[x3d-public] Hyperlink and form support

[Leonard Daly]

Hi Albert,

I think it is important to make a distinction between a single web page 
(that is in 3D) opening up many 2D pages and allowing the user to 
interact with each of them separately and a browser application (e.g., 
replacement for Chrome) that does more or less the same thing.

This example presents a potential security issue. In the first case, the 
page application (typically written in JavaScript) must maintain 
separation between the various external pages; otherwise one page could 
capture the contents of another site. The existing structure in the 
browsers works at preventing this sort of operation by preventing 
JavaScript from one page from accessing the contents of another page 
(unless permission is explicitly granted). This extends to even 
off-screen rendering and grabbing of a web page to use as a texture in a 
3D world.

The second case is very much like a tabbed display where the some number 
of tabs are all shown in a single display separated by 3D virtual space. 
In this case it is the responsibility of the application to maintain 
separation between the various content.

All of the discussions I have heard from multiple sources indicate that 
the browser and/or underlying libraries would actively prevent the first 
case (single 3D web page opening up multiple 2D pages). The security 
considerations are too great and expecting the browser to manage all of 
the interfaces is too difficult at this time.

If you are developing your own application for a 3D display that does 
this, then the responsibility is now on you to ensure that there is no 
cross-site information transfer. In this case, you would need to figure 
out what happens to a web page that wished to use some resource of the 
device (e.g., WebGL with canvas, or device camera) and how that 
interacts with other demands for  those same resources. For example, 
imagine three different web pages, one was displaying a virtual world 
with audio (all computer generated), one was using the device camera for 
remote viewing, and one was an AR (camera + computer generated) experience.

If your intent is to use multiple pages from a single site, then you do 
not have that issue; as it is assumed that the a single 
person/organization has control and access to all of that information. 
You could have a general page for 3D content that also loads and 
positions various 2D pages (like a list of participants or chat 
windows). The ability to handle the interaction on the various 2D pages 
would be controlled by the application providing the 3D environment 
context and the browser.

Leonard Daly



> Thank you all for your answers. I appreciate that.
>
> To avoid too much emphasis on the example given: the broader question 
> is about having the 3D web specification prepared to support 
> business-type processes for AR/VR devices.
>
> It seems to me the most natural view is to see 2D webpages as special 
> cases of the 3D web,  where all elements are flattened to one plane. 
> So in 3D space one should be able to position and use both the active 
> elements like textareas, checkboxes etc, and more or less passive 
> elements like div's anywhere within the scene (Leonard, your first 
> case scenario). This of course includes combinations of these elements 
> on flat plains up to full 2D pages viewable in 3D.
>
> Currently the resolutions for devices like HoloLens and I think also 
> Magic Leap (didn't try ML yet) are good enough for viewing 2D web 
> pages floating in 3D. This would allow companies to easily extend 
> their websites for 3D access by using X3D, and/or an equivalent 
> declarative syntax, in combination with their existing 2D content.
>
> Albert Jan Wonnink

-- 
*Leonard Daly*
3D Systems & Cloud Consultant
LA ACM SIGGRAPH Past Chair
President, Daly Realism - /Creating the Future/
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://web3d.org/pipermail/x3d-public_web3d.org/attachments/20181217/34be4405/attachment.html>