Mozilla Announces Deal to Bring Firefox Reality to HTC VIVE Devices(blog.mozilla.org)
blog.mozilla.org
Mozilla Announces Deal to Bring Firefox Reality to HTC VIVE Devices
https://blog.mozilla.org/blog/2019/01/08/mozilla-announces-deal-to-bring-firefox-reality-to-htc-vive-devices/
62 comments
Firefox from what I gathered will be used as the de facto browser in in-VR web browsing. If you read the VIVE promotion article on VIVE homepage you'd know that they are planning a kind of VR metaverse as an interface to the games you play - instead of picking an option from a flat surface you'll "walk" into the app/game. But of course this doesn't change the facts that web is flat so any in-app or in-game web will be displayed in some fashion using the Firefox Reality engine.
Second Life has browsers in-world, and they're Chrome subprocesses. These allow you to put any web page on a surface in the 3D world.
Doing a "VR metaverse" is a big job. There are about a half dozen of them now, and they all have tiny user counts.
Doing a "VR metaverse" is a big job. There are about a half dozen of them now, and they all have tiny user counts.
I don’t know. They maybe hope they’ll be first on the next big thing which they think is VR. But who knows if VR is going to be big, who knows if it needs a browser and what keeps chrome from eating the cake anyways after Firefox has scoped out the territory.
I think this is a very risky bet unless they make net money on this which they can funnel into more promising projects.
I think this is a very risky bet unless they make net money on this which they can funnel into more promising projects.
You and I have very different definitions of risky :)
Worst case, the effort of a relatively small team has been wasted, and Gecko and Servo will have received a few improvements as a result of their work. Best case, VR makes it big and Firefox is its go-to browser.
Then consider the case where they're not in the VR space early. Best case, they've saved a relatively small team's effort. Worst case, VR makes it big and Firefox is as marginally relevant as it is now on mobile.
Worst case, the effort of a relatively small team has been wasted, and Gecko and Servo will have received a few improvements as a result of their work. Best case, VR makes it big and Firefox is its go-to browser.
Then consider the case where they're not in the VR space early. Best case, they've saved a relatively small team's effort. Worst case, VR makes it big and Firefox is as marginally relevant as it is now on mobile.
I don’t believe that being early as a browser gives you that much of an advantage in such an unexplored space. Differences between browsers in this space will be big for a while and people will switch back and forth.
Maybe the cost of this project for Mozilla is so small that it doesn’t matter, but I think it isn’t
Edit: also I said that the bet itself is risky, not that it’s putting Mozilla’s existence in jeopardy
Maybe the cost of this project for Mozilla is so small that it doesn’t matter, but I think it isn’t
Edit: also I said that the bet itself is risky, not that it’s putting Mozilla’s existence in jeopardy
Is this going to be a text-free browser?
Trying to display/read small text in VR really shows up the limitations of current tech.
Trying to display/read small text in VR really shows up the limitations of current tech.
Mozilla needs to focus on their main product. I finally "upgraded" Firefox on an old 4 GB Windows 7 machine. It's bigger, slower, and crashes every time it exits.
Hey i am using Windows 7 and Firefox as daily 'driver' here.the lastest firefox is rocking here as i can tell.
I don't have a clue about VR stuff, and this is the first time I'm hearing about "Firefox Reality". Why do you need a special browser? What's different about this one?
A web browser can browse to site that display VR content. A VR browser has to handle various aspects of this:
1. Displaying the VR content itself. Most can do this already but it's not always switched on by default. 2. Render the browser chrome - the UI around the VR content in a way that's sensible and usable in VR. 3. Handle transitioning from VR web contentto a non-VR content in a sensible way. (where does the browser Chrome live when the "web page" is covering 360 degrees in both directions? VR content can't be contained in a rectangular frame)
1. Displaying the VR content itself. Most can do this already but it's not always switched on by default. 2. Render the browser chrome - the UI around the VR content in a way that's sensible and usable in VR. 3. Handle transitioning from VR web contentto a non-VR content in a sensible way. (where does the browser Chrome live when the "web page" is covering 360 degrees in both directions? VR content can't be contained in a rectangular frame)
A parallelizable layout engine should make it a lot easier to get the kind of framerate needed for vr. Mozilla have used this as a test-case for Servo. I’ve previously criticized them for it as an extreme niche market and a diversion from mainstream use cases, but happy to swallow my words if it goes well.
I was indeed wondering about how niche this was. Are there any reliable usage statistics for, say, how many people currently use a VR browser daily?
(And to forestall the inevitable comment, I understand the potential of VR, and have since the 1990s VR wave. What I'm wondering about is actual sustained (that is, non-novelty) use.)
(And to forestall the inevitable comment, I understand the potential of VR, and have since the 1990s VR wave. What I'm wondering about is actual sustained (that is, non-novelty) use.)
It's rather chicken and egg at the moment but if you believe in the web and you think VR is an important new medium it's important that the web doesn't get left out of the party.
Something has to be the connecting tissue of the multiverse and we'd better hope it's not Facebook or something equally proprietary.
Something has to be the connecting tissue of the multiverse and we'd better hope it's not Facebook or something equally proprietary.
Sorry if I wasn't clear, but the theory that "VR is an important new medium" is exactly what I wanted empirical data to test. I was not asking about potential; I was asking about current sustained use.
Yeah. I was trying in a roundabout way to say "Current usage is probably low but this is why I think it's still an important strategic move". It's definitely about having faith in VR as a medium rather than "oh wow. Look how successful VR is already!" - because it isn't.
Thanks, but when I said I was trying to forestall the inevitable comment about potential, it was exactly that sort of faith-based reply I was hoping to avoid.
I'm finding it hard to imagine what kind of evidence would be of use in a scenario like this. Some kind of Gartner hand-wavey focus-group thing?
You don't have to imagine. I gave a clear example: DAU statistics for a VR browser. Or you could look at basically any successful technology, look at it on the way up, and ask what the early signs were that it was seeing significant use beyond novelty purchase. (That is in Moore's model getting out of the "innovator/techie" market segment and into pragmatic use.)
For literally my entire adult life people have been talking about VR as the coming big thing. I'm 100% over hearing about that, especially in response to very specific questions about actual use.
For literally my entire adult life people have been talking about VR as the coming big thing. I'm 100% over hearing about that, especially in response to very specific questions about actual use.
Not quite sure what this means - is Firefox Reality going to be the default browser on the VIVE?
Yes. It will be the pre-installed default browser on Vive devices.
I still don't understand what this means. It's like saying Firefox will be the default browser of Dell monitors. The HTC Vive is a display.
Unless they're referring to HTC's software suite (Viveport I think it's called), which very few people use because it's some of the worst software I've ever seen. For example, it somehow managed to jam the entire 2.4GHz and 5GHz bands in my apartment while it was running. Not even sure how it's possible to do that by accident.
Unless they're referring to HTC's software suite (Viveport I think it's called), which very few people use because it's some of the worst software I've ever seen. For example, it somehow managed to jam the entire 2.4GHz and 5GHz bands in my apartment while it was running. Not even sure how it's possible to do that by accident.
On desktop, it means that if the user chooses to install the Viveport suite, that home experience and the suite will provide Firefox Reality as the default browser.
For all-in-one devices, Firefox Reality will be preinstalled and the default browser.
For all-in-one devices, Firefox Reality will be preinstalled and the default browser.
Can confirm the PC software is terrible and should just never be installed.
I think this is actually about their freestanding arm based headsets as this is the only target Firefox Reality builds for (android probably).
I think this is actually about their freestanding arm based headsets as this is the only target Firefox Reality builds for (android probably).
(this announcement is most likely about mobile vr headsets, not the PC side)
I'm pleased about this because -
1- Firefox Reality is actually "Servo" ie the Rust based browser Mozilla has been developing for years. edit - UGGGH this may not ship with Servo enabled :(
2- This is a stamp of approval from both Firefox and HTC that Servo is ready for consumers at least for this use case (which is very demanding low latency graphics!)
3- Actual Firefox means far higher standards of privacy and integrity then some cancerous Chromium embedded scenario.
I'm concerned enough to think twice about buying anything from Vive mobile because -
1- Their Vive PC software is awful shovelware.
2- It's not clear what the OS environment around the browser would be in an HTC branded mobile headset. If it's a clean version of Android that can be hacked on and replaced then we have something.
3- WebXR/WebVR still has no useful way to actually browse the real web, there is no "HTMLTexture" to allow you to say even look at Wikipedia through a portal. As a result you are trapped in basically WebXR Demo hell...
I'm pleased about this because -
1- Firefox Reality is actually "Servo" ie the Rust based browser Mozilla has been developing for years. edit - UGGGH this may not ship with Servo enabled :(
2- This is a stamp of approval from both Firefox and HTC that Servo is ready for consumers at least for this use case (which is very demanding low latency graphics!)
3- Actual Firefox means far higher standards of privacy and integrity then some cancerous Chromium embedded scenario.
I'm concerned enough to think twice about buying anything from Vive mobile because -
1- Their Vive PC software is awful shovelware.
2- It's not clear what the OS environment around the browser would be in an HTC branded mobile headset. If it's a clean version of Android that can be hacked on and replaced then we have something.
3- WebXR/WebVR still has no useful way to actually browse the real web, there is no "HTMLTexture" to allow you to say even look at Wikipedia through a portal. As a result you are trapped in basically WebXR Demo hell...
Firefox Reality is the brand we're using for the suite of browser products across VR and AR devices.
Today, Firefox Reality is available for standalone VR devices (like the HTC Vive Focus) and is a native Java application that talks to the GeckoView library, which is made up of components from the mobile version of Gecko, the Firefox web engine.
Firefox Reality for desktop will similarly be based off of the Firefox Desktop product pieces.
Note that there's a bunch of "Servo inside" - from small Rust components to the Quantum work including the new CSS style system to the forthcoming WebRender component.
I don't have anything to announce today about a Firefox Reality-branded browser based on top of Servo.
Today, Firefox Reality is available for standalone VR devices (like the HTC Vive Focus) and is a native Java application that talks to the GeckoView library, which is made up of components from the mobile version of Gecko, the Firefox web engine.
Firefox Reality for desktop will similarly be based off of the Firefox Desktop product pieces.
Note that there's a bunch of "Servo inside" - from small Rust components to the Quantum work including the new CSS style system to the forthcoming WebRender component.
I don't have anything to announce today about a Firefox Reality-branded browser based on top of Servo.
Probably worth mentioning, Firefox VR is also going targeting Magic Leap: https://blog.mozvr.com/a-new-browser-for-magic-leap/
IMHO the strategy of unlocking a lot of VR and AR content via a browser makes a lot of sense. The ongoing Rustification of Firefox, things like Web GL, WASM, etc. It all makes sense for VR/AR applications to build on that. Solving the empty room problem where there is no meaningful content or where all the interesting content is in somebody else's walled garden is a key challenge.
VR requires a dedicated UX and I can imagine independent hardware vendors are maybe not so eager to e.g. rely on Google for providing them with that UX. Also, Firefox seems to be leading here rather than following.
IMHO the strategy of unlocking a lot of VR and AR content via a browser makes a lot of sense. The ongoing Rustification of Firefox, things like Web GL, WASM, etc. It all makes sense for VR/AR applications to build on that. Solving the empty room problem where there is no meaningful content or where all the interesting content is in somebody else's walled garden is a key challenge.
VR requires a dedicated UX and I can imagine independent hardware vendors are maybe not so eager to e.g. rely on Google for providing them with that UX. Also, Firefox seems to be leading here rather than following.
Shameless plug, but for WebXR on Magic Leap github.com/webmixedreality/exokit has been the go-to for a while.
I wrote it because I found browsers too slow to move on features like html to texture (powered by webrender) and multi-site blending, which become quite manageable if the browser core is written in JavaScript.
I wrote it because I found browsers too slow to move on features like html to texture (powered by webrender) and multi-site blending, which become quite manageable if the browser core is written in JavaScript.
> 1- Firefox Reality is actually "Servo" ie the Rust based browser Mozilla has been developing for years.
Firefox Reality currently uses GeckoView, Gecko repackaged in a WebView-like embeddable Android component. The Firefox Reality team is still working on experimental support for Servo:
https://github.com/MozillaReality/FirefoxReality/#experiment...
Firefox Reality currently uses GeckoView, Gecko repackaged in a WebView-like embeddable Android component. The Firefox Reality team is still working on experimental support for Servo:
https://github.com/MozillaReality/FirefoxReality/#experiment...
hgq(1)
Why would they deal with HTC and not just Steam in general.
I have had Vive since day 1. I am never touching anything HTC again. I'm happy enough with the device and technology I have- but the HTC software is horrendous and there's no place for it. I've heard just horror story after horror story of HTC customer service / repair, wanting $200 for a new controller etc.
Vive Pro is a complete joke, in price.
HTC hasn't done anything positive since the release of the original vive.
I have had Vive since day 1. I am never touching anything HTC again. I'm happy enough with the device and technology I have- but the HTC software is horrendous and there's no place for it. I've heard just horror story after horror story of HTC customer service / repair, wanting $200 for a new controller etc.
Vive Pro is a complete joke, in price.
HTC hasn't done anything positive since the release of the original vive.
You might notice HTC Vive seems to be distancing themselves from Steam, including focusing pretty much entirely on their own Viveport platform. Though I'm not sure why, that seems to be the mentality behind their current direction.
To non-gamers, Steam looks kinda wacky - there's crazy popups, really intense graphics, nonstandard UIs, and it has to update most times it launches. It's also blocked on lots of networks - businesses, schools, etc. I personally love Steam and its craziness, but that's the main reason - HTC wants Vive to become mainstream and to capitalize on other uses for VR besides gaming.
HTC Vive is focusing on Viveport for the same reason Epic, Ubisoft, and Oculus have: They want to take the sales away from Steam. We are watching the same thing happening with Netflix verses every television company (CBS, HBO, etc) making their own streaming service fiefdom. Netflix and Steam beat the other companies for distribution by years and they don't like that. Re-fragmentation.
A commonly voiced complaint about current VR gear is that it's too expensive for mainstream adoption. HTC doesn't have a Facebook backing it, so it can't just subsidize the hardware to build market share - unless it finds another revenue stream. The obvious candidate (see game console economics) is the complementary good [1]: VR software. Hence HTC Studios and Viveport.
[1] https://en.wikipedia.org/wiki/Complementary_good
[1] https://en.wikipedia.org/wiki/Complementary_good
But, why?
I'm going to offer a different explanation, but first, the standard one: Mozilla doesn't want to miss a paradigm shift, especially one where the only players currently in it are very proprietary. They learned their lesson with mobile.
One thing that many of us, myself included, are very excited about regarding the Servo/WebRender/Quantum projects is the ability for a fast, parallel web rendering engine to open new possibilities, such as how V8's JIT significantly pushed the envelope of what was possible. Now that JavaScript is acceptably fast, and we have WebAssembly for cases where control over memory layout is necessary, network speeds are the bottleneck of most applications. WebRender's big concept of rendering the entire web at "maximum FPS" is pretty awesome, but it's not pushing the envelope by itself, since framework and application developers have been treating "minimize DOM manipulation" as an important optimization. There's clearly potential with WebRender, but if a majority of used webpages are already rendering at an adequate framerate due to framework-level optimizations (and prior browser optimizations like Electrolysis and Stylo), it may be viewed as superfluous, despite potentially huge advantages.
I think the research in this direction is beneficial, and as a result, I'd like the Servo project to be allowed to continue developing cool stuff. As Servo won't replace Gecko in Firefox, it needs to try more challenging tasks to continue to push its envelope. WebRender has a ton of potential in the WebVR space, since a Web Components-based VR system would rely heavily on a performant DOM. Such a project would push Servo to continue development in preparation for a more functional future web.
One thing that many of us, myself included, are very excited about regarding the Servo/WebRender/Quantum projects is the ability for a fast, parallel web rendering engine to open new possibilities, such as how V8's JIT significantly pushed the envelope of what was possible. Now that JavaScript is acceptably fast, and we have WebAssembly for cases where control over memory layout is necessary, network speeds are the bottleneck of most applications. WebRender's big concept of rendering the entire web at "maximum FPS" is pretty awesome, but it's not pushing the envelope by itself, since framework and application developers have been treating "minimize DOM manipulation" as an important optimization. There's clearly potential with WebRender, but if a majority of used webpages are already rendering at an adequate framerate due to framework-level optimizations (and prior browser optimizations like Electrolysis and Stylo), it may be viewed as superfluous, despite potentially huge advantages.
I think the research in this direction is beneficial, and as a result, I'd like the Servo project to be allowed to continue developing cool stuff. As Servo won't replace Gecko in Firefox, it needs to try more challenging tasks to continue to push its envelope. WebRender has a ton of potential in the WebVR space, since a Web Components-based VR system would rely heavily on a performant DOM. Such a project would push Servo to continue development in preparation for a more functional future web.
WR doesn't help much with DOM update related issues, WR comes into play after that, when it's pixel time. WR is pushing the envelope for creative (ab)use of DOM elements as graphics primitives, like in that spinning cubes demo :) but on a more serious note, leveraging the GPU makes a lot of sense because it's been sitting there mostly unused (only used for layer compositing) when browsing, and it's more efficient at graphics operations.
That said, Servo itself has a fresh DOM implementation and I think there was work on making it really fast.
That said, Servo itself has a fresh DOM implementation and I think there was work on making it really fast.
In retrospect, you're correct. I should have said "WR reduces the cost of the paint/composite cycle".
Because both AR and VR still have the possibility of enormous growth.
Mozilla as an organization these days seems to be hunting for that one trick that will let them magically become big again.
I don't want a trick. I want a good web browser. From its beginning, Firefox got popular because it was just a good web browser, not because of any gimmick. If anything, it was the lack of gimmicks that made it good. Unlike previous Mozilla products, it was just the browser.
I'd be curious to see an interview with their executive team. Their solutions just don't seem to align with people's problems any more.
(Last year, I might have guessed they're just abandoning the desktop web browser market, but today they're the only remaining completely independent implementation, so they've lucked themselves into a position of great importance. A lot of people who never use Firefox still want it to succeed.)
I don't want a trick. I want a good web browser. From its beginning, Firefox got popular because it was just a good web browser, not because of any gimmick. If anything, it was the lack of gimmicks that made it good. Unlike previous Mozilla products, it was just the browser.
I'd be curious to see an interview with their executive team. Their solutions just don't seem to align with people's problems any more.
(Last year, I might have guessed they're just abandoning the desktop web browser market, but today they're the only remaining completely independent implementation, so they've lucked themselves into a position of great importance. A lot of people who never use Firefox still want it to succeed.)
To the best of my knowledge, Mozilla relies on Google for the vast majority of their operating budget. That is a precarious position to be in. So of course they're searching for other sources of revenue.
I'm glad you want Firefox to succeed. I do too. (I'm posting this from FF!) But unless you can find a few dozen million friends to pony up actual cash for it, every year, Mozilla's going to have to keep looking for other options.
I'm glad you want Firefox to succeed. I do too. (I'm posting this from FF!) But unless you can find a few dozen million friends to pony up actual cash for it, every year, Mozilla's going to have to keep looking for other options.
Why do you consider a VR browser to be a gimmick? And why is this announcement not relevant to building a "good web browser"? They want to build a good web browser for VR, as well as mobile, desktop, and any other platforms which are relevant today and hopefully tomorrow. Perhaps you are only interested in the desktop version, but it's still probably in their best interests to develop browsers for other upcoming markets. Were you also upset that they spent valuable resources building an Android version?
Smart move on Mozilla's part to start working on VR. The desktop browser space has been saturated for so long, but making a bet on a new space might help them get a toehold on a whole new world.
This doesn't mean they are not focused on desktop, but strategically this is a solid bed that may or may not pay off, but will be key for long-term survival.
EDIT: plus having commercial sponsors for the eventual Servo-in-VR (as noted elsewhere, not shipping yet) will be key to getting Servo itself ready for production.
EDIT 2: Can we get material properties on HTML elements please? I want <img>s with bump-mapping and specular highlights. :)
This doesn't mean they are not focused on desktop, but strategically this is a solid bed that may or may not pay off, but will be key for long-term survival.
EDIT: plus having commercial sponsors for the eventual Servo-in-VR (as noted elsewhere, not shipping yet) will be key to getting Servo itself ready for production.
EDIT 2: Can we get material properties on HTML elements please? I want <img>s with bump-mapping and specular highlights. :)
In particular, Firefox Reality is just a consumer of GeckoView.
Working on Firefox Reality has beneficial effects on Gecko's general ability to be embedded.
Working on Firefox Reality has beneficial effects on Gecko's general ability to be embedded.
Embedding in other Android apps is not really what we expect in general by being embeddable. Servo for instance is easy to embed since you just need to provide it a GL context and feed it events. This is far from being the case with Gecko unfortunately. Chromium is pretty good also especially with the Ozone layer.
From what I've heard, it's not an Android-exclusive effort, and eventually Gecko will become embeddable again on desktop too. Camino Resurrection anyone? :)
While you're correct in your assessment, and I hope this comes to fruition, this same paragraph could have been written about Firefox OS. That said, I'm not a detractor, I wish Firefox OS had also worked out. And I'm certainly very glad that Mozilla is still willing to approach new ideas and markets, even after getting burnt.
The Firefox Reality team is much smaller than the FirefoxOS team was (i.e., by more than an order of magnitude).
The FirefoxOS legacy is alive in KaiOS which has 50M+ users now and is growing. Indeed a very poor decision from Mozilla's leadership to kill the effort too soon.
I have KaiOS on a Nokia phone that I intended to use for email and 4g hotspot. I didn't expect much but it literally a POS, super slow, impossible to customize (i.e. hotspot quick launch), and no apps in the marketplace. It would be more usable with a pre symbian OS in my opinion. People that rave about KaiOS I figure are like that those that endless recommend SICP or 'Clean Code' as the book of gods and probably never read it.
Cool, going to try this on my Daydream setup right now.
Chrome on Daydream presents a super simplified UI. I'd love to be able to open a few tabs, move them around spatially, adjust the size, etc.
There is so much potential for improving browsing experiences in VR.
Chrome on Daydream presents a super simplified UI. I'd love to be able to open a few tabs, move them around spatially, adjust the size, etc.
There is so much potential for improving browsing experiences in VR.
That's what I've been wanting since VR started. I just want to be able to place browser windows around me in space, and enter input into whichever one I'm looking at. Bonus points when I can also have a VLC window somewhere, file browser, etc, but mostly just want a stable usable VR-space window manager for now.
What device are you using? I'm using a Pixel 2 XL and had to sideload it because the Play Store shows it as incompatible. It is broken, though, can't type more than one letter. There's a ton of promise though. I haven't found a good VR browser yet.
Somewhat considering actually working on the project, I know they're on Github but don't know if they are looking for help.
Somewhat considering actually working on the project, I know they're on Github but don't know if they are looking for help.