patenting geometry

hi guys, this may be of interest to all us.

http://www.nzarchitecture.com/blog/index.php/2011/08/29/patenting-g...

Replies to This Discussion

Permalink Reply by Evert Amador on September 4, 2011 at 2:50pm

After thinking about this for a couple of days since I first read Daniel's original post and subsequently Daniel Piker's post, I personally think that what they have done is wrong.

I'm very grateful to the papers they've published and I've had the opportunity to read and I understand why people/companies patent inventions (not discoveries) as a way to get a payback for their investments but I don't agree this is the case for researchers and academics, for start and for consultants/developers who are supposed to get paid for solving a specific problem, similar to me when, as consultant, I get paid for solving/designing something, I don't invent the thing I design it using experience and knowledge I've acquainted throughout my life, including perhaps, reading papers published by academics. For me it goes beyond good ethics, scientific knowledge and impediments for the advance of culture as a whole.

It is even worse when you consider that they've only found one way to optimised freeform surfaces and because of that they've patented the results of probably many other ways (kangaroo physics being one) to solve the same problem. It seems they didn't think it through carefully as with the continuos advances of computation and computational geometry there will be other ways to "find" " get to" this kind of results in the future and the wording of the patents themselves is very ambiguous (at least the english translation) and as I've said before when they get paid for consultancy and probably to follow a research agenda in the first place.

I have to say I'm very disappointed reading about this sort of things but at the same time, I hope this is a good opportunity for all of us to consider it and raise our voices about what we all spent our time and effort doing because for me it's definitely worthwhile doing.

Cheers

Evert

Permalink Reply by Nathan Miller on September 5, 2011 at 12:25pm

Here is my unknowing "violation" circa 2009 with source code available in the custom component (ghx needs updating) http://nmillerarch.blogspot.com/2009/10/quadrangular-flat-panels-by...

and a more recent video: http://www.vimeo.com/21027915

No literature related to those involved with Evolute was consulted here... discovering flat panel solutions for free-from shapes seems to be a common and obvious enough of a process.

I haven't built anything of significance with this particular approach (except maybe a study model or two)... Do I owe a % of the construction cost for chipboard and glue?

According to the Evolute post, they are patenting the constructed result, not the process... Who do we owe for construction innovations such as the arch, the vault, the flying buttress....

Permalink Reply by Steve Lewis on September 5, 2011 at 6:51pm

Hey Nathan,

Yeah I think this whole patent thing stinks. it's not like anyone is using their algorithms, connection, details or software to build a structure. My main gripe is that it is about as a patent troll bullying the hobbyist to innovate. In the current economic climate we need as much innovation as possible.

I would argue that a lot of free form geometry is derived from classical mathematical or biomimicry precedents. I am sure evolute benefited from many years or earlier hobbyists work before hand.

Permalink Reply by Nathan Miller on September 5, 2011 at 4:16pm

To add another line of dialogue to this... I wanted to provide the means to compare/contrast the Evolute patent to other precedents....

Taking the tools out of the equation, Evolute has been granted a patent for a "SUPPORTING STRUCTURE FOR FREEFORM SURFACES IN BUILDINGS" The full description of the patent (including diagrams) of this supporting structure can be found here...

We can compare this to a patent for Geodesic Structures created by Bucky himself... (This was mentioned in the discussion section in the initial link...)

Also for comparison, here is another example of a BirdAir patent for a tension braced dome structure...

Permalink Reply by Michael Pryor on September 6, 2011 at 12:30pm

Permalink Reply by robert cervellione on September 6, 2011 at 8:27pm

I think a good president for all this is the marching cubes patent . A lot of people were thought it was ridiculous to patent the marching cube algorithm, but it lead to the generation of the marching tetrahedron which was a better solution. they both produced the same type of geometry but using different approaches . because marching tetrahedrons was a different solution (even though it made the same geometry) it did not infringe on the cube patent.

Permalink Reply by Vicente Soler on September 7, 2011 at 1:59am

I recommend you read Frédéric Bastiat's 1850 essay "That Which Is Seen, and That Which Is Not Seen".

You are only taking into account the seen effects of the patent, but you are ignoring what would had happened if all the creative energy that was directed into circumventing the patent was directed into something else.

Permalink Reply by Ángel Linares on September 7, 2011 at 4:19am

And with this context exposed, we could now introduce the "Open Source Architecture" concept...¬¬. People talking about open designs, open architecture process, etc, and Evolute trying to patent free form architecture ¬¬....

Come on!, we live in 2011, is not time to patent everything or make humanity run slowly to the future to make more money. You can't patent pancakes...you could patent a really close process to cook it, but if I develop a new cooking process with some small variations in that process, this will not break your patent (IMO)...If I was Daniel I wasn't put that lines in my code...

We have enough with stupid Apple patents...take this rubbish out of architecture and creative design circle, please...

Keep working hard guys...this doesn't worth...:(

Permalink Reply by Morten Norman Lund on September 7, 2011 at 5:28am

Anybody up for designing and building a piece of architecture (nothing to big) that violates the patent with the sole purpose of having this settled in a court of law.. just to set the example that architecture can’t be patented..

We risk that this turns into the standard of the industry.. Architecture and computational design should be (remain) an open source industry..

Best regards Morten (an engineer in architecture working in R&D for a architecture firm)

Permalink Reply by Charles C Vincent on September 7, 2011 at 5:32am

Maybe one of these guys could have a hint:

http://www.ted.com/talks/cameron_sinclair_on_open_source_architectu...

http://www.ted.com/talks/lang/eng/yochai_benkler_on_the_new_open_so...

http://blog.ted.com/2011/05/20/fellows-friday-with-gautam-john/

http://www.ted.com/talks/clay_shirky_on_institutions_versus_collabo...

Permalink Reply by Milos Dimcic on September 7, 2011 at 5:40am

We can all agree this is a bit ridiculous...why?

Well, Evolute uses known optimization techniques (like Gauss-Newton Algorithm, Conjugate Gradient Method and similar) to optimize a function for a set of variables (mesh vertices in this case). As a result they get planar faces and they want to have an exclusive right on them. Hmmm...a bit tricky...

If I present a problem in form a function and use some (known) mathematical methods for the optimization of that function it is not logical to claim the right for everything that resembles my result. Ill give you an example:

In my PhD I used Voronoi Diagrams, relaxed them over a free form surface and used Genetic Algorithms to optimize these structures statically (I provided the link down). So since I only combined known methods, I dont think I have the (moral) right to patent the structure that was an end result, formed by relaxing a Voronoi Diagram (which i called Voronax)...and I never will.

I dont think this kind of patenting can "survive" but it raises questions...patenting a method I can understand (not really, but OK)...but patenting a geometry of a built structure (end result) of such broad spectrum, regardless of how I got to it (even if I live on an island, without internet, and have never heard about Evolute)...no...

What is there to do?

link:

( http://www.programmingarchitecture.com/index.php? option=com_conten... )

Milos Dimcic

Dr.-Ing.

www.programmingarchitecture.com

Permalink Reply by Florin Isvoranu on September 7, 2011 at 8:15am