zhengj2007
commented
9 years ago From alanruttenberg@gmail.com on September 20, 2011 12:03:19
On Tue, Sep 20, 2011 at 11:09 AM, Bjoern Peters bpeters@liai.org wrote:
There may be different uses for image. The original definition was meant to capture the kind of data produced by 'imaging' techniques' such as photography, x-ray machines, electron microscopy.
Which the current definition handles quite well.
How about:
'physical 2D-image'=def: 'a data item that consists of a 2-dimensional representation of the spatial arrangement of one or more material entities and one or more of its inherent qualities'
This is a much broader term. It includes paintings, for instance. Is that your intention? It would also include encrypted images (which display as snow). There is a lot buried in the "representation" word there - precisely what you consider to be adequate to qualify as a representation is not, I think, currently in IAO. In addition, there is no need that an image is of one or more material entities - it can be a visualization of only qualities.
Based on what I've heard so far, my priority is on getting an appropriate label for the class we have, which I will consider to have an adequate definition until there is a viable alternative. I will note that it does not include nonlinear transformations such as what one might see in a google captcha, nor nonlinear projections such as one in which polar coordinates are layed out on a rectangular grid. I am content with the former. The latter might require some thought as to whether to broaden our term or to define a more general term.
To map out the larger "image" term is a big job, I think, but would be aided by the creation of a list of exemplars that help delineate the scope.
ps. Please respond to issue reports by adding comments to the issue on the issues list so we can more easily track them.
-Alan
- Bjoern
Do you mean that we should think about defining 'image' through reference to perception?
Janna
On Tue, Sep 20, 2011 at 3:27 PM, Barry Smith phismith@buffalo.edu wrote:
I think the solution is to have image defined in the MFO (Mental Functioning Ontology), and then use that definition as a starting point for IAO 'digital image artifact'
On Mon, Sep 19, 2011 at 11:58 PM, information-artifact-ontology@googlecode.com wrote:
Status: Accepted Owner: alanruttenberg Labels: Type-Defect Priority-Medium
New issue #121 by alanruttenberg: Problems with definition of "image"
https://code.google.com/p/information-artifact-ontology/issues/detail?id=121 >>> Raised in a discussion relating to IAO at the UB Applied Ontology meeting, by Barry Smith.
On Mon, Sep 19, 2011 at 6:01 PM, Barry Smith phismith@buffalo.edu wrote:
This breaks multiple rules
definition "An image is an affine projection to a two dimensional surface, of measurements of some quality of an entity or entities repeated at regular intervals across a spatial range, where the measurements are represented as color and luminosity on the projected on surface."@en
It is not Aristotelian It uses terms far more complex than the term to be defined
Fair criticisms. I will certainly change the label to be more specific. I do think the type captures an important class of information content entity, so the tasks are to:
a) Find a better label for this class
b) See whether we can come up with a definition that corresponds better to the common meaning of "image" as it would be used in the information sense (I'm not sure your shadow qualifies, but I am thinking about it).
c) Consider whether the definition can be simplified. Given the choice of being clear and "good enough to be wrong" and using simpler terms (some issues here - I might challenge you to define a "simpler than" comparison that can operate across terms as different than these) I would err on the side of being clear as to what instances the term denotes.
I will add an issue to our issues list to record this after I hit send.
-Alan
-- information-ontology@googlegroups.com To change settings, visit http://groups.google.com/group/information-ontology >>>
-- information-ontology@googlegroups.com To change settings, visit http://groups.google.com/group/information-ontology >>
-- information-ontology@googlegroups.com To change settings, visit http://groups.google.com/group/information-ontology >
information-ontology@googlegroups.com To change settings, visit http://groups.google.com/group/information-ontology >
From alanruttenberg@gmail.com on September 19, 2011 23:57:42
Raised in a discussion relating to IAO at the UB Applied Ontology meeting, by Barry Smith.
On Mon, Sep 19, 2011 at 6:01 PM, Barry Smith phismith@buffalo.edu wrote:
Fair criticisms. I will certainly change the label to be more specific. I do think the type captures an important class of information content entity, so the tasks are to:
a) Find a better label for this class
b) See whether we can come up with a definition that corresponds better to the common meaning of "image" as it would be used in the information sense (I'm not sure your shadow qualifies, but I am thinking about it).
c) Consider whether the definition can be simplified. Given the choice of being clear and "good enough to be wrong" and using simpler terms (some issues here - I might challenge you to define a "simpler than" comparison that can operate across terms as different than these) I would err on the side of being clear as to what instances the term denotes.
I will add an issue to our issues list to record this after I hit send.
-Alan
Original issue: http://code.google.com/p/information-artifact-ontology/issues/detail?id=121