Re: EMOTE

[Niclas Olofsson <niclas.olofsson@ismobile.com>]

"Bullard, Claude L (Len)" wrote:
> Has anyone besides Rex and myself reviewed the
> EMOTE paper (Chi, Costa, Zhao, Badler)? 

No. But I took a look at the purpose of it and I feel somewhat confident
with my current opinions on the following...

> Should HumanML only capture a description of the conditions and rely
> on implementations to communicate these to the EMOTE middleware which then 
> communicates to H-anim the requirements for key frame adjustments, etc
> to get the right "Shape and Effort"?

As expressed elsewhere (www-vrml) I rate EMOTE and similar technologies
as middleware in respect to HumanML. In fact I hardly see it as a
middleware either, but rather a very thin (but extensive) layer, mostly
to enhance animations. But this is probably not true (please forgive me,
I will go read the paper when I'm finished writing).

HumanML middleware are implementations, yes. They are the engines that
put life into the spec/profiles/genres. But HumanML can of course do
without runtime. Doing pshyco profiles does not nesseccarily mean render
or run. Simulating pshycos however do. A typical place for a
HAnim/HumanML middleware is in a VE MU, like deepmatrix or blaxxun-mu.
Putting a HumanML driver inbetween two HAnim would put you in a position
to say that you probably are doing something innovative. This position
is open for filling right now ASFAIK.

So my answer to your question is yes and no. HumanML does not
nesseccarily need implemenetations to be useful. Sean asked if HumanML
was Esperanto ala digital. I say NOT, but I also say it can probably
serve as a Esperanto digital framework.

I'll go read the paper now. But sometimes it works better for me not
thinking to much about things. Having wrong can be quite constructive,
at least as long as you can live with it yourself. I'll be back to
correct myself later :-)

Cheers,
/Niclas