Thanks, this is one of the most rational comments, still not convinced but open to seeing.
In the above comment, there is a screenshot of a web page being shown, it’s a real-time screenshot, but not interactive.
We used a similar method on our collaborative canvas in Golfcraft (btw open-source), which in fact could be the one with the most resolution in DCL, players were able to paint with a given small delay, I leave the timelapse here: https://www.youtube.com/watch?v=8AHtUGuUo5Y
Who’s browser runs this and will it be a security risk? I would still need to sign into my accounts to do anything. So it would have limited use. I don’t see a lot of benefit for this.
No, you won’t be able to send a signature, for now, you will see pages like using guest mode. and if at any moment you do, it wouldn’t be insecure to send a signature.
It does not have a lot of benefits, but it covers a necessity and is not a 240K grant, but everything sums.
Ok, spend some hours today for a first prototype with cache and scroll.
First time takes few seconds to load the page, just because the method used that waits until all network requests are resolved, this could be configurable. Also using rust headless chrome can improve performance.
Second time it loads fast because the site image is in cache.
Added a scroll up/down with E/F and UVS over an image that contains the full height… not the best method to see a clear image, but we can split in chunks and it will be more nitid.
next test could be to add clicks, improve method to have clearer text, add configurable options, etc.
Other option could be to always request external navigation when we click on a link out of the configured domain.
When one considers how this can be a shared experience, the use-case should be quite clear.
Many users will want to create living environments, in which they can invite others to spend relaxing time.
Simple lounges and living-rooms, decorated with taste, where one can catch up with the happenings of the lives of others.
And where one can, among others, share some website find or other on that fancy DCL web-browsing computer !
Even if it doesn’t end up being used much, the simple “coolness factor” coming from DCL being able to do this should be enough to tip the scales.
As long as it is understood that only anonymous browsing can be supported, and browsing of simple websites (future development is always possible I suppose), the requirements don’t appear to be too demanding.
The only uncertainty I have is about the apparent need of an external server for the rendering; could this not be done by library code within the scene logic itself ?
No, I don’t think so, a server is necessary, which will provide images for the attendees.
Unfortunately, we cannot render a canvas as texture or similar. Maybe in other web-based metaverses, but not here.
Could this maybe even be made a Smart-Item ?
Not sure if appropriate. afaik, smart items execute code for the owner, not for others. Several people could use a smart item, but I don’t think the benefit could be on a smart item for personal use, but for a scene to showcase information from a web site, or maybe a classroom to show slides to students.