AW: [Insight-users] Fuzzy classification & 3D : PATENTS & OPEN SOURCE

Wed Aug 30 09:53:54 EDT 2006

Hi Zein,

    Filters in ITK are *NOT*  "...Slowly becoming Patented..."

The FuzzyConnectedness filters were patented before being added to the
ITK toolkit. It was an unfortunate oversight of our early process for
accepting contributions to the Toolkit.

Now that we have the Insight Journal we are more selective on the
background of the contributions that are accepted into the Toolkit.

Once we realized that several algorithms were patented, we moved them
in to the Code/Patented directory and added the CMake options for
blocking them. The purpose is to prevent Patented algorithms from
infecting the rest of the toolkit.  This was done in October 2004,
so it is not new at all. They have been clearly indicated as Patented
algorithms for almost two years now.

There is *NO* intention to remove the Freedom of the ITK Toolkit, at
least on the part of ITK Developers and the Insight Software Consortium.

However, we *cannot* control what the authors of algorithms do with the
Intellectual Property of their work.

Please note that there are two aspects of intellectual property that
are relevant for ITK:

              1) Patents
              2) Copyrights

We can regulate the details of Copyrights to make sure that the toolkit
is open to the public on the terms of the BSD-License. Copyright law
relates to the "expression of ideas" while Patent law relates to the
"ideas" themselves.

The big problem that our community is confronting is the double standard
of research groups for wanting to *"Sell the cake, and eat it"*. This is
seen in particular in the dubious practice of *Patenting and Publishing*
E.g. Filing for a patent and the next day submitting the same work for
a publication.

Such detrimental practice *must* be repudiated by the community, because
it is simply a mechanism for using Technical Journals and Conferences as
*Marketing* platforms.  By publishing in a Journal, authors are inviting
the community to use their algorithms,... but then, later come with the
news that the use of that algorithm also requires a license, or a an
agreement for paying royalties.

As long as Universities and Companies continue to celebrate the Frenzy
for Patenting the simplest spark of an idea, we will have to live in
paranoia in order to continue supporting Open Source and Open Science.

We have *too many bad things* to say about Patents, and look forward to
remove patented algorithms from the ITK Toolkit.  The Visualization
Toolkit VTK already went through a process of "Purification" and removed
all Patented algorithms from the code. VTK can claim now to be "Patent"
free. We should strive to reach the same status for the ITK Toolkit.

In the meantime it is a task of our community at large to demand for
disclosure of Intellectual Property status on *every* paper that is
submitted to a Journal.  For example, a condition for publication
should be for the authors to submit a letter stating that "to the best
of their knowledge" this Algorithm X described in this paper is free
from Intellectual Property burdens.

Of course, if authors want to still *Patent and Publish*, then Journals
should charge them extra fees for "Advertisement", because that's the
status that such a paper will have, and of course that status must be
made clear to readers and subscribers. E.g. Those papers should go in
the "Marketing" section of a Journal issue.

You as a subscriber to Journal, or as a reviewer of Papers must do your
part on demanding that authors disclose any Intellectual Property
string that may be attached to their papers.

You as a author of papers must do your part by voluntarily disclosing
the IP status of your submitted work. So that this practice becomes
standard in the publishing process.

You as the user of Open Source should do your part by *NOT USING*
patented algorithms, and by discouraging other from using those
algorithms.

You as contributor to Open Source should do your part by proposing
and contributing alternative algorithms that are not Patented and
could do the equivalent task of Patented algorithms.

    Regards,

        Luis

------------------
Zein Salah wrote:
> Hello Luis,
> 
> Since when are the Fuzzy Connectedness filters patented ? I am wondering 
> why some filters of ITK (and VTK) are slowly becoming patented! Is there 
> actually a tendence towards "unfreeing" these toolkits?
> 
> Best wishes,
> Zein
> 
> ->8<---------->8<---------->8<---------->8<---------->8<---------->8<-
> Zein Salah 
> Universität Tübingen, WSI/GRIS, Sand 14, 72076 Tübingen 
> Email: salah at gris.uni-tuebingen.de
> Tel.: (07071) 29 75464 (GRIS) , (07071) 25 69 43 (privat) 
> Fax: (07071) 29 54 66
> 
> 
> 
>>-----Ursprüngliche Nachricht-----
>>Von: insight-users-bounces+salah=gris.uni-tuebingen.de at itk.org
>>[mailto:insight-users-bounces+salah=gris.uni-tuebingen.de at itk.org]Im
>>Auftrag von Luis Ibanez
>>Gesendet: Dienstag, 29. August 2006 15:16
>>An: sami_bourouis at yahoo.fr
>>Cc: Insight Users
>>Betreff: [Insight-users] Fuzzy classification & 3D
>>
>>
>>
>>Hi Bourouis,
>>
>>Yes,
>>you can apply the FuzzyConnectedness segmentation methods in 
>>3D images.
>>
>>
>>Note however that those methods are *Patented* and you are not allowed
>>to use them without a license from the patent holder.
>>
>>Note also that academic research is not excluded from the activities
>>where patent rights must be respected.
>>
>>If you need Fuzzy-like segmentations, I would encourage you to use
>>LevelSets instead of FuzzyConnectedness, so you don't have to deal
>>with the obscurities of Intellectual Property.
>>
>>
>>    Regards,
>>
>>
>>       Luis
>>
>>
>>-------------------
>>bourouis sami wrote:
>>
>>>Hi Luis,
>>>thank you very much for your answer about 
>>
>>itkReadITKImage3DSegmentShowVTK.
>>
>>>Please, I have another question about classification method.
>>>Can I apply classification method and Fuzzy segmentation method 
>>>whose described in ItkSoftwareGuide.pdf *to 3D image*.
>>> 
>>>thank you
>>>
>>
>>_______________________________________________
>>Insight-users mailing list
>>Insight-users at itk.org
>>http://www.itk.org/mailman/listinfo/insight-users
>>
>>
> 
> 
> 