RE: [cgmo-webcgm] Proposed agenda for 10/10 telecon

["Bezaire, Benoit" <bbezaire@ptc.com>]

Hi Ulrich,

Thanks for the update! I'm looking forward to reading the proposal.

This group will have to talk about timing, especially if you guys are proposing animation along a path.

Can I rename 'animation object' to 'animation definition'? Does that make sense or am I interpreting your description in the wrong way?

I would like for the animation definition to be inside the CGM; but I think I'm alone on this one, most people in the group were talking of have the animation entirely done in scripts outside the CGM. But if I understand where you are going; you have a control path that needs to be defined somewhere, correct?

This is the SVG section I think people should review:
http://www.w3.org/TR/SVG11/animate.html

As for the XCF and animation; I don't have a clear picture of all of this fits together yet.

Regards,
Benoit. 

-----Original Message-----
From: Ulrich Laesche [mailto:ulrich@ematek.de] 
Sent: Wednesday, October 10, 2007 7:38 AM
To: 'CGM Open WebCGM TC'
Subject: AW: [cgmo-webcgm] Proposed agenda for 10/10 telecon

Dear All,

I finally got two people together who will finish the transform proposal based on the ideas and your replies to the questions mentioned below.
As usual, my vision of simplicity for this proposal faded after digging into the details, but I still think it's not too complex.

The initial idea, I presented to development, was to simply provide a however complex transformation matrix to achieve things like movement, stretching, rotation etc.  This matrix would then get an id and could be applied to a grobject.  However, this idea was immediately rejected by the dev team as being too crude.  I was told, to achieve movement we would at least need something like a movement path.  They want to adapt the PostScript path definition for that and are working on describing it.

Of course, this brought up the issue of timing for the movement along the path but I told timing is no issue for now and that we should settle on discrete steps.  For example, a defined path and a number of n discrete steps would result in a beginning and end position and n-1 intermediate drawing positions on equal distances.  In addition, the transformation matrix still needs to be defined as you would like to rotate or resize the object along its path.  The dev team is working on combining these two methods regarding their mathematical and programming interface presentation.

From our point of view this approach covers the use cases we discussed, moving an object, opening/closing a switch, simulate fluid flow in a straight pipe.  For the latter two the path would be of zero length and the transformation matrix would be applied to a none moving object.
Programming examples are definetely needed here as a proof of concept.
Regarding timing, I guess this is a question of implementation.  It can't be done a as sequence of complete redraw operations but must be dealt with on the base of bitmap copy operations.  But does this need to be defined in the standard?

Beside the general question if this approach makes sense, I have three questions to avoid working into a wrong direction.  First of all, where does the animation object (let's call it like that for now) fit into the CGM file?  The animation object needs an ID, that's clear.  However, should it be a separate object so that it can be accessed by an arbitrary number of grobjects?  Or should it be tied to a specific grobject?  With the definition of a path (with fixed coordinates, I
guess) it is questionable if more than one object can make use of an animation object anyway.  But I don't know.

Secondly, Lofton, you mentioned some time ago that we should "at least"
have a look at SVG.  Did you refer to the standard in general or was there a specific section you had in mind?  I believe it was the latter.
If this is true, could you refer to a section in the standard that's worthwhile reading?  I want to avoid getting our team distracted and to look into directions which are probably too ambitious for this implementation (it happened already before when copying the group's mail exchange).

Thirdly, Stuart, what do you expect to be in the XCF file if you want to achieve animation from outside the CGM?  Do you see anything in this approach that wouldn't work in the perspective of offloading it to the XCF?  I believe it's all a question of standardizing the syntax.

Of course, comments are welcome from everyone.

Regards
Ulrich


-----Ursprüngliche Nachricht-----
Von: Cruikshank, David W [mailto:david.w.cruikshank@boeing.com]
Gesendet: Sonntag, 7. Oktober 2007 10:42
An: CGM Open WebCGM TC
Betreff: [cgmo-webcgm] Proposed agenda for 10/10 telecon

Please reply with regrets if you can't attend.
passcode for the telecon is: 6961530

Dial in numbers:
Toll numbers:
FRANCE +33-1-70-70-84-56
GERMANY +49-69-2222-3468
UNITED KINGDOM +44-20-7108-6391
USA +1-210-795-0625
Toll Free Numbers:
FRANCE 080-510-0984
GERMANY 0800-101-7056
UNITED KINGDOM 0800-279-9632
NORWAY 800-10129
USA 866-617-3597

Thx...Dave
 <<20071010_WebCGM_TC_Telecon_agenda.doc>> 

Technical Fellow - Graphics/Digital Data Interchange Boeing Commercial Airplane 206.544.3560, fax 206.662.3734 david.w.cruikshank@boeing.com