[semantics-public] [now semantics-public] Unable to meet today on X3D semantics; looking ahead
John Carlson
yottzumm at gmail.com
Mon Dec 30 17:24:20 PST 2019
So it sounds like we will be using SPARQL engine, with something like Alexa
fronting it. That is okay with me, as long as we have a large supply of
appropriately indexed scenegraphs.
Perhaps we should step back and examine my use cases for X3D?
1. Generate a virtual world with an appropriate level of entropy, possibly
with a very basic assistant, ala Clippy, followed by transformations.
Ideally I would have a sliding sensor which would adjust the level of
entropy.
2. Have a large supply of alternate scenegraphs at appropriate levels of
entropy (LOE) or alternatively have a way to generate them on the fly. See
“No Man’s Sky.”
3. Support laser (coherent), incoherent (noise), and mixed lighting.
Adjust lighting with sensor.
4. Provide for geometry to go from solid to liquid to gas and back
controlled by sensor.
5. Modify objects that are procedurally generated to move from geometric
to chaotic to stochastic to quantum generating equations.
What I really want to do is *show* people what I’m thinking instead of just
talking about it. Right now it still seems best to talk about it.
Beyond merely X3D, I would like to show this sliding entropy in language,
emotion and HAnim.
What’s at the root of this is generating a world with a mixture of order
and disorder(chance).
So it seems like I am still querying the ontology to see if such features
as entropy, randomness, laser light etc exist. Right now, it would seem
like searching tooltips would be best. How might I search for them with
SPARQL or preferably Alexa?
I have really mostly been on my rump. I’m guessing if my nurse doesn’t
get me my medication by tomorrow, I’m going to have a long night or two.
On Mon, Dec 30, 2019 at 5:49 PM Don Brutzman <brutzman at nps.edu> wrote:
> Hi John.
>
> Shifting this general discussion from semantics at web3d.org (member-only
> meeting scheduling) to semantics-public list.
>
> I was referring to separate geometric ontology work being referred to on
> the CAD mailing list. Some interesting references were posted there.
>
> Everything you describe sounds interesting but is far too abstract for me
> to ponder. I think we get our best mileage out of focusing on:
> - performing X3D scene-graph queries (with a modest start) to produce
> properties of interest (needs work)
> - performing metadata queries (needs work)
> - performing multiple-namespace queries with other information (such as
> separately derived geometric properties) in combination with X3D triples.
>
> At that point we will be peeking up on the plateau of general 3D search.
> For upcoming efforts, staying focused on improving X3D queries can best
> help.
>
> As quoted on the X3D Ontology page:
>
> * "The answer to your question is the response to the query." Jim Hendler
> and Dean Allemang
>
> ... and a converse is true also, if your query can't get a response then
> maybe the question isn't well formed. Suggest we stay focused on trying to
> "ask the right questions" using SPARQL and the X3D Ontology in order to
> best understand how to effectively apply it.
>
> So, for example, deconstructing "bouncy" might mean writing an X3D SPARQL
> queries for
> - "is there a PositionInterpolator that starts at one SFVec3f position
> with Y=0, goes up in +Y direction, then returns to Y=0 height."
>
> and moving on to "bouncing" would add something like
> - "is there a TimeSensor connecting the PositionInterpolator to a
> Transform via a set of ROUTEs?" and "is the TimeSensor triggered/active?"
> etc.
>
> Defining animation queries like these is certainly within the realm of the
> possible today. Some would be great diagnostics.
>
> Sticking to practical measurable concepts like these seems quite
> valuable. We don't need others agreement to characterize the essentials of
> what exists in X3D now. There is a whole set of scene-graph model
> characteristics we might be able to figure out answerable queries like
> - what shapes share the same color?
> - what shapes are transparent?
> - what shapes are invisible due to scene-graph structure (hidden by
> Switch, follow-on child in ProtoBody, within LOD etc.)
> - what functional animation chains are present (optional trigger,
> TimeSensor, Interpolator/Sequencer, target node, connecting ROUTEs
> throughout)
> - what nonfunctional animation chains are present (i.e.
> Interpolator/Sequencer without the above)?
> - what functional/nonfunctional animation chains use a Script or
> ProtoInstance instead?
> - does metadata exist? do metadata nodes have references?
> - what shapes are red (green, blue, a certain color)
> - do any geometry nodes share the same coordinates?
> - are there any textures? any transformed textures?
> - are there any Viewpoint nodes? Inline nodes? etc.
> - are there any (possibly insecure) url values with http vice https?
> - are there any url values with possibly insecure/blacklisted addresses?
> - are there any nodes present with missing (or perhaps only default)
> values?
> etc. etc.
>
> So, lots of practical inquiry lies ahead. Have fun with X3D semantics! 8)
>
>
> On 12/30/2019 9:46 AM, John Carlson wrote:
> > Previously we said we are working on: "Reasoning and Inference inquiries
> include Structural, Conceptual, Functional and additional aspects of 3D
> models." May I suggest that Conceptual, which previously meant Spatial
> among other things, be moved up, and Symbolic or Sign (Semiotic might be
> okay) aspect be added below and a new Spatiotemporal aspect be added below.
> Conceptual will be above the other 4 aspects, and together, they all will
> be called "conceptual aspects." We can also think about adding
> metadata/querying for emotions by concept for various sound files, if not
> already present.
> >
> >
> > That said, people have different ideas of space. My wife wouldn't even
> describe space to me. Perhaps we should ask people to describe "space"
> before adding that aspect. I do know that people have very different ideas
> of time, but that X3D has very specific ideas. Perhaps we should allow
> people to query space-time in their own ways, and provide a smorgasbord
> like approach?
> >
> >
> > I'm kind of fudging here, because the general theory of relativity has
> not been combined with the special theory of relativity.
> >
> >
> > I would encourage us to develop more concepts that are
> animation/emotion/spatiotemporal based, like "bouncy."
> >
> >
> > I'm not sure what the term functional means in this context (structural
> is to formal, as ??? is to functional). I like behavioral or change. I
> think we should harken back to our mathematical roots for this one, and
> speak of relative position, velocity and acceleration: "Which car is the
> fastest going around checkpoint 5?" might be a query.
> >
> >
> >
> > On Mon, Dec 30, 2019 at 11:08 AM John Carlson <yottzumm at gmail.com
> <mailto:yottzumm at gmail.com>> wrote:
> >
> > Well, I've been doing a bit of mathematical anthropology, looking at
> maps of mathematics. I would encourage us to look at these, and think
> about providing high-level microtheories based on these maps.
> >
> > https://www.youtube.com/watch?v=OmJ-4B-mS-Y (there are static
> versions as well)
> >
> > Classifying geometry with topology may be interesting for search
> purposes. How else might we use topology?
> https://images.app.goo.gl/8S8rjXvoHcAXVDAU7
> >
> > There are also maps of science, philosophy and physics which are
> likely too broad to cover just in one paper.
> >
> > But coming down to mathematics, it appears that there are 4 main
> subject areas from pure mathematics in our ontologies found in the map of
> mathematics: structure, change, space, and symbol (numbers). I believe
> these should be top-level ontological objects, as follows Structure for
> Geometry, Symbol for visual symbols, Change for Scripts, Routes, Animation
> and Interpolation, Space for manipulations with transforms.
> >
> > Specifically, for structural output graphics, I would say there are
> 3 types of "topologies," namely, hypermovie, hypertext, hypershape and the
> parent hypermedia (we can have hypermovies of hypermovies, hyperdocuments
> and hypershapes, etc.) I studied these 3 basic media types under
> PostScript first, using X/NeWS and printers. We should also consider
> media like hypertactile, hypersmell, hypertaste, beyond hypersight and
> hyperhearing (hyperemotion anyone?).
> >
> > Next I would like to study structural entropy. How might
> structure&change like particle systems and classical physics be
> ontologized? How might we explore levels of entropy? There's perfect
> order, or pure mathematics, stochastic/chance/probabilistic order, chaotic
> order, and quantum order (someone can put in forms of disorder if they
> like). How might we ontologize mixtures of order and chance? Can we map
> "hurricane" or "earthquake" to graphics easily?
> >
> > Frankly, I am not ready to crack open "Maps of Meaning" yet.
> >
> > John
> >
> >
> >
> > On Mon, Dec 30, 2019 at 9:21 AM Brutzman, Donald (Don) (CIV) <
> brutzman at nps.edu <mailto:brutzman at nps.edu>> wrote:
> >
> > Am traveling today, unable to join. Next week should be OK.
> >
> > TODO items:
> > - Add Jakub’s queries in IC3D paper to X3D Ontology suite
> > - Add paper to archive
> > - Post tweet if any photos might be available
> >
> > Reviewing the recent documents on feasibility study of ontology
> for geometry and topology is really interesting and encouraging. We have a
> lot of great things to do!
> >
> > For our next meeting, let’s discuss what we want to do for Web3D
> 2020 in Seoul 24-26 JUN for greatest influence. Paper, tutorial?
> > https://www.web3d.org/event/web3d-2020-conference-seoul-korea
> >
> > Thanks for a tremendous 2019, looking forward to a productive
> and Happy New Year together. 🧨💥😎👍
> >
> > v/r Don
> >
> > Sent from my handheld device
> > --
> > semantics mailing list
> > semantics at web3d.org <mailto:semantics at web3d.org>
> > http://web3d.org/mailman/listinfo/semantics_web3d.org
>
>
> all the best, Don
> --
> Don Brutzman Naval Postgraduate School, Code USW/Br
> brutzman at nps.edu
> Watkins 270, MOVES Institute, Monterey CA 93943-5000 USA +1.831.656.2149
> X3D graphics, virtual worlds, navy robotics
> http://faculty.nps.edu/brutzman
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/semantics-public_web3d.org/attachments/20191230/038907e9/attachment-0001.html>
More information about the semantics-public
mailing list