> you better keep it because I'm gonna use it again. =0D
> Oh yeah this is a chair if you want to forget about details and use you=
r own =0D
> favorite chair... If you want details this is specifically =0D
> a desk chair number 12345=0D
=0D
Sounds good to me. Here's another thought that might add weight to the ar=
gument for =0D
=FF"standard" objects.=0D
=0D
Lets suppose that VRML spaces become as popular as Web Pages. There are v=
ery few =0D
=FFpeople with the necessary skills or patience to construct 3D environme=
nts of any =0D
=FFcomplexity from scratch. It therefore seems likely that there will be =
a HUGE market =0D
=FFfor 3rd party object libraries allowing "artistically disadvantaged" p=
eople to more =0D
=FFquickly prepare their own VRML spaces. =0D
=0D
Now I admit that this is purely a 3rd party issue and I am sure that ther=
e will be =0D
=FFcommercial and free object libraries appearing on the scene fairly rap=
idly and that =0D
=FFthese will form the basis of many early VRML spaces. However...=0D
=0D
If such libraries become commonplace (as I'm sure they will) then it woul=
d make =0D
=FFexcellent sense to allow the VRML language to specify common objects. =
This should be =0D
=FFpossible in terms of both the generic chair example above and specific=
chairs from =0D
=FF3rd party libraries. I would suggest the addition of (yet) another fie=
ld to the =0D
=FFWWWInline node to allow specific naming of objects. ie:=0D
=0D
WWWInline {=0D
name "http://somewher.com/chair1.wrl"=0D
hint cache=0D
stob "chair"=0D
style "desk 12345"=0D
=FFuse "VRMLSoft:Chair99"=0D
}=0D
=0D
Having said all of that, there will be world authors who very painstaking=
ly build =0D
=FFthere worlds in just such a way. The fact that Jo Punter then sets the=
ir browser to =0D
=FFuse standard objects could leave their carefully sculpted environment =
looking (and =0D
=FFfunctioning once events and animation are added) rather differently th=
an they had =0D
=FFintended. =0D
=0D
It would seem logical that if we are to allow standard objects to be used=
in place =0D
=FFof the authors original object then we must also allow/require the aut=
hor to specify =0D
=FFa bounding box into which the standard object should be scaled prior t=
o display.=0D
=0D
Imagine for example that an author creates a tall, narrow backed chair an=
d that it =0D
=FFis essential to the function of the world that the back of the chair e=
xtends to a =0D
=FFcertain "height" within the world. The standard chair object is almost=
certainly =0D
=FFdifferently dimensioned and therefore its use may impact the world's f=
unctionality. =0D
=FFIf the author is able to specify a bounding box for any of his/her cri=
tically sized =0D
=FFobjects which may be replaced by standard objects this problem would b=
e at least =0D
=FFminimized.=0D
=0D
In summary my final version of the opening VRML statement would be:=0D
=0D
WWWInline {=0D
name "http://somewher.com/chair1.wrl"=0D
hint cache=0D
stob "chair"=0D
=FFbbox "xxx yyy zzz"=0D
style "desk 12345"=0D
=FFuse "VRMLSoft:Chair99"=0D
}=0D
=0D
although the bbox parameters could be optional suffixes to the stob field=
=2E=0D
=0D
Any thoughts ?=0D
=0D
=0D
=0D
+--------------------------------+-----------------------------+=0D
=A6 Dave Durbin =A6 [email protected] =A6=0D=
=A6 Visual Systems Solutions Ltd. =A6 100102,[email protected] =A6=0D=
=A6 Who ? =A6 [email protected] =A6=0D=
+--------------------------------+-----------------------------+=0D