Re: WWWAnchor as a WWWAnchor's child. Clarification needed.

Michael Pichler (mpichler@iicm.tu-graz.ac.at)
Thu, 23 Nov 1995 11:46:41 +0100


> From: Mario.Juric@public.srce.hr (Mario Juric - XV Gimnazija)
>> Is the browser's behaviour defined in cases where WWWAnchor
>> is a child of WWWAnchor ? Take this for example :

#VRML V1.0 ascii
WWWAnchor {
name "http://blah.blah/~blah"
DEF ClickMe Cube {}
# you may wish to add this here to make it better to see:
Translation { translation 2 0 0 }
WWWAnchor {
name "http://bu.bu/~bu"
DEF ClickMeToo Cone {}
}
}

>> What would happen if I clicked on the cone ? Will it take me to
>> http://blah.blah/~blah or to http://bu.bu/~bu ? I don't think
>> this is clear in the VRML 1.0 spec, and some modelers (Fountain)
>> generate files with cases where WWWAnchor has a WWWAnchor child.
>> Maybe this should be clarified for 1.1 spec ?

>>>>> "J" == James Waldrop <sulam@construct.net> writes:
J> Good question. This came up at one point and got locally clarified,
J> but I don't know if it got into 1.1 (either Bernie's list or the one
J> the VAG produced).

J> The general upshot, if I remember correctly, is that the first object
J> in the hierarchy wins. Of course, I might be wrong about this.

I think the result was ginnermost wins", but I may be wrong too.

J> I suggsst you try it out on WebSpace, WebFX, Scout, and WorldView.
J> That'll give you a rough consensus on what the supposed behavior is.

I tried it under UNIX with webspace (chooses innermost), webview
(chooses outermost), and the upcoming VRweb 1.1 (chooses innermost;
includes WWWAnchor support, available in a few days).

Although a clear spec should be preferable to "try all browsers and
expect that what the majority does to be right", I can imagine how
busy all people are at the moment.

J> James

Michael

| DI Michael Pichler ______ mpichler@iicm.tu-graz.ac.at |
| IICM, Graz Univ. of Technology / / Tel: +43-316-873-5627 |
| Schieszstattgasse 4a / /'95 Fax: +43-316-824394 |
| A-8010 Graz, Austria, EU-rope `esality inside' |
|--------------------------------------------------------------------------|
| Browse through space with VRweb: http://hyperg.iicm.tu-graz.ac.at/vrweb |


  • Next message: Gabriel Sigrist: "TGS WebSpace crashes on Solaris"
  • Previous message: Michael Pichler: "Re: WWWAnchor as a WWWAnchor's child. Clarification needed."
  • In reply to: James Waldrop: "Re: WWWAnchor as a WWWAnchor's child. Clarification needed."
  • Next in thesad: Bernie Roehl: "Re: WWWAnchor as a WWWAnchor's child. Clarification needed."