). Pattern recognition is an invention; putting it on a device is not. Even if we say "yeah, let's grant patents on integration," there is still prior art for this particular one since it is so broad.

*EDIT: it has been pointed out that this patent was filed in 1996.

"Yeah, I don't buy that this is an invention in the 21st century, with no prior art."

Maybe not in the 21st century, but it might be in the 20th century which is when the patent was filed (1996).

http://patft.uspto.gov/netacgi/nph-Parser?Sect1=PTO1&Sec...

Fair enough, but I've read research papers on pattern recognition that predate even 1996 (the field was pretty hot in the 80s before computing power caught up with theory)

The server isn't even a remote server... it's just a service reachable by an opaque API on the phone.

Why are you under the impression that you need to build the device you describe yourself (which as other posters have pointed out, is just the standard patent description of a computer)? Are you implying I'm allowed to infringe on a patent as long as I outsource the building of some of the parts?

If we have a patent which is violated if the bundled software includes a feature but isn't if the software is separately installed later, by what argument is that a rational patent? Where is the huge philosophical leap between a tool within the platform and an extension which plugs into the API?

The simple answer is, as we all know, that there isn't, and that they've wrapped some fancy language for a general purpose computer around regular expressions (which very substantially predate the patent) and managed to get the patent granted. This is ludicrous.

I don't understand patents all that well, so perhaps somebody could help me understand: how are the things patented not totally absurd?

If you think you can patent utterly absurd things... well, you understand patents pretty well.

I don't understand the question - please rephrase.

I'll ruin your perfectly good joke by actually explaining what I meant:

Some patent applications really make sense to me. I read 'em and I say, "Wow, that was pretty clever. I never would have thought of that." Whether or not we should allow patents is a different question, but if we do, I get why the patent is valid.

These, though, I don't get. I know patents claims are a special language, so maybe I'm missing something, but I'm pretty sure a NeXT OS box circa 1991 running common apps covers all this.

Much clearer - thank you.

I don't understand how some of these are even patentable. The use of an output device? Really?

Does the linking patent apply only to phones? If not, I've been in violation of this one several times in the past.

If you've written a line of code you've violated several software patents. The question is will anyone come after you.

That's why the system is so ridiculous and broken.

Isn't that basically what a wiki does? I'd also suspect that there were similar solutions for the old Psion 3 handheld.

Reply on news.ycombinator.com