Told to remove blog posts describing patented algorithm

http://sites.google.com/site/redcodenl/

by willvarfar • 16 years ago

429 147 16 years ago

sites.google.com

I'm not a lawyer, but it seems to me that he is looking at this backwards.

If his goal is not to actually implement the algorithm, but just to discuss or understand it, he can use the lawyers to his advantage. He's already received the two US patent numbers, which contain published (and freely publishable) text describing the algorithm in great detail. And, since the lawyers are interested in the international implications, it should be simple for them to provide a list of (or better yet, the full text of) all international patents granted or applied for.

My point is not that software patents (or patents in general) are good or bad or harmful or useful. My point is that, for better or worse, a patent represents an exchange-- the inventor is granted a limited-time monopoly on the implementation of an innovation in exchange for publication of the details.

willvarfar2• 16 years ago

didn't the Dutch lawyer say he could get fined? If it's your nuts on the block, would you have the guts to stand up to them?

alextingle4• 16 years ago

He should find a different lawyer.

jrnkntl• 16 years ago

I'm not sure but the lawyer doesn't explicitly say that it will be a 'fine'. The translation is a bit off but actually he says that the lawsuit will deal around a substantial amount of money.

Confusion• 16 years ago

No, he just gave the usual advice: if you don't comply and they decide to litigate, it could become an expensive affair, just because you are forced to defend yourself.

joe_the_user2• 16 years ago

"Landmark also requests that in the future you do not ship, deploy or post any portions or versions of this code in its current state or in any modified state."

Here, the lawyer is demanding that he not discuss the method whose "publicness" is the public good supposedly given in exchange for the patent.

mkramlich• 16 years ago

welcome to the inane world of software patents

binspace2• 16 years ago

It's really only the threat of legal action that matters. Who wants to deal with lawyers, legal expenses, and energy spent defending your position, even if you are right?

It's too bad such anti-innovative behavior is even legal.

pyre• 16 years ago

If the suing company loses, they forfeit their right to do business in the country they filed in. Sounds reasonable. It would at least stop this penny ante bullshit.

loup-vaillant17• 16 years ago

Remember Paul Graham's words: "One thing I do feel pretty certain of is that if you're against software patents, you're against patents in general." http://paulgraham.com/softwarepatents.html

I made up my mind, and currently, I think the whole patent system should be abolished.

Some may think we should be safe if software patents are abolished, but we're not. Make a bit of Arduino based hardware that happen to infringe a patent, and you're screwed. In short, patents are hampering the rise of Free Hardware, just like it does Free Software.

It is past time we examine the effects of patents rigorously, or listen to people who have.

nudge8• 16 years ago

How would you encourage the development of new pharmaceuticals (the process of which costs vast, vast sums of money and can take decades) if, as soon as the new drug is done, anyone can come and just copy the formula?

I'm not pro-patents on everything, but pharmaceuticals is a case where the patent bargain seems to produce benefits in terms of drugs that otherwise might not be developed.

mjgoins• 16 years ago

"Remember Paul Graham's words:"? Seriously? This messiah thing has got to stop.

openfly• 16 years ago

I agree.

alextingle4• 16 years ago

This is nonsense. Publishing the details of a process cannot violate a patent. The patent system was created specifically to enable the public to see and discuss the latest state of the art. Are these clowns going to try sending take-down notices to the patent office's web-site next??

(Besides, algorithms cannot be patented in the EU, so it is impossible that this code violates any patent.)

hcho7• 16 years ago

Algorithms can be(and is being)patented in the EU. It is the software you cannot patent.

alextingle4• 16 years ago

[Citation needed]

loup-vaillant17• 16 years ago

Wait a minute: what is the difference between an algorithm and a piece of software? I don't see any.

pbhjpbhj4• 16 years ago

You can patent software. I used to examine and grant software patents on behalf of the UKPO (now UKIPO) Comptroller (sic).

The UK practice had substantially changed since I left (after Macrossan, http://en.wikipedia.org/wiki/Aerotel_v_Telco_and_Macrossan%2...), the UK Patents Act has harmonising terms such that various sections are interpreted to match EPC articles. In the EU software patents were ruled out "as such" and traditionally the loophole was that as long as a "technical effect" (beyond that expected from mere computerisation) was demonstrated in the application of a patent then the software patent was allowed.

Generally the EPC and UKIPO treatments have differed as software patents were actually ruled out "per se" in UK Patents Act 1977 S1(1)(d) referencing 1(2)(c) and UK has been stricter in interpreting this (see Macrossan).

UK law also states that [pure] business methods and mathematical algorithms are unpatentable (again Section 1; see examiners manual - http://www.ipo.gov.uk/practice-sec-001.pdf) amongst other things.

However historical quirks meant that compression algos were allowed in certain areas (imaging) that wouldn't be allowed unless they were couched in the right way (that's my interpretation, I couldn't understand why they were granted).

The Macrossan tests are spelt out clearly in the above Patent Examiner's manual link as is the UK practice (but this can be amended by court decisions).

http://en.wikipedia.org/wiki/Software_patents_under_the_Euro... and the corresponding UK page appear to be thorough treatments for those that wish to read more.

10ren3• 16 years ago

I think they were going for "encouraging infringement of a patent". Which publishing certainly does do; though if that was the sole ingredient, the patent office would also be guilty (as you say).

deadsy• 16 years ago

Right. It's hard to see how publishing source code in and of itself constitutes a make or use action, so they could go for "inducing infringement". As I understand it, you would need to have an infringer for that to be true.

borisk3• 16 years ago

Letters say nothing about description, the CEO only wants the code out.

BTW I thought there are no software patent laws in EU or EU member states. Just USA and Japan were on the road to insanity some time ago.

ComputerGuru• 16 years ago

Read the last letter. It says clearly that his description of the algorithm in the blogpost is in violation of the patent laws.

ANINDIAN7• 16 years ago

dude that guy is surely bullshitting Description of an invention is actually the purpose of the patent! Patent is there to allow other to get knowledge about your invention while protecting your "Commercial" interests!

borisk3• 16 years ago

"While we trust your good intentions, yes, we would like you to refrain from releasing the code at all and to remove the blogpost explaining the algorithm."

It doesn't say directly the description is a problem. IMHO it's the whole blogpost, because it contains code.

ra2• 16 years ago

I'm no lawyer but my advice would be to object to the patents in your jurisdiction (and the US, if possible) by writing to the appropriate local patent offices. State that their work isn't novel as it's remarkably similar to your own. If your work is based on "public domain" knowledge (for example techniques that have been previously published in a book, or on the web), it is important that you refer to those.

I'd also speak to a different IP lawyer, one who is familiar with software patents. You shouldn't have to pay for an initial consultation.

EDIT: You could also speak to your local EFF. They'd have better advice.

ANINDIAN7• 16 years ago

WTF Just write at end your implementation:

License: Research Purpose Only

Any patented invention can be used for research pupose!

To quote from wikipedia: A patent being an exclusionary right does not, however, necessarily give the owner of the patent the right to exploit the patent.[13] For example, many inventions are improvements of prior inventions that may still be covered by someone else's patent.[13] If an inventor takes an existing, patented mouse trap design, adds a new feature to make an improved mouse trap, and obtains a patent on the improvement, he or she can only legally build his or her improved mouse trap with permission from the patent holder of the original mouse trap, assuming the original patent is still in force. On the other hand, the owner of the improved mouse trap can exclude the original patent owner from using the improvement.

Add few new features

jjs• 16 years ago

> Add few new features

Or devise some novel features that would allow this marvelous invention to interoperate with existing software!

Then patent the combination and fuck them with it. >:D

pbhjpbhj4• 16 years ago

>Any patented invention can be used for research pupose!

This is a grey area under the EPC - research is allowed but commercial use is not.

Consider: If the invention is a measuring device used in a particular area of research then that doesn't mean you can rip off the invention and claim it is for research purposes.

Generally I'd say if you built the invention just so you can use it then that's commercial.

ANINDIAN7• 16 years ago

Create a Fake Email address and post it on a Posterous blog. And then we will take care that it is found out for perpetuity! Also we can try to destroy his page rank by making reference to his name!

ANINDIAN7• 16 years ago

Any way you are currently on second page of Google Search! Better Post it on a .in Indian Web Address Patent laws in india do not allow software patents + no one gives a shit about C&D!

andraz• 16 years ago

Patents cannot be used to prevent someone publishing research, source code, explaining how things work, etc. Infringement happens when the patented matter becomes part of commercial product, is being used commercially or sold. Not by publishing either textual description or source code. An example of this is x264 codec which is perfectly legal (to distribute and work on) even in countries that happily allow patenting everything under the sun. The infringement happens when it becomes used as part of the product or service.

Patent system's primary 'deal' with the inventor is "time-limited monopoly in return for a disclosure of how you did it".

Leaving aside the question on patentability of software and algorithms, I am sure talking to FFII, EFF and similar organizations might help gathering traction for a proper pushback.

billswift2• 16 years ago

Publishing doesn't violate a patent, but it doesn't require commercial activity - building your own device, or writing your own software, for your own use constitutes a violation.

ANINDIAN7• 16 years ago

Please some can someone with good knowledge of proxies and etc post it one 4Chan or Posterous or similar site and then post it as Question on http://metaoptimize.com/qa/

Which will get the word out to practitioners in this field ! I currently cant do that since i am on a *.edu ip

ANINDIAN7• 16 years ago

cant we start http://en.wikipedia.org/wiki/Streisand_effect effect?

SoftwarePatent3• 16 years ago

IAAPL (I am a patent lawyer), but I am at the very beginning of my career. Anyway, non-lawyers often are too credulous when they get nastygrams from the other side. Don't believe everything they say to you! It's amazing what non-lawyers will believe and do just because they get a mean-sounding email.

That being said, one of the fundamentally unfair things about the legal system is how expensive trials are. Even if the other side here has no case, they might be able to haul you into court and run up expenses for you. Maybe you can become a cause celebre of the anti-software patent blogosphere?

noonespecial2• 16 years ago

Welcome to HN. I hope you'll stick around and share what interesting tidbits you are able from cases you take on. Your fresh perspective will be highly welcome.

It is a shame that the patent system is essentially inaccessible to the "little guy" without becoming a celebrity charity case or piquing the interest of someone like the EFF. Here's hoping that you can do much in your coming career to remedy this.

chopsueyar• 16 years ago

Do you find it unusual the letter was sent by the CTO and not the legal representative of the company?

SoftwarePatent3• 16 years ago

No, that does not strike me as unusual. In a small company, people are forced to wear many hats. Matters like this can usually be "resolved" through a few emails or a phone call. If the CTO were to call his outside counsel and direct her to write the nastygram, outside counsel would charge $300+/hr. to do so.

Ardit20• 16 years ago

If they go to court and he does win the case wouldn't the cost of his lawyers be paid by the loosing side, so, he in that case would not lose anything and perhaps might even be awarded nominal damages if the other side pursued the case friviously.

hga• 16 years ago

Hmmm, yes, how do the courts in the Netherlands do this? That wouldn't be true in the US but I gather very few if any other countries follow our model (http://en.wikipedia.org/wiki/American_rule).

SoftwarePatent3• 16 years ago

This depends on many factors. 1) What country the lawsuit takes place in. Unfortch I only know US law. Here in the US, the factfinder (a lawyer word for either the judge or the jury) can order legal costs to the winning party under certain circumstances, but those are rare. I have heard anecdotally that in Britain legal fees are always paid by the losing side.

cduan• 16 years ago

If you have US patent numbers, you can get the related European patents either from either http://ep.espacenet.com or http://www.patentlens.net. For example, http://www.patentlens.net/patentlens/patents.html?patnums=US... shows related patents for one of the two US patents mentioned.

vibragiel• 16 years ago

I was wondering how to contribute with my bit of pressure. I'm a regular Shazam user, and was thinking of sending them an email telling that I don't want to make business with companies that deal with bullies like Landmark Digital Services, that I'm uninstalling their app (even though I paid for it) and will be campaigning against it among my friends.

But then I thought Shazam might just be another victim of this messed up patent system. On the other hand, they're making profit out of it and favouring and legitimating Landmark Digital Services on the way.

What do you think?

billswift2• 16 years ago

According to this comment http://news.ycombinator.com/item?id=1497450 Shazam sold the patents to Landmark, which the parent of the linked comment pointed out is owned by BMI. Shazam deserves any flack it gets and more if those are true.

mikeytown2• 16 years ago

Reminds me of when I accidentally violated a software patent in less then an hour. http://forum.doom9.org/showthread.php?t=139913

pbhjpbhj4• 16 years ago

You didn't do it in less than an hour, you did it with your whole careers worth of knowledge acquisition + 1 hour of implementation.

If as you say your steps were a priori obvious and trivially implemented and were at the time that the relevant patent was applied for (or relevant invention made in the US) then the patent is very weak and a citation against it should be easy to find.

What appears to be the case is that professionals working in the field for a substantial time were not able to solve the problem easily (eg 2Bdecided's comments).

In short you should give yourself more credit.

TIL: AVIsynth is a script based video editor.

maurits• 16 years ago

Okay, so I am not a lawyer, but I don't get it. One of the most know audio matching systems is from Kalker and Haitsma, published 2002. http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.103...

dtf• 16 years ago

Oh dear. I can see the open-source and pirate communities coming down on these guys like a ton of bricks. I wouldn't be surprised to see a CDDB-style distributed database and server and accompanying mobile apps up by the end of next week. Shazam's CTO may well have been better off turning a blind eye.

It's got all the elements hackers love: software patents, an underdog, free speech and an interesting technical problem.

jakevoytko• 16 years ago

I doubt anything will happen. Stories with similar elements don't usually end with an underdog victory [1]. If the software was already released, people could create "I'm mirroring the Shazam code, come and get me Shazam!" blog posts. The news cycle would have been well-fed, and might have generated enough hype to cause the Shazam CTO to learn to leave the issue untouched.

However, now someone must develop the code. It'll take time. When it is done there will only be passing interest in the whole issue. This is interesting enough that it'll get some action on blogs over the next week, but I don't see this becoming a long-term win for the original author. Just Shazam and the blogs.

[1] One example is "How I Got Sued by Facebook". http://petewarden.typepad.com/searchbrowser/2010/04/how-i-go... . It combines elements hackers love: the rights to your information, an underdog, free speech, and an interesting technical problem (indexing all of Facebook). But no public databases of Facebook user data sprung up as a result.

weirdwonka• 16 years ago

The reason for that is that people who have capabilities of doing that know that its not a good thing. The data can be used for re identification attacks against user privacy.

sh1mmer• 16 years ago

Hasn't MusicBrainz.org being doing this in the Open Source for a long time? At least longer than I personally remember knowing about Shazam at all. All of their code is some variant of GPL.

Jun8• 16 years ago

AFAIK, MusicBrainz just relies on metadata associated with a music file and its length. It doesn't actually derive an audio fingerprint from the music, like Shazam does.

DanielRibeiro• 16 years ago

Alistair Cockburn (agile authority and creator of Crystal methodologies) comented about this on twitter ( http://twitter.com/TotherAlistair/status/18053308892 ): Simply putting Shazam code on a website wouldn't be patent infringement (AFAIK) because the patent itself does that and is public reading.

willvarfar2• 16 years ago

Its not Shazam who are making the fuss though, its Landmark (http://www.landmarkdigital.com/about) who are owned by BMI (the record company - RIAA?)

ahi• 16 years ago

Shazam sold them the patents.

kljensen• 16 years ago

Patent +applications+ cannot be infringed.

srparish• 16 years ago

Andrew Tridgell's talk on patent defense could be useful for him to read: http://news.swpat.org/2010/03/transcript-tridgell-patents/

impeachgod• 16 years ago

Assholes. I am living in Ukraine, and I am willing to host any patent-infringing code or documents on my Ukrainian servers.

koenigdavidmj• 16 years ago

You could always link to the patents, since they describe the process in so much more detail.

reader5000• 16 years ago

This illustrates one of the major problems with patent law: uncertainty. Nobody, including the pros, can tell the poster whether the Shazam claim is legitimate or not. Therefore, under the threat of untold financial penalties, real or not, Shazam can simply bully whoever they want.

Although I know nothing about European patent law and almost nothing about US law, here is the relevant US code:

"Except as otherwise provided in this title, whoever without authority makes, uses, offers to sell, or sells any patented invention, within the United States or imports into the United States any patented invention during the term of the patent therefor, infringes the patent," (emphasis added). [1].

It seems to me the issue would be whether writing the code for patented software constitutes "making" it. Who knows. The patent system is such a joke.

[1] http://www.law.cornell.edu/uscode/usc_sec_35_00000271----000...

iuygtfnh5• 16 years ago

>it seems to me the issue would be whether writing the code for patented software constitutes "making" it.

That's the problem with software patents. The patent must be published - that the whole point. But in software writing down the algorithm is also making an implementation. It's like the pornography laws that defined looking at an image on a computer screen = producing an image and therefore equivalent to being present when the act took place.

voidpointer3• 16 years ago

Great example of the problem with software patents. Landmark has a method for music-matching. They would love to be the only one having access to that technology. The confusion starts where the patent system is meant to only protect how their particular implementation matches music and not the concept of music matching at all. They on the other hand argue, that the method described by the author is similar to what they have patented - after all, it uses the same fundamental CS concepts such as hash-table lookups etc...

As other mentioned, the idea with patents was to offer people a limited time monopoly on their innovations in exchange for publishing them so others could use/enhance these ideas instead of reinventing the wheel over and over. I have not seen one instance where this works out for software...

iuygtfnh5• 16 years ago

>Landmark has a method for music-matching.

No they have a patent the idea of music-matching, not the implementation.

I could have a patent on the idea of a car that defies gravity - and you pay up when you actually invent a flying car.

wake_up_sticky• 16 years ago

No you couldn't.

voidpointer3• 16 years ago

This is not how it works. In the patent application you have to describe how your invention works.

In software patents, the implementation is described in such general terms that a lot of the time, any implementation can be said to use the method described in the patent. This is my biggest problem with software patents: once you abstract your solution enough, it isn't of much use to anybody, hence the value of the patent to the general public is null while it still allows you to block out competitors that want to find a solution to the same/similar problem although your patent is of no help to them actually solving the problem.

It is this skewed relationship that drives me crazy. Licensing a software patent only allows you to come up with your own solution a lot of the time, it does not save you the development cost.

If you license a patent for some industrial process it saves you huge amounts of research cost so licensing the patent is actually attractive.

Software should only be patentable if the patent provides actual value in solving the problem it claims to solve. Since useable software implementation are so specific, this may rule out patenting software at all.

joe_the_user2• 16 years ago

they have a patent the idea of music-matching, not the implementation.

This is indeed the exact opposite of how the patent system is supposed to work. But it is none-the-less a description of how the patent system is more and more winding up working.

nzmsv• 16 years ago

This is the outcome that patent trolls hope for. And this has been going on even before software patents. Look up the story about Feynman and the nuclear airplane.

wazoox2• 16 years ago

The problem is with the (utterly broken IMO) US legal system (that unfortunately tends to "colonize" more and more foreign legal systems). Because they have a patent, they can efficiently scare away many hobbyists or commercial developers who could compete with them with litigation costs.

abstractbill• 16 years ago

This is quite amusing to me personally. I am named as the inventor of a patent which makes claims that appear practically identical to the Landmark one (mine is owned by Macrovision, for whom I used to work). I wonder how many times the damn thing has been patented!

protomyth• 16 years ago

I still wish some congresscritter would pass a law simply stating "if you can copyright the implementation, then you can't patent it"

frognibble• 16 years ago

The LAME MP3 encoder uses technology covered by patents. The LAME developers distribute the code in source code form only. The developers say that it is legal to distribute the code because the source code is a description of the patent and not a working implementation of the patent. See last question at http://lame.sourceforge.net/tech-FAQ.txt for more info.

wazoox2• 16 years ago

I'm pretty sure you could release the code copyright and the code to the FSF, and let it go. I think it's important to struggle against software patents, now more than ever!

cjbprime• 16 years ago

Irritating. The patent tradeoff is disclosure in exchange for monopoly, and Landmark's pretending the disclosure part doesn't have to exist.

dpatru• 16 years ago

The problem here is that it's too easy to make and use a software invention once you have a sufficiently detailed description of it (source code). An analogous situation would be the invention of a new homemade bread. A detailed recipe of the bread in a cookbook would make it very easy for anyone to make use of the invention and very hard for the inventor to stop infringement. In such a case, the law must take sides: either it protects the public's right to share information over the right of the inventor to benefit from his invention, or vice versa.

It seems to me that the law sides with the public by forcing the inventor to choose between patent and copyright. If the inventor chooses patent, his invention has protection, but the public is free to share information about it. Indeed, to obtain a patent, the inventor must fully disclose the best way he knows to make and use the invention. On the other, hand, if the inventor chooses copyright, his particular expression is protected, the public is not free to share it (republish it), but the underlying ideas behind his invention are not protected.

So for instance, a musician could patent his unique method of strumming a guitar and he could copyright his music recordings and explanations of his strumming method, but has no way of preventing public discussion generally of his strumming method. His invention does not mean that there can be no more learning and innovation in the field of guitar strumming.

Similarly, a software maker can copyright his expression (source and object code) of his work, and patent the use of his work, but he cannot (should not) prevent public discussion of his work generally.

To hold otherwise would mean that intellectual property (IP) holders could stop all learning and progress in the areas in which they hold their IP. But the purpose of IP is exactly opposite. The US Constitution says that Congress shall have power "To promote the Progress of Science and useful Arts by securing for limited Time to Authors and Inventors the exclusive Right to their respective Writings and Discoveries." (US Const., Art. 1, Sec. 8, Cl. 8)

russell• 16 years ago

IANAL, but I would hazard that you are wrong on nearly all points. There are very specific rules on the form of a patent application. It used to be/may still be that you had to include stylized drawings of a machine that accomplished the function. Ten thousand pages of code wouldn't cut it. Generally the patent is abstracted from the actual machine. A couple of the problems with software patents is that the abstraction is too great and the claims are too broad. It's the claims that trip up future developers.

The musician example is pretty much bogus, because method patents have been pretty much disallowed, though there is a movement to get them accepted. Method/process patents would bring all innovation to its knees.

Also a misconception about copyright. You can have a copyright without publishing the code, although if you publish it without a copyright, it may be in the public domain. Code copyright is used as a defense against duplication, not innovation.

> Similarly.. Yes indeed.

dedward• 16 years ago

One take-away from this is this, though:

When it comes to patents - the fact that your software is non-commercial, or that you invented it independently without knowing about the patents does not matter.

Unless you can prove you created the infringing work before the original patent applicants did, in which case you can invalidate and/or take over the patent, patent rights are fairly absolute, and have nothing to do with derivative works or anything like that (which is why software patents suck so much....)

Patents are granted via application, and apply unilaterally, regardless of how the infringing works were created. The only defences in general against patent infringement are a)your product does not actually infringe on the patent because it's different enough and b)attacking the validity of the patent itself.

Contrast with Copyright, which is granted automatically, and independent creation of similar works (even identical) is permissible, as long as they were truly independent and not copying each other.

(This is why clean-room reverse-engineering techniques, where one group revere engineers the software and documents it's function, and a completely separate group who's never seen the original work writes new code to those specifications) work for software when it comes to copyright, but not against patents.

fleitz• 16 years ago

Even if you could prove such a thing it's still very expensive to do so.

joshfraser• 16 years ago

lots of companies send nastygrams like this even if they have no intention of following through with legal action. it costs them nothing (except reputation on HN) and there's a good chance the recipient will get scared and give in without a fight.

there are important issues on the line here and it makes me sad that so many people feel like they need to walk away because of the possible legal expenses.

i know it's easy for me to say this as i've never been sued myself, but for once, just once, i'd like to see someone fight back against something that is so clearly wrong.

pier0• 16 years ago

"why are they targetting me and not the already published implementation of Dan?"

Probably because Dan didn't title his article Creating Shazam in Java

geebee• 16 years ago

This is so depressing.

I have to say, the replies from engineers who have spent some time figuring out how the patent system works have been enlightening and interesting to read.

But none of this matters unless a well funded group (maybe the EFF, maybe some businesses who want to challenge this patent) decides to pick up the cause and fight. There is no way a programmer who coded something in a weekend and put it on his blog is going to be able to go up against a well funded patent holder.

The merits are irrelevant, seriously irrelevant.

This is what depresses me so much about the US legal system. While the law around software patents is absurd, the problem isn't really that software patents exist per se (1), it's that stupid patents are granted and that it is brutally expensive in terms of time and money to fight them.

(1) I do consider software patents to be a failed experiment. Sometimes the implementation is so bad that we can say "tried it, didn't work" even if in theory it could have been done better.

malandrew• 16 years ago

I'm not a lawyer and I may be wrong but can't we use the patent vagueness against itself in a sort of "patent aikido" defense.

It's seems like it would be trivially easy to release this source code given the way the patent is described.

This source code does pattern matching for audio, sound and music. I looked at the patents in question. They only discuss this method for use with "audio files", "audio samples", "sounds", and "music".

However the patent doesn't cover other types of waveforms with changing frequency and amplitude.

With that in mind, couldn't you just change the language used to describe what you code does so that in now describes other types of waveforms that are NOT sound, music and audio?

It seems to me that patent lawyers can either patent the implementation or they can patent the application of their implementation. Simply changing the application of the implementation you coded in a weekend would mean that the patents cited no longer apply to the code he wants to publish.

Also, if these lawyers have never seen the code in question. How are they going to know you released it if it somehow gets released anonymously?

caf• 16 years ago

You know, what's kind of interesting about this whole thing is that it sounds like a case could be made that Shazam themselves are infringing copyrights on the songs in their database. After all, in order to make the database of hashes, they had to make a temporary, unauthorised copy of each song recording.

Similarly, the client-side software makes an unauthorised copy of the music as it records it from the environment to do the test. This seems like it could be contributory infringement on the part of the software writers.

Does anyone know if the record companies ever tried to go after them on these grounds?

sswam• 16 years ago

The basic technique they are using, to generate a Fourier transform and distinctive signature from a segment of music, and look it up in a hash, is completely obvious. It should not be patentable. The gory details of their algorithm might be patent worthy (although I don't like patents at all). But Roy did not use their algorithm, he invented something similar. Greedy trolls, hate.

It might be a nice idea to integrate music identification with CDDB I think.

Google has a similar system that they are using to prevent the use of copyright music in youtube clips (and they also have patents for their method). I suppose many other systems exist to do the same things. Roy should publish his code. Death to trolls.

Told to remove blog posts describing patented algorithm

Related Stories