Mozilla, Virtual Reality, and the Dawn of the Metaverse? | Hacker News Reader

Mozilla, Virtual Reality, and the Dawn of the Metaverse?

wemo.io

46 pointscpeterso19 comments

We have the Metaverse. It's called Second Life. It works reasonably well. Second Life already supports the Oculus Rift. See "http://www.youtube.com/watch?v=gWE91IB9QMI". They still seem to be struggling with the control side of the user interface, but somebody will think of something. Maybe a Wii controller. (Basic problem of virtual reality: looking around, moving, and shooting work great. Everything else is hard.)

From Mozilla, we have "Now imagine a Minecraft that under the hood was HTML, CSS, and JavaScript". Ugh. There was a previous idea like this. It was called Web3D, and it's VRML 97 in XML syntax. "A spinning logo in 40 characters" was claimed. Anybody remember Web3D? Anybody? Hello? Also, putting on headgear to browse the web seems a bit much. You can build it, but will they come? The web is going in the other direction, towards small mobile devices.

(Mozilla's video won't play in Firefox 33 on Ubuntu Linux. Their Javascript video player crashed with a type error at line 1790 of "vjs.js". And those guys think using Javascript will make things better. The annoying thing about the Mozilla Foundation is that their core product keeps getting worse. Firefox crashes more now than it did three years ago. The number of unfixed bugs increases over time. Mozilla has a huge technical debt; the code base goes back to the Netscape days, and it shows.)

And Second Life is owned by a single, proprietary interest. And that's great, but that's not the Web. The beauty of the Web is that nobody owns it all and yet it still all works together. Even one of Linden Lab's former founders is trying to help open VR up on the Web: https://highfidelity.io/

I tend to think the problem with VRML was less to do with markup and more to do with 1997. Did anything 3D look great in 1997? Not really, and even then it still impressed us. But we have the power today to make--while not the most cutting edge graphics--certainly anything better than was available in, say, 2010, and we can make it cross-platform, open, and on the Web.

Doesn't that sound like something we should at least try to do? Just once?

> And Second Life is owned by a single, proprietary interest.

Linden Lab's/Phillip Rosedale's stated goal was world economy creation. Second Life is to be thought of as a web protocol that is undergoing standardization, and is stated to become fully open-source when that standard is finalized.

>And that's great, but that's not the Web.

Well, 'There', the predecessor to Second Life which created most of the control scheme concepts and user content creation schemas that were employed in Second Life was actually a technology demonstration between a collaboration of Forterra Systems and the US Army for private training simulation and remote work.

If a system which was co-created by a defense group eventually becoming standardized for public use isn't 'the Web', I don't know what is.

I filed https://bugzilla.mozilla.org/show_bug.cgi?id=1097572 to track the issue you're seeing with the JS player - I think it's vid.ly not air.mozilla.org that's not working, but it's worth looking into.

Can you send me more info on your crashes? I work on crash collection and reporting. If you put about:crashes in the address bar and hit enter, you can click on some of the crashes there and send the links to me.

Is this the Ubuntu Firefox build or did you download it from Mozilla? (they do their own builds)

Sure. There will be a SERVICE CHARGE of $125 for custom debug assistance services.

(Sorry, you can't expect end users to debug your product. That's your job. It's probably because I'm using Flashblock and Ghostery to block ads and tracking, and Mozilla tried something those won't allow. I don't have time to debug every defective web site.)

The site in question is not hosted by Mozilla, but embeds video hosted by vid.ly, which air.mozilla.com also uses.

You certainly don't have to go out of your way on my account! If you'd like help with crashes or anything else, let me know or file a bug.

Nobody has been able to reproduce the problem you hit with the video, although I appreciate knowing about it. It's pretty difficult to test arbitrary sites with different sets of addons and other local config, so I certainly appreciate if you don't have time/desire to dig in further.

I'll pony up the $125 right now. Give me a PayPal or Bitcoin address to send it to. I want to see your about:crashes

No one wants to go out and compare their project to Second Life though, since SL is (to put it mildly) a case study in what people will do given the ability to create anything they want.

That's the Metaverse. That's what it does.

This is a completely different approach to X3D/VRML. Everyone involved is acutely aware that a declarative approach won't work.

Every project's unfixed bug count increases over time, unless the project is dead. Likewise, technical debt is a problem that every large successful project has.

DK1&2 owner here, I really don't see how the metaverse is better than our current metaphor of browsers/webpages when it comes to sharing information? We already exist in the 3D world and we invented the web browser to make information sharing more efficient. Having to negotiate a 3D world instead of scrolling with my mouse wheel seems like the less efficient metaphor.

Web browsers were invented to load "pages", or text in discrete documents. And until recently, that's pretty much all it could do without plugins. Now that we have good WebGL implementations, near-native javascript performance, websockets, etc. we can start moving beyond static, flat, rectangular, discrete pieces of content.

>> we can start moving beyond static, flat, rectangular, discrete pieces of content

to what? I've tried myself to imagine what that could be but then I realized I was trying to find a problem for the technology than the other way around..

Only useful for browser games, tv and adult entertainment but that is pretty much it.

When I see statements like this, I'm always reminded of this gem: "No wireless. Less space than a nomad. Lame."

Who knows?

What does sticking HTML and JS into every hole have to do with the Metaverse? It's about Web, not legacy languages.

It's what we have. It's filtered out that HTML5 is an application platform. It succeeds where Java tried and fell almost completely on its face: write once, run (mostly) anywhere. It started with ubiquitous and came to functional, rather than the other way around with other cross-platform development systems. Take that for what it's worth. But that's all it is.

I'm excited for VR on the Web (i.e. using the Web as a deployment platform for VR). It necessitates finally pulling through on the promise of HTML5. Remember, back when the iPhone was first released, there was no app store for 3rd party apps. It was supposed to be all HTML5 apps. But at the time, HTML5 couldn't do enough to make it worthwhile. Today, there are very few things I can't do in the browser that I can do as a "native" app.

As for the Web on VR (using VR to render Web content), I think it's unlikely that we're going to be reading HN in an Oculus Rift anytime soon. Again, I go back to the example of the smartphone. Most of the activity has been around making new classes of apps that work best in the form factor. No shit, Sherlock. We're not going to see any serious effort to adapt our current understanding of the Web to VR, and I'm completely fine with that. We will, however, see new designs on the Web that can upgrade or degrade as needed between VR and 2D displays. People poo-poo'd the mobile web, disbelieving we'd ever fit a site designed for the desktop on such a tiny screen. And then we invented responsive design. The Web on VR will have it's own responsive design patterns.

This stuff about "markup languages for Minecraft" is... I think kind of lacking in vision, because you don't need VR to make that happen, nor is it even a thing that anyone actually wants. Markup is text, text is 2D, and 2D and 3D mesh poorly. I think it's why 3D desktop metaphors never worked: because they were all mapped to 2D user interaction devices. And there is a reason you don't program on your smartphone (in general, actually I manage it on occasion, tweaks to my site in an emergency), because there isn't a keyboard worth doing it. VR is going to make keyboarding too difficult for most people. A certain class of VR form factors (the smartphone-based ones) are going to make it impossible. And that's actually a good thing, because they open up new possibilities in free-roaming, augmented reality in the process.

Carpenter gets into one of my biggest problems with the development community with regards to VR right now, "95Hz is what the baseline is for presence." No it isn't. That's Oculus Rift marketing material to get everyone thinking of Oculus Rift as the Porsche of VR hardware. Currently, they are, but they're trying to establish a brand expectation. I am seeing a lot of people completely avoid VR because "it's not 95Hz yet" or "the latency, the latency, THE LATENCY!", like it's going to hurt them. Yeah, things are a little rough right now, but it's mostly on the developers now to make good looking graphics. The hardware is mostly there. We used to say "60Hz is the baseline for smooth video in gaming" and then went right on playing at 30-45Hz, as much as our machines would handle. If you wait for the hardware to get the rest of the way there (and it will, it doesn't have far to go), then you'll miss the boat.

Yes, faster framerates and lower latency are always better. Don't come at me quoting some study you were not involved with at all talking about experiences you never had. There is a difference between virtual reality--where we talk about nebulous concepts like "presence"--and stereoscopic view apps--where you're still looking at a display. If you want to talk about VR, there is a hell of a lot more to it than just frame rate and latency, and a lot of people are completely ignoring that there is still a lot of work to do in simple things like matching the right field of view for any given user, satisfying body awareness, motion sensing, working with 3D ambient-audio, or even just plain making an app that is interesting and fun to use, regardless of whether or not it's VR.

And even if you don't go for full VR, there is still a lot to do with stereoscopic view apps. I think most people who will end up developing apps for stereo-displays will end up making stereoscopic view apps rather than VR apps, just because it's so much easier, especially right now with the state of the art in interaction devices and development tools for VR being so thin. A traditional, mouse-and-keyboard- or gamepad-oriented FPS-style game with cartoony graphics and a stereo view is not VR but it's still a lot of fun. You don't need 95hz for that.

Reply on news.ycombinator.com