Open Source Patent Donations?

patspam writes "As a software engineer I come up with patentable ideas every now and then, ideas which I'm not interested in pursuing myself but which I'd like to keep out of the hands of private entities/patent trolls in my own personal effort to defeat software patents. Should I patent the ideas and donate them to some sort of open source foundation? Or just blog about the ideas so that the 'prior art' exists in the public domain? What's your strategy for fighting against restrictive software patents?"

Comments

As far as US is concerned

[hansraj]

you could just blog since the system is not first to file but first to invent. Moreover it also saves you patent fees that you would have to pay if you actually try getting a patent.

Re:As far as US is concerned

[mavenguy]

This is a common misconception. This will not defeat someone else who conceives and files for a patent later in the US. From 35 USC 102:

A person shall be entitled to a patent unless--
...
(g)(2) before such person's invention thereof, the invention was made in this country by another inventor who had not abandoned, suppressed, or concealed it. In determining priority of invention under this subsection, there shall be considered not only the respective dates of conception and reduction to practice of the invention, but also the reasonable diligence of one who was first to conceive and last to reduce to practice, from a time prior to conception by the other. (Emphasis added)

You need to actively work on it and have reduced it to practice (the latter which is satisfied by filing a patent application). Your triumphant opening of the envelope might garner you bragging rights, but no destructive effect on the patent rights of others.

Re:As far as US is concerned

[IAmGarethAdams]

Unfortunately however, this is unlikely to be admissible in any serious context.

Most postal services have absolutely no problem with letting you send an empty, open envelope to yourself; after which you have an open, registered and dated empty envelope which you can later fill with a pre-dated proof of invention and seal.

Re:As far as US is concerned

[thegrassyknowl]

you could just blog since the system is not first to file but first to invent.

Since when has that pesky little thing called 'prior art' stopped the patent office granting patents, and since when has it stopped the legal system upholding those patents when they are used to sue the pants off the original inventor or anyone who is using his invention free of charge?

Re:As far as US is concerned

[Instine]

"not abandoned, suppressed, or concealed"

Do you think thats an inclusive or an exclusive or?

Re:As far as US is concerned

[ddrichardson]

No.

Re:As far as US is concerned

[SLi]

I've been told that a common trick for large companies in some countries when they hit a thing they figure out might be patentable but they are not interested in patenting it is write a rudimentary description of it and file it as a patent. At least here they've told me that is enough to get the application published in some public PTO journal, but it won't be examined until the examination fee is paid, which they just "forget" to do. So then they have very officially published prior art to similar claims for free.

Re:As far as US is concerned

[beavioso]

IANAL, but I think you're trying to say that if this person writes it down in a journal, or a blog, that this person will not be able to stop someone from patenting it.

If that's what you're asserting, then you're wrong. There's a court case about prior art, which ruled something to the effect that prior art is anything that is accessible by the public (i.e. a PhD dissertation available through a public university library only needs to be available for access, and you do not need to prove that someone actually looked at it or read it after it's been cataloged). Once an examiner finds a reference with a varifiable date that predates a filing by one year, then 35 USC 102(b) is used and it creates a bar for patenting (e.g. the person seeking a patent cannot swear back and come up with evidence that they in fact invented more than one year ago).

In the end, I'm almost positive that blogs can be used as evidence, especially if they provide enough detail and motivation to do something. It might even be best to let archive.org to archive your blog, so that the dates have a second varifiable date attached to it.

Now, if all you meant to address, was that a person cannot get a patent after abandoning, suppressing, or concealing the subject matter, then that is correct. But, I haven't heard of many 102(g)(2) rejections.

Re:As far as US is concerned

[greensoap]

First off ---- I am not a lawyer. The information contained within should not be relied upon. If you are in need of legal counsel, please seek out a licensed attorney in your state. I do not intend to create an attorney/client relationship with any reader. I am just a law student and I am not qualified to provide legal advice, but hopefully I know a little more than the average joe.

The parent is flat out wrong. The USPTO, or a court, can use a publication under 102(b) or 102(e) to show that an invention is not novel. Regardless of the status of the a patent on the prior art. 102(g) is related to if an inventor is trying to get a patent on his invention, not whether a publication can be used against another as prior art.

Perhaps I should explain. If there exists a publication describing the subject of a patent application and that publication was published more than one year prior to the application (also applies to if the subject of the patent was in public use in the US or for sale in the US), it is an absolute bar to the patent. 35 USC Section 102(b).

Blogging about an invention is a publication and can be used to destroy novelty for further inventors. The problem is that the USPTO is less likely to find your art than if it is an issued patent or an application.

If the As a suggestion to the original post, maybe a provisional application would meet your desire:
http://www.uspto.gov/web/offices/pac/provapp.htm

Provisional applications are published after 18 months, and if the application matures into an issued patent, it will count as art from the day it was filed. You still have to meet some of the more stringent requirements though, so this may not be the right solution. Also, if you don't intend to get a patent on the idea, it is better to publish something as well as file the provisional. That way there is a publication farther back in time than just the publication of the application that will never mature into an issued patent.

But as I said before, the parent is wrong and misunderstands the import of 102(g). The designation of informative on that post is misplaced.

Re:As far as US is concerned

[dwye]

A person shall be entitled to a patent unless--

...

(g)(2) before such person's invention thereof, the invention was made in this country by another inventor who had not abandoned, suppressed, or concealed it. In determining priority of invention under this subsection, there shall be considered not only the respective dates of conception and reduction to practice of the invention, but also the reasonable diligence of one who was first to conceive and last to reduce to practice, from a time prior to conception by the other. (Emphasis added)

There has to be more to it than that, as well, or I could patent the horse collar and horseshoes, or even the use of subordinate clauses in Indo-European languages (as no one claims these "inventions" anymore). Assuming that I cannot, there must be a way to define "prior art" that recognizes ideas or practices in the public domain for ages, let alone to place one's patentable inventions into that public domain (which seems to be the article writer's goal).

BEWARE THE LINK

[IBBoard]

I've not clicked the link, but it's another Yahoo redirect link to an on.nimp.org address. Proceed with caution/insanity.

Re:BEWARE THE LINK

[IBBoard]

I realised after posting that it would appear as a post with no obvious parent and I should have been clearer. The "beware" is for the GP comment, not the summary (noticed the "parent" link under my comment ;) )

legal advice

[simonharvey]

What's your strategy for fighting against restrictive software patents?" Well I suppose the first would be not going to slashdot for legal advice, but then IANAL.

Re:legal advice

[sporkme]

My company is vastly intrigued by your refreshing concepts. Please email all of your code to: code-stealer slashtot.org (sorry for the obfuscation... can't be too careful these days). Once our firm receives your code, we will review it for viability and email you back with our offer. IANAL but I am totally "Al." You can trust everyone online, but be aware, many profitable arrangements are ruined by "best friends" who feign suspicion of legitimate businesses like mine, only to steal the data to resell to Korea. Just think about it. Tens of dollars await you my friend!

wtf legal advice on /.

Establishing Prior Art

[codegen]

I publish to a peer reviewed academic journal or conference. In most
of the conferences I'm involved in, we are always looking for
more industrial contributions.

sorry mate, somebody beat you to it.

[apodyopsis]

sorry mate, somebody beat you to it. the "toilet snorkel" is already patented..

http://www.totallyabsurd.com/toiletsnorkel.htm

Some considerations

[kanweg]

Publishing is a fine way to create prior art, effectively inhibiting any further patents for that very idea. However, make sure you add as much detail as possible to give a patent attorney as little leeway as possible.

If you blog it, the content may get lost. Is it possible for you to (crudely) implement your ideas and put that code on sourceforge or some other repository, together with a description. In the comments of the code you can elaborate on things not implemented. In such a repository, the ideas may be longer lived, and more people may see it (and if necessary bring it to the attention of a patent office).

Bert
Patent agent

Re:Some considerations

[n3tcat]

I thought too much detail would in effect give the patent troll a "way out" as they can then just change one of the many details and become unique.

Patent Commons vs Prior Art

[Anonymous Coward]

Patent Commons can be pooled and used defensively (aggressively in defence) to defend Open Source. A patent commons that defends one particular cause can help find infringement in order to scare off someone accusing others of infringement. It can cause a ceasefire.

Prior Art helps block patents in general, but as we saw with the Amazon shopping cart patent it can take years and effort (from New Zealanders) to remove them.

So Patent Commons is more powerful, but it involves the patent submission process, and it can be costly. I suggest the patent commons if you're got the resources.

(I am not a lawyer)

Re:Patent Commons vs Prior Art

[JavaRob]

How much does it cost to submit a patent? From what I understand, the legal documents are extremely complicated (so you need a lawyer) and the total cost will be somewhere from $5K up to $25K and possibly beyond depending on the complexity of the thing you're patenting (though the higher numbers will only come into play for highly-complex physical devices).

I absolutely agree that a donated patent can be quite useful -- but keep in mind that it's a sizable donation that *might* be valueless if there's some prior art you didn't find in your search.

The prior art approach seems like the way to go for most of us. It sounds like you just need to publish the idea with enough detail that someone could implement it, to qualify as prior art:

In most patent systems, in order to anticipate a claim, prior art is expected to provide a description sufficient to inform the average worker in the field (or the person skilled in the art) of some subject matter falling within the scope of the claim. Prior art must be available in some way to the public, and many countries require the information to be recorded in a fixed form somehow. A blog would seem to qualify; any way to more reliably prove the date of publication? Get it into the wayback machine, maybe?

Creating "Prior Art."

[mlwmohawk]

To establish prior art you need to publish. The problem is that magazines and journals are fairly selective about content because they have to pay or the content has to fit their market or be interesting in some way. Then there is credibility, if you write an article about recommendations or motor control methodology, something you've done as a hobby project, a magazine or journal may not choose to publish because it can not properly verify the content.

We need to create an "on-line" and perhaps paper "journal" that will accept all technical submissions and publish them in a way that fits the USPTO's definition of "published" to establish prior art. That way *all* ideas get covered easily.

I've written a lot of articles and only been published a few times, its hard to get published. There are lots of would be authors out there and a lot of subject matter being written about.

Re:Creating "Prior Art."

[ortholattice]

To establish prior art you need to publish. The problem is that magazines and journals are fairly selective about content because they have to pay or the content has to fit their market or be interesting in some way. Then there is credibility, if you write an article about recommendations or motor control methodology, something you've done as a hobby project, a magazine or journal may not choose to publish because it can not properly verify the content.

This is highly dependent on the journal. In the case of trade journals, a lot of them are actually dying for "real" articles buried in their mass of ads, and ads thinly disguised as articles, to attract readership. In the past I have submitted many "test cases" to a trade rag on electronics test and all of them were published without question. In some cases they were patentable ideas I specifically wanted to make public domain, because it wasn't worth it to me to go through the patent expense but I also didn't want to be prevented from using them in the future. As a side effect I became well-known in the field, very helpful for my consulting work.

On the other hand, I have submitted a couple of ideas to the "Design Ideas" section of EDN magazine. Both were rejected, even though I thought they were reasonable. I think the problem there is simply that the column is very popular and they have more submissions than they can publish, so it depends on the whim of the editor that day.

Anyway it depends on the journal. I guess my point is that if you're going to write it up anyway, might as well submit it and see what happens, doesn't cost anything. If it does get published, you'll make a name for yourself and have something for your CV, as well as establishing prior art. For best results find a little known or start-up trade rag in the field that's 90% ads. And if it doesn't get published, you'll still have your write-up for some on-line thing as you suggested.

Worthless advice, here's why.

[Yoozer]

Moreover, also write the idea, go to post office and mail it (sealed) to yourself (and may be also another one to your trusted legal advisor) by registered mail.

Do not do this. See http://www.copyrightauthority.com/poor-mans-copyright/
This trick works with anything else, too, so it worthless advice.

Re:Worthless advice, here's why.

[asuffield]

And if you want to accomplish something along those lines (proving that you had a document on a certain date) then that is precisely why notaries public exist in the first place. Take whatever it is along to your local notary and pay them their fee. In the eyes of the law, the document has now been proven to exist in your possession on that date. Unlike the ridiculous games with envelopes that you see in the movies, it actually works.

publish

[nguy]

You need to publish your invention in an archival format. Write it up for Dr. Dobbs or some other magazine.

Publishing it on the Internet is not enough; it doesn't count as prior art.

Even if you publish in an archival format, companies will often still patent almost the same thing and then worry about fighting it out in court. There are all sorts of ways of basically invalidating your publication for the purpose of counting as prior art, but it's still the best chance you have.

Patenting is pretty hopeless: it's enormously expensive, and trying to enforce a patent is even more costly. Patents are not useful for inventors or open source, they are only useful as legal ammunition for big companies and law firms to play games with.

Re:publish

[asdjlfhgas]

Who modded this up? Publishing on the internet does count as prior art, if it is accessible via archive.org or is dated such that there is generally no reason to believe it was backdated. A dated comment on slashdot.org would constitute prior art as there is the presumption that the date was not altered for deceptive intent. It may be more transient over the course of decades as opposed to an actual publication, but it certainly would be usable as prior art if found.

The us is currently "reforming" its patent laws

[plasmacutter]

I haven't seen anything on this in a while, but It passed one of our houses already.

From what little i've tracked on it, theyre "reforming" the patent system by handing it even more firmly to trolls:

e.g.:

switching to first to file
putting severe restrictions on who has standing to challenge a patent
putting severe restrictions on the period during which it may be challenged before it can never be invalidated period.
(goodbye EFF patent busting campaign anyone?)

I'm sure more "reforms" have and will be added to this bill.

Given this, many other horrid sellouts, and the recent news our new bush appointed head of the fed refuses to impose regulations economics experts nationwide say are absolutely necessary because he holds right wing dogma above economic reality (surprise surprise), I would like to be first to cry "WOMEN AND CHILDREN FIRST!"

the lifeboat is across the great lakes to the north folks, those of you in steerage might want to start making your way there now.

The nature of software does not support the....

[3seas]

...patent ideology

What is universally accepted as not being patentable:

abstract ideas,
natural law
physical phenomenon

Mathematical algorithms as often added to this but in essence are included in the three primary.

The reason these are so universally accepted as not being patentable is that its near impossible to enforce patents against such things.
i.e. patent on gravity... you cannot use gravity without paying me royalties. But if you try and avoid gravity and walk off a cliff, paying royalties to some fool is teh least of your concerns.

How these applies to software:

Abstract ideas is rather obvious as software is the art of abstraction creation and use if anything is to be considered an art of abstraction.

Natural law is that abstraction is naturally a characteristic of human ability. Without it we could do nothing more than preform like any other animal incapable of developing technology, medicine, clothing, etc..

Physical phenomenon - thru the application of non-physical based abstraction we cause physical movement. If I tell you to go to the store and get a gallon of Mayfield milk, I have used abstract ideas communicated and received by you to act upon. If I/we did not have such ability, Mayfield milk would not exist to begin with.

Now many claim that software is in essence mathematical algorithms, but the basis of computer technology, the way it works, defines this constraint as a computer is based on "calculating" defined in terms of numbers (binary based). However, a radio station (fm or am) uses a carrier signal that cares not of what is carried over the signal, be it music, news, talk, noise, etc... And in the same way this basis of computing can as well be used as a carrier of abstraction far beyond math. Yet the carried its still inherently made up of the three primary things universally consider not being patentable.

This software patent fraud that has been going on has, as any attempt to contradict physics and nature, very bad effects, only due to the inherent nature of abstraction the skill of abstraction manipulation (human use and ability to deceive via abstraction manipulation) there are those who have remained fooled by such deception and unfortunately are in positions to indirectly tell people they have to walk off a cliff rather than recognize they are human with inherent abstraction skill and as a natural human characteristic it is to be expected and even encouraged to be what we are and make the best of it.

There is a way to not only prove all of this but in so doing build up a foundation of common knowledge solution direction of anyone "skilled in the field", non novel, and other aspects supporting non-patent-ability.

Humans can be deceptive for a long time and in an environment that even proves otherwise. This might be called the "human denial factor". Examples of this are well known, ie. Galileo and teh exoneration in the early 1990 of his views. But a lot of good such exoneration does Galileo now. Likewise the Hindu-Arabic decimal system took three hundred years to over come the far more mathematical limiting roman numeral system. Lets face it, only a fool would think nothing cannot have value (re: zero place holder), yet accountants using roman numerals were elite.. (sound familiar?)

Anyways, this idea of asking for funds/donations for to support software patents, regardless of the claimed intent of obtaining such patents, is supporting dishonesty and does not help to resolved the real issue of genuinely recognizing the honest nature of software.

for more see: Abstraction Physics

Send them to me

[realkiwi]

Send your ideas to me I'll look after them free of charge!

Bill G.

oh wait maybe I shouldn't have signed...

Honest opinion?

[dpx420]

Sell out. Patenting the ideas is going to cost you time and money, whereas if they are truly worth the attention of the 'evil corporations' you stand to make a substantial gains from making them available to a company with the required resources to put them to use. Around here a higher than average subset adhere to strange personal religions that financial benefit from your own ingenuity is somehow immoral, and that the world is better off if real companies can't use these ideas and make them a practical reality (but that's ok, some guy sitting in his parents basement will knock off a buggy implementation in 10 years time, for freeeeeee man). You decide which of these outcomes you would rather see.

There are only small number of people for which simply blogging their ideas would:
a.) get them taken seriously, or even noticed at all,
b.) be worthwhile for them personally in terms of personal reputation and the longer term benefits of that.
Those are people who are already recognised within specific (usually comparatively small) technical communities, often with freelance careers for which reputation is important. I'm going to guess that you are not in this position because all this would be obvious to you otherwise. It is also rather naive to think that articles simply posted on a web page will stand up as prior art in courtroom against competent lawyers. This is simply pragmatism.

I'm going to make another assumption, that you are posting this question out of a genuine uncertainty of how to reap the results of your own creativeness (which is something that anyone deserves). If you are simply seeking group endorsement (perhaps subconsciously) then my reply probably hasn't been of much help to you.

build a prototype

[FudRucker]

build a prototype version .0.1 with well known friends of GNU/GPL and get it licensed under the GNU/GPL-3 and release it on SourceForge, then you know it is anchored somewhere...

Publish it as a patent application

[Woodpeckeruk]

The best way to ensure that i) it is published; ii) that it will be available for patent searching in perpetuity; and iii) you don't have to spend too much money, is to write it all down in detail and send it to a patent office as an application. Provided you pay the initial filing fees, all patent applications are published. You can speed up publication by specifically requesting it, rather than waiting for 18 months. As an example (since I am a UK patent attorney), what I would recommend is filing an application consisting of pretty much anything at all, paying the £100 fee and then letting the application lapse after publication. Even easier than that, simply post your invention disclosure to the UK-IPO, Newport, Gwent, UK including the words "I would like to apply for a patent for the following", making sure to include some contact details, and the UK-IPO will let you know what else they need.

If I remember right, the US has a similar 'defensive' patent system for doing precisely this, but I would rather leave the details to someone else.

Re:Publish it as a patent application

[theantipop]

This is the best answer so far in regards to the actual question. If you want your idea to be used as a defensive measure against future patents, you want it to get the most coverage by the examiners who will be searching. Previously published patents and patent applications are, with few exceptions, the primary source and the major focus in the course of the search of an examiner.

You may think publishing it somewhere on the internet will be good enough until you realize it's not feasible to search every whackjob blog out there for claim language that you probably haven't reproduced. Also, archive.org doesn't actually archive a lot of these small, personal websites so it would be very difficult to reliably date the material.

Ban this troll, and some other advice!

[greginterrupted]

Can the moderators look up this troll's IP address and ban it? Or even the range of IPs? We have all seen these nimp.org links on Slashdot for several days at least, and it is annoying. I hate to give the troll attention by posting this but we should be able to ban this hater from Slashdot with minimal effort. I would hate to think that it is a dupe account of a meaningful poster, but we can take that chance given the content that the troll is posting as Anonymous Coward.

I fell for the link in Firefox and their popup blocker didn't catch it, even though I'm running the latest version and I have auto-updates turned on. Good thing my speakers were turned off.

If it hits you:

1. It's easy enough to hit ctrl-alt-delete and bring up task manager to close the instance of your web browser if you run windows.
2. If you run linux you're probably knowledgeable enough to do a kill or kill -9.
3. The design of the nimp.org link actually helps you to close all of the web browser windows that pop up. The default behavior of windows is to "group similar taskbar buttons" together. You may have noticed this from the old tabless-web browsing when you had many instances of (gasp) IE or another browser open. Windows would annoyingly consolidate all of the windows into a single button on the taskbar. To switch between windows, you needed to click on the button and a drop-down menu would appear. So the nice thing about this grouping is that after enough nimp.org browser windows show up, windows consolidates them, and you can simply right click on the one button that represents the dozens of web browser windows. Select 'close' from the drop down menu and you're good. In fact, I keep the "group similar taskbar buttons" feature turned on just for trolls like the AC who posts the NIMP links. That, and tabbed browsing lets me run only one instance of a web browser instead of running one instance for each open web page. If you want to change the behavior, right click on the taskbar and select properties, and you'll see a menu with the 'group similar taskbar buttons' checkbox.

Just wanted to bring some attention to this instead of sweeping it under the rug, and also wanted to provide advice on how to close your browser easily if it hits you.

open source patent pools and patent agents

[j0nb0y]

I've seen some discussion on /. of creating open source patent pools to protect open source projects from patent infringement.

I want to point out to everyone thinking about such a thing that you *do* *not* have to go to law school and become an attorney to file prepare/file/prosecute patent applications. All you have to do is pass the patent bar exam.

A law degree is not required to sit for the patent bar exam. All you need is a technical background (which most /.ers have). There are a list of degrees that will automatically qualify you to sit for the patent bar. Included are various engineering and science degrees. Computer Science is included, but to automatically qualify, the school had to have had a certain accreditation at the time you graduated. You can also qualify with *any* computer science degree under category b option iv, but that requires a transcript along with course descriptions for 40 credits worth of qualifying courses.

If open source projects want to start filing a lot of patent applications, then having a project member become a patent agent would be a good idea.

I'm a first year law student, and I'm planning on taking the patent bar this summer.

IBM defensively published for 50 years;

[davidwr]

IBM Technical Disclosure Bulletin.

IP.COM provides a way for you to publish your own work and add it to their searchable prior art database.

You may also be interested in the Patent Commons.

Re:IBM defensively published for 50 years;

[Woodpeckeruk]

Although IBM technical disclosures are a valuable prior art source for patent examiners, IBM stopped publishing a few years ago (in order to get more flaky patents granted) and it now costs money for mere mortals to access it. IP.com charge a minimum of $200 for your publications, charge more for longer documents, and do not provide a guarantee that their database will be either searched by patent examiners or that your publication will always be around. The patent system, however, is guaranteed to always be around (at least for as long as patents are granted), so once a document is published there it is much much more likely to come up in a future patent search. A UK application at least is still cheaper, as all it costs you is £100 and the postage. As for the patent commons, that just looks like company PR to me. If a company genuinely wants to give everyone access to their patent, the best way is for them to abandon it.

Are you employed by someone?

[timothyf]

Better make sure your contract doesn't say anything about the company owning inventions that you come up with while you're employed by them. Bottom line, talk to a lawyer, not Slashdot.

Your employee probably already owns your ideas

[jpswensen]

As for whether you own your own ideas or not, many companies make you sign away patent right for anything you come up with (while working for them, whether on the job or not) that is related to their business interests. Sometimes very skilled or well-known people can weasle their way out of clauses like this, but for the most part you probably signed away your right to patent tech stuff when you signed you offer acceptance letter. go read it.

Those solutions wont work

[shaitand]

Apparently people still don't understand. You can patent an invention, you can copyright a tangible representation of an idea, but you can not protect an idea through any IP method.

Software is really a representation of an idea and should be copyrighted, not patented. Since patents are granted as if software were an invention you do still actually have to make the invention to qualify for a patent. I can scream my idea from the heavens all day long and that doesn't qualify as prior art, only implementing the idea would qualify as prior art.

Re:Those solutions wont work

[Woodpeckeruk]

Perhaps you should count yourself in amongst those people who still don't understand. The vast majority of patent applications describe inventions that have not been made. There is no legal requirement to 'make' the invention (the US abolished theirs over 100 years ago), only a requirement to describe it sufficiently to allow a person skilled in the art to make it. Software inventions are treated just the same as any other invention. If, as a programmer, someone gave you a really good idea about how to do X on a computer, would they need to tell you the exact code to enable you to do it? The other side of patents works the same way: provided a prior disclosure is 'enabling', it will prevent someone later on getting a patent for the invention.