[x3d-public] HAnim only skeleton renderable ? task list

[Don Brutzman]

[copy HAnim Working Group]

Actually I do not think these MFNode fields are a problem at all, they are something that helps advanced functionality without hurting anything.

An easy and efficient way to make these fields nonrenderable in an implementation is to simply wrap them in a hidden switch, i.e. <Switch level="-1"></Switch>

One could make the case that they are superfluous, since the information already exists in the Humanoid.  However it is hard to reach there, requiring lots of scripting effort to "walk the tree" and find them.  Having such a list makes is much easier to create ROUTE connections at run time.  Conceivably an author might only want to only expose a smaller set of the joints and segments and sites, so having that list adds capability.

Am planning to give a good scrub on the following, hopefully several before SIGGRAPH:

a. All scenes in example archives checked against HAnim 2.0 DIS
b. Update the HAnim prototypes and perhaps add such checks
c. X3D Schematron rules to report differences and suggest additions to those MFNode fields - these help a lot.
d. X3D-Tidy rules to add the USE nodes to a Humanoid if missing
e. X3D Examples archive regression testing
f. X3DJSAIL testing
g. X3DJSONLD testing, if John wants
h. X3DOM and X_ITE testing, other X3D players as well
i. Rig the Shay D Pixel skin to a skeleton, showing a human-compliant LOA
j. Test Motion node generation integration and animation examples
k. Re-evaluate these issues

So plenty more TODO before specification changes can be justified.  Also no blockers to continuing progress now, only good reinforcements to come.

I've learned over the years that there is a lot of sophistication built into the HAnim design that is not always self-evident, so we will do well to continue trusting the collective wisdom of that powerful design.  No doubt we will appreciate that elegant design even further as we implement and evaluate that recent release for X3D v4.

Have fun with HAnim!  8)


On 6/22/2018 1:23 PM, Joseph D Williams wrote:
>   * I will see if and how it is possible to special case the nodes in these fields in x3dom, perhaps by removing them from the list of drawable objects.
> 
> That is the biggest problem, that these are a special case of USE that does not exist anywhere else. Any processor is supposed to not render this allegedly useful list from the user code, or substitute a notation that does not whack use of USE.
> 
> Of course, big blasting tool-tips or all caps in the spec are not truly reasonable solutions to the issue.
> 
> Thanks and Best,
> 
> Joe
> 
> *From: *Andreas Plesch <mailto:andreasplesch at gmail.com>
> *Sent: *Wednesday, June 20, 2018 5:15 AM
> *To: *Joe D Williams <mailto:joedwil at earthlink.net>
> *Cc: *X3D Graphics public mailing list <mailto:x3d-public at web3d.org>
> *Subject: *Re: [x3d-public] HAnim only skeleton renderable ?
> 
> Hi Joe,
> 
> ok. Most of the HAnim scenes in the web3d examples archive probably populate these fields with USE references. My vague sense is that they do that only out of a perception that this is required or recommended, not for a functional purpose, but I did not go through those in detail.
> 
> I will see if and how it is possible to special case the nodes in these fields in x3dom, perhaps by removing them from the list of drawable objects.
> 
> On a spec. tangent, 19774-2/V2.0 Annex B, here
> 
> http://www.web3d.org/documents/specifications/19774-2/V2.0/MotionDataAnimation/ExampleKeyframeAnimation.html
> 
> and 19774-1/V2.0 Annex F
> 
> http://www.web3d.org/documents/specifications/19774-1/V2.0/HAnim/Design.html
> 
> define a containerField='skeleton' for the example Humanoid whose parent is the Scene. This is probably an editing oversight.
> 
> -Andreas
> 
> On Tue, Jun 19, 2018 at 6:52 PM Joseph D Williams <joedwil at earthlink.net <mailto:joedwil at earthlink.net>> wrote:
> 
>       * So my question is if only the nodes in the skeleton field should be
> 
>     rendered while the USE references in the joints and segments fields
> 
>     should not be rendered and are provided exclusively for other purposes
> 
>     (like kinematics) ? I could not quite answer that question from the
> 
>     standard language.
> 
>     Forget these fields they are leftovers from when the skeleton was not included in the user code and the fields just defined which of the skeleton parts in the ‘standard’ model that was not even accessible to the actual user, these fields were all he got to author, and some animation.
> 
>     Anyway, unless somebody can clearly show the need for these, then should be dumped, and never treated as required. When it became feasible to permit the skeleton into the actual user code, these fields became excess but for some reason nobody will delete them.
> 
>     Thanks again,
> 
>     Joe
> 
>     *From: *Andreas Plesch <mailto:andreasplesch at gmail.com>
>     *Sent: *Tuesday, June 19, 2018 5:35 AM
>     *To: *X3D Graphics public mailing list <mailto:x3d-public at web3d.org>
>     *Subject: *[x3d-public] HAnim only skeleton renderable ?
> 
>     In my first attempt to improve the implemention of  HAnim nodes and
> 
>     fields in x3dom, most (all?) shapes making up the Nancy humanoid from
> 
>     Nancy, native tags, were rendered multiple times, on top of each
> 
>     other.
> 
>     The reason turned out to be the USE references in the joints and
> 
>     segments field of Humanoid.
> 
>     Since a USE node places another instance of a DEF node into the scene
> 
>     graph, all the Shapes in the joint and segment nodes were duplicated.
> 
>     So my question is if only the nodes in the skeleton field should be
> 
>     rendered while the USE references in the joints and segments fields
> 
>     should not be rendered and are provided exclusively for other purposes
> 
>     (like kinematics) ? I could not quite answer that question from the
> 
>     standard language.
> 
>     -Andreas

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