Giving Up On Patents

From: John Cowan (Feb 22 2010, at 23:10)

I'm in favor of patents, provided the holders pay a substantial annual fee to the government (i.e. we the people) to retain the rights, and once lost, they are lost forever (subject to obvious limitations for simple errors, minors, etc. etc.)

"Tax bads, not goods." Legal privilege is a bad, but efficiency is served by selling a reasonable amount of it to the highest bidder.

[link]

From: Martin Probst (Feb 23 2010, at 01:26)

John: how would you decide on a price tag for "substantial"? The patents are supposed to protect small innovators/inventors, so you can't set the price to high. Setting it too low won't change anything at all.

It's also worth remembering that there is no such thing as a natural entitlement to patent protection. Patents are a tool we invented because we thought it would provide net benefit to the people. Turns out it doesn't, so we should better get rid of it.

I always wonder if this is also the case for non-software related fields. I guess not, software has protection by copyright law so it doesn't need the patents as bad, and e.g. bicycle manufacturers seem to be getting along ok with the patents.

[link]

From: James Fuller (Feb 23 2010, at 01:53)

I completely agree and would widen the scope beyond patent law ... as someone who has spent the last 5 years as family representative dealing with both the criminal and civil court cases of family member killed whilst riding his bike (was hit by a drunk driver) ... a few fundamental truths come out of any interaction within the legal system of most western democracies.

a) lawyers, solicitors, barristers, and expert witnesses are making money irregardless of what happens ... also it seems to be in their interest to continue litigation as they continue to make money as the clock continues ticking

b) there is sufficient artificial complexity in the legal system that resembles the 'lock in' akin to enterprise software development e.g. larger entities win by default

c) timescales in legal systems do not match the rhythm of natural human lifespan ... or put another way, everything takes far too long. The longer the time window of a legal challenge the more chance for duplication, loss of information, stale information, etc etc etc

d) law and their underlying systems do not scale well

The ultimate problem though is 'whats the alternative', even after 5 years working in a fundamentally broken system I can see that law and legal systems are a requirement for a civil and peaceful society. The problem is that the systems are broken ... effectively most law is operating on 18th century models and have evolved little over the ensuing 2 centuries.

Getting back to patent law ... I think its a system that needs revolution as well as new systems to underpin it to ensure a stable and scalable foundation ... but there is no incentive from within this group of people to effect change. I wonder where this change will come from ?

[link]

From: Sean Hogan (Feb 23 2010, at 03:05)

Give non-inventors a chance to have a vested interest in the success or failure of a patent.

Perhaps something like a float on the stock exchange.

- The inventor registers his (patent-pending) invention. The inventor has a 50% share in the patent by default. Shares become available (but the inventor can buy them all for some prohibitive amount).

- You can buy and sell shares in a patent which gives you a share of any licensing fees (assuming the patent is granted).

- Bounties proportional to the perceived value of the patent are available to parties that contribute to the refusal of a patent.

- In the event that a patent is granted but not all shares are sold, the remaining shares are owned by the government.

Probably a completely impractical scheme, although more viable now than in the past.

[link]

From: dr2chase (Feb 23 2010, at 04:46)

I'm of two minds here. First, that there is no use asking for individual action; there is an endless supply of lawyers willing to do patent work for a buck, so all you're asking is for people with a conscience, to be chumps. What we've got, is bad rules, and you fix bad rules by changing the rules.

Second, I fear you are proposing to throw out the baby with the bath water. I've worked in little companies that did acquire software patents, and been on the wrong end of them. Some of them are quite good -- for example, the Purify patent (5,535,329, Reed Hastings).

Some are not so good -- for example, the object code instrumentation patent (Parasoft, Kolawa, 5,581,696) which is mostly invalidated by an article by Steffen, in 1992 Software Practice and Experience, which is even cited in the Purify patent, and both patents have the same examiner -- Thomas Beausoliel. In this case, I think I might fault the examiner, or else the patent office's processes for reading, understanding, and remembering prior art.

The difficulty is that these bad patents are well-nigh impossible to fix, once granted; there is a presumption of validity, and the patent-holder gets various advantages in any challenge.

[link]

From: David Workman (Feb 23 2010, at 05:05)

My personal view on patents is that they are useful in fields where the timescales to develop a product are on a par with the timescale the patent protects the product for. For example, medical drug development can take decades. If a patent isn't obtained that covers the length of the development, competitors can reverse engineer the drug during trials, etc.

But for software, where an idea can go from conception to production in a year or less and the basic concepts are so misunderstood by so many (so 'inventive steps' can't be properly checked) then the patent system is definitely horribly broken and a major impediment to real progress.

[link]

From: Ed Sweeney (Feb 23 2010, at 07:17)

Thanks for posting the links. The current system is trash.

I think it will be a slow generational change for us stop believing that whoever shouts "pay me!" first and loudest should get rich even if their patent has no more substance to it than a kid calling 'shotgun' first to ride up front, ie we award patents to giant corporations for things that any high school kid would intuitively know to do facing a certain problem.

The point that the main players are mostly folks that don't actually make anything says it all.

[link]

From: JulesLt (Feb 23 2010, at 12:36)

While complete dismantling sounds an attractive solution, it's always a lot easier to be a revolutionary or a critic, than to create a system that works - and of course the big question with all proposals to remove IP laws is 'how do we fund the production of content, research, etc'.

As it happens, we do have some good comparisons of what society looks like with weak or no IP protection, and even more so with weak or state-owned copyright. Anecdotal evidence does suggest that societies with some form of IP protection tend to invest more in R&D that societies without it.

It's difficult to know if this is a chicken and egg situation i.e. is strong IP law and high R&D a symptom of a certain kind of society rather than a cause, for instance?

(Of course, a great form of IP protection is trade secrecy, and hiding all your source code and customers data on remote server is a very good way of doing this).

I note that buried in Stephan's piece is the suggestion that software should actually be excluded not just from patent (where I'd agree) but also from copyright.

That does strike me as wrong, because it completely removes the model by which a independent developer or software house can produce a game and sell it, without otherwise locking it down by DRM or tying it to a server-side component.

Equally, proposals like '10 years copyright' also strike me as wrong. They make sense for software - after all few firms will even support software older than 10 years, but I think a lot of the problem with copyright is around having a fixed term.

A book from a new author, a book from an established author, a stage musical, an orchestral composition and a pop song, all have very different business models - some expecting a short-term pay off, and some looking for a return over a decade, if not more.

In fact, one of the biggest problems with the creative industries is exactly the move towards a model based around short-term returns. The Blockbuster vs the Classic.

I think the other thing I really dislike is that underlying a lot of 'free' economics is a preference for Service based economics i.e. the value is not in the art or craft itself, but in performing, appearing, subscription, repeat business, selling secondary goods at added value i.e. not the value of the goods themselves as a piece of labour and production - otherwise you would be better just producing the goods.

To me it seems an extension of the Thatcher / Reagan era economics that devalued labour and production, into devaluing their intellectual equivalents.

(Hmm - how about the concept of IP as something that is purchased - you get an automatic free period, but can buy additional terms . . at a progressive rate??)

[link]

From: David (Feb 23 2010, at 17:01)

Is there a better model somewhere else? Canadian Intellectual Property Office, International Patent Law, other?

[link]

From: Jose_X (Feb 23 2010, at 20:45)

Some major problems that apply to all patents. [Software patents simply mean the situation is aggravated because there are a lot more inventors (lower barrier to entry) and the long monopoly period is that much more intolerable and to that many more people.]

Say you invent something wonderful. You outline it and then build the entire thing. You invest lots of time and energy (and possibly money, etc). You do the hard work.

Along comes someone that writes up your invention as their claim 1. Then they find 99 more claims that leverage claim 1.

Now in court, when someone goes to the trouble and costs to show that prior art exists for claim 1, the other 99 claims still stand despite being normal applications of claim 1 to new features of current society.

"All" the hard work may have been done by claim 1 and by the proof of concept invention, but now not even that inventor can use the invention for new interesting things (covered in 99 claims that would survive). The inventor can't extend their invention nearly as might happen naturally and into new areas because, while s/he was developing it up to point A, someone else with more savvy and money was patenting 99 important extensions to it.

Of course, a second major problem with patent claims is simply that they are so broad. It isn't just this last inventor that would be affected, but potential many many others that were working on things only related in that a few key properties were shared (captured in claim 1 or claim X) but most details where different. All of these people will have their work also go into the trash, especially if they finish after the patent application was filed.

It is the person with the broad idea that gets the patent first and can (for the next 20!!! years) stop or control all brilliant ideas that take off from that basic description conceived one fine rainy day when Mr. or Mrs. Patent Leech was browsing web forums and noticed something interesting. This means that before you have your invention stolen from you as just described, you actually might even have been pre-empted by someone with little more than the vague idea and huge patent ambitions. It takes much less effort to jot down the general idea and submit it to the USPTO as the would-be enforceable patent claim (and send as well some bogus detailed implementation description you never intend to implement and which might actually be a horrible design) than it is to actually build a decent product of that idea. So people building things will be beat to the punch by those writing up the summary.

Another huge problem is that independent invention is not recognized.

Another huge problem is that 20 years is way too long for many types of patents (eg, most things done in software), even assuming these "inventions" were legitimate contributions to society in the first place and were significantly narrower than the grossest of patent claims.

For the case of software, we have another obvious problem. Patents are incompatible with open source and disliked by most open source developers. Clearly open source is a great contribution to society. That patents damage this contribution is a clear strike against it or against any law that claimed to "promote the progress of science and useful arts."

Why must a monopoly be granted? There are alternatives. Why hurt everyone else while at the same time promoting extreme laziness on the part of the monopoly holder? Why not find other ways to give breaks and small levers to the inventor of record without tying up everyone else's hands.. and for 20!!! years, no less?

Why pretend that anyone is that smart that what they did would not be done as needed by the rest of society and usually within a short period of time? Obviously someone had to be first. Someone is always first. Even the greatest most prolific genius can't be first except in a small number of cases (and only if s/he gets into gear writing up patents rather than developing the produce or ideas further); however, that genius can then be barred from a huge number of other inventions, even inventions that leveraged significant amounts of the work of that genius. Sorry, but the super genius was not first in those cases. His/her wrist could not write fast enough to beat all others also writing patent claims. Now super genius can be barred for the next 20 years. [Society and progress frown from the sidelines.] Of course, the really sad part isn't that this one super genius would be barred form many inventions, but that a great many other very smart and dedicated people (including some other super geniuses) would also be barred from most of these as well.

Since when is it a good strategy to tie the hands of everyone else to come along that path over the next 20 years? In particular, the person most capable of developing that idea might not be the first to it or the one to file the patent. No one invents in a vacuum. Why does 100% credit and 20 year monopoly automatically go to the first to file the general description (or any description for that matter)? Why a monopoly? why total credit and full exclusion of all others? Why 20 years, no questions asked?

Many amazing contributions to science, math, music, literature, law, business, etc, have been done by people expecting little more than some money and perhaps fame. Creation as well as recognition are huge motivators. This is clearly seen in quality FOSS -- a phenomenon most patent supporters would not long ago have claimed was essentially impossible -- oops.

[link]

From: Jose_X (Feb 23 2010, at 21:38)

It's simple to come up with many ways to promote sane innovation once you get beyond the myth that "a patent is a gift from a god who thus deserves 20 years of absolute control over the described broad class of inventions."

It should also be pointed out that even without any special rewards, many great things will get created, especially if the cost to create them can be distributed among many collaborators. [Software obviously comes to mind.]

Growth is incremental. Always has been. Nothing has changed for patent authors except that the potential for easy riches and a degree of absolute power has made it easy for their judgment to falter and remain in a coma. In fact, the main problems come from a small number of people and corporations. And, certainly, not every patent monopoly is as bad as the next.

BTW, it only makes sense to taylor the incentive to the class of inventions. It also seems reasonable to consider tiers of some sort. It could also be useful to allow options to the inventor on the incentive they could accept (these would be balanced). Also, we should look towards monopolies as a last resort since likely we are failing if we have to resort to this except narrowly and for not too long of a period of time.

[link]

From: Robert Young (Feb 24 2010, at 07:28)

The issue isn't with patents, per se, which can be used for the intended purpose of rewarding innovation and expense of development. The last is key to why a patent system was created: the idea for an invention might require a few hours sweat of the brow, but bringing "the thing" to market could be quite expensive. Patents were allowed for ideas which were made manifest as physical things produced.

As it stands, Pythagoras could patent this theorem. That is not what the Founders had in mind.

[link]

From: MIke (Feb 25 2010, at 16:30)

Patents are working the way they always have. If you look at the history of any industry during its inception you see the same flurry of patents and the same grumbling about the system. Things will settle down.

By the way, the U.S. and Canadian patent systems are for all practical purposes identical.

[link]

From: Robert Young (Feb 26 2010, at 05:42)

@Mike:

Patents are working the way they always have

No, they are not, and that's the crux of the matter. The patent office(s) are now issuing patents to algorithms and processes. That was not the case prior, nor was it the intent of the founders, who did not include such in their law. Algorithms and processes existed back then, and have for all time; the insertion of a computer does not change the fact. The patent process has been perverted.

[link]

From: Robert Young (Mar 01 2010, at 05:53)

>> The "algorithms" were implemented in and implicit in the mechanical components of the machines.

And it was just this implementation that justifies patentabililty. That's the whole point. Just drawing on a screen by some algorithm means that paintings, drawings, and a whole manner of other expressions become patentable.

>> Ultimately it just became silly to require machines to be built to get patents for processes.

Nothing silly about it. For all the crowing about "original intent" from the Right Wingnuts, which they credit only when it suits their ends, the "original intent" of the founders was clear: it had to be a thing, you had to make the thing (at least a plan for doing so), and algorithms and processes were not patentable.

Reductio ad absurdum is a perfectly fine logic/proof mechanism. In the case of patents of software or other algorithms, would you countenance Pythagoras or E=mC^2 being patented?? If not, then how is either of those materially different from single-click or any of the myriad "business process" patents?? Again, the founders had the opportunity to specify that algorithms were patentable; such existed in their time. They chose not to. Who are we to overthrow our founders?

[link]

From: Adrian Sampaleanu (Mar 01 2010, at 19:25)

Every so many years you we have scattered discussion threads where it becomes obviously clear that the patent system is a mess and should be scrapped. We complain over and over, but nothing changes - or it does, for the worse. Why is it that there is no forum through which all the people that see what the fix should be can be heard?

A union of software engineers and other similarly impeded IP producers could, even if not generally useful, possibly provide a common voice. I know I would pay the dues with the cleanup of the patent mess as its only platform. Who can lead the way here?

After close to two decades in the business it's sad to see that innovation is so successfully blocked by vague and broadly scoped patents. We're in an age where we've got so much prior art to build on and yet we're held back from doing so by patent shackles. Obama wants to add stimulus to the economy? Kill software and process patents and we will see a huge boost in innovation.

So seriously - how do we band together? We need representation!

[link]

From: patno (Mar 08 2010, at 05:10)

Logic is not patentable!

Software neither needs patent attorneys, nor artists or parasites.

The software-patent-discussion suddenly is led by voices heatedly, which are not sufficiently familiar with the issue. This pushes me to despair.

Maybe I can contribute something to a better understanding. In 1982, as CEO of a system-house and expert of patent-law, I got a member of a committee to formulate a German software-patent-right. Initiated by Prof. Erich Häußer, the then president of the German patent-office, all its experts were busy. Also patent attorneys and other scientific experts (Fraunhofer, Max Plank) were involved.

There were many conferences and public discussions in order to accomplish the law. However, all the big players at that time (IBM, Bull, Siemens, Sperry, ICL) were not interested at all. One held it for useless. Presumably, one wanted to avoid the employee-inventor-protection-issue. - The same was then in the USA!

Nevertheless, the topic was further-discussed by us still for decades. With Prof. Häußer I was in contact until shortly before his death. The result is:

The industrial legal-protection in Germany and Europe distinguishes conceptually between the patent-right on the one hand and the copyright on the other.

1. There can be patents for a mechanism, a Manu faction-procedure, a recipe or another form of the materialization of an invention. The term is restricted, the protection inclusively - Monopoly: If one is not hindered by other monopolists (frequently through patent-claims and miscellaneous tactical meanness), one gets a monopoly.

2.For immaterial results like books, music, pictures etc. there only is copyright-protection. It runs essentially longer, but the protection is not so extensive. On hits, you can live yourself, sometimes, only the grandchildren can live on it.

Software as immaterial property is subject to the copyright at best. This was the point of view of Prof. Häußer.

3. However, it gets even worse: Programming artists make themselves suffer and the rest of mankind. Until now not-mathematical programmers are an underestimated time-bomb (David Parnass). One must keep artists away from the software-production!

Programming is no art, it is a new mathematical discipline.

However, mathematical knowledge, procedures and formulas are subject to the "free-stand-necessity": They cannot be protected by law. They are created by realization-joy and are given to those, who understand to enjoy them!

More than 25 years of expert-discussion of hundred of bill-blueprints showed:

·Logic is not patentable.

·Software-patent-law cannot formulated conclusively, it remains patchwork.

·"technicity", the kernel of current software-patent-blueprints, is an insubstantial phantom of hungry patent-attorney-brains.

·By a prospective law, the free utilization of the current mathematical knowledge of the mankind might become impossible.

·It is the Open-Source-Killer par excellence!

The technical and economic advance would be sensitively hit: blockade of economic growth, a relapse to early-forms of the mercantilism.

Therefore it must be prevented, that the existence-basis of the mankind is withdrawn by the injudiciousness or the greed of some irresponsible guys.

Our future is endangered in the kernel!

[link]