[webvr] Why does entering VR need to be on user interaction?

Michael Chang flux.blackcat at gmail.com
Thu Sep 22 20:15:49 UTC 2016


Oh I see, so

A. Already in VR to a new site that's supposed to be viewed in VR **stay in
vr**
B. Not in VR but clicked on a link that's supposed to be viewed in VR
**undecided behavior**

I'm trying to think of use cases for this...

For example, an Ikea catalog site where every link opens up a VR view into
a new product (although why wouldn't you just build the catalog in the
first place to avoid having to take off the headset).

We'll fire off a onvrdisplayactivate event early in the page lifetime
> (maybe look for a meta tag too? TBD) and if the page calls requestPresent
> in response to that event we'll keep the user in VR and let the new page
> start presenting.


Can't ads still easily jack this event?

On Thu, Sep 22, 2016 at 1:10 PM, Brandon Jones <bajones at google.com> wrote:

> So there's a difference in that we have a concept (not quite implemented
> yet) of "VR linking" where I'm in a VR page and I do some interaction that
> moves me to another page that also supports VR. We'll fire off a
> onvrdisplayactivate event early in the page lifetime (maybe look for a meta
> tag too? TBD) and if the page calls requestPresent in response to that
> event we'll keep the user in VR and let the new page start presenting.
> Obviously the browser has to make present this transition nicely to the
> user and provide appropriate context (indicate what domain you're moving
> to, etc.)
>
> So the question posed by Florian is: Do we allow the same behavior if the
> user isn't already in VR but the site they're headed would have gone into
> VR mode automatically in the above scenario? Contextually it's more
> disjoint for the user, but depending on how the linking page presented it
> that transition could make sense.
>
> On Thu, Sep 22, 2016 at 12:58 PM Michael Chang <flux.blackcat at gmail.com>
> wrote:
>
>>
>>    1. Visite some site (an ordinary web page)
>>    2. It has a link of the form <a href="vr-thingy.html">Try this VR
>>    thingy here</a>
>>    3. You click the link, you land on a different page, whose sole
>>    purpose is to do VR
>>    4. You have to click "Enter VR Now" on that page to get into VR
>>
>> I thought that was what Kip was talking about about 15 replies ago with
>> onvrdisplayactivate event?
>>
>> On Thu, Sep 22, 2016 at 8:47 AM, Brandon Jones <bajones at google.com>
>> wrote:
>>
>>> On Thu, Sep 22, 2016, 2:18 AM Florian Bösch <pyalot at gmail.com> wrote:
>>>
>>>> I believe there is a situation in which the user interaction to enter
>>>> WebVR requirement might not be intuitive.
>>>>
>>>>
>>>>    1. Visite some site (an ordinary web page)
>>>>    2. It has a link of the form <a href="vr-thingy.html">Try this VR
>>>>    thingy here</a>
>>>>    3. You click the link, you land on a different page, whose sole
>>>>    purpose is to do VR
>>>>    4. You have to click "Enter VR Now" on that page to get into VR
>>>>
>>>> It could be argued that user intent to enter VR was already decided on
>>>> step #3 with a single click.
>>>>
>>>
>>> This is definitely an interesting use case to consider. I've want to
>>> discuss it with the other implementors, but it doesn't seem too
>>> unreasonable.
>>>
>>> I guess it largely comes down to trying to infer the users context. If
>>> I'm already viewing WebVR content and I link to new WebVR content it's
>>> highly likely that I want to see the new page in VR immediately. The
>>> assumption coming from the other side is that if I'm not in VR yet it's
>>> (especially at this point) a pretty safe bet that I don't want to be
>>> accidentally tossed into VR. The "link as a user gesture" concept straddles
>>> those two environments.
>>>
>>>>
>>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mail.mozilla.org/pipermail/web-vr-discuss/attachments/20160922/63dc480e/attachment-0003.html>


More information about the web-vr-discuss mailing list