Changes To Look Forward To With Firefox 52

[liam]

That "someone" must've slept under a rock for years then 'cause that person missed the whole webOS era (and ChromeOS, if you will, but webOS was a mobile platform rather than laptop platform so more in line with FFOS). So it had been tried before and they could've easily looked at it rather than looking at FFOS for that. (not hating on FFOS though, the whole UI was pretty neat IMHO, esp. with the never-released 3.0)

I confess that i did forget about webos. Webos was released four years prior to fxos. The web changed a massive amount in that period and, i think, webos had two things working against it: 1) technical debt that they couldn't shake, 2) hardware wasn't quite good enough.
If there's a difference between the two projects if might be Mozilla's desire to push their stuff "upstream" (w3c/whatwg/tc39). So, aside from simply determining the api gap, you also need to actually implement them and, if successful, standardize them. I don't think fxos even attempted to push all their apis upstream, but you can see, from the old web api page, that the intent was there.
ChromeOS has been too wedded to pnacl, and now with the Android compatibility, it would appear that the team is looking at integrating with Android rather than relying on making "native" web apps (they is, ones that rely on the standard functionality provided through the browser, of browser-like interface). I'm saying despite the recent announcement regarding increased web app functionality in Android.

[ⲣⲂaggins]

Has Firefox fixed this age-old bug?

On Linux, we don't have a fast compositing path right now for WebGL. This means we do a readback of each frame before sending it to the compositor. This is really slow, and the reason for most reports on Linux regarding framerate differences between Chromium and Firefox.

No. So Firefox's WebGL is basically half as fast on linux as it is on any other platform, and has been like that for the last 4 years. Most notably this affects Google maps/earth/street view. I think I'll be sticking with chromium...

[AppTeF]

- No support for HTML5 input types datetime, datetime-local, date, time, month, week.

In next release part of date/time will be OK but some blocker remain feel free to work on it

https://bugzilla.mozilla.org/show_bug.cgi?id=1286182

https://wiki.mozilla.org/TPE_DOM/Date_time_input_types#Roadmap

[Azrael5]

Meh, Chrome blacklists accelerated hardware decode on Linux largely for political reasons. It's not even flag-enablable because Google thinks Linux GPU driver situation is awful. The feature is enabled in ChromeOS

Graphics Feature Status

• Canvas: Hardware accelerated
• Flash: Hardware accelerated
• Flash Stage3D: Hardware accelerated
• Flash Stage3D Baseline profile: Hardware accelerated
• Compositing: Hardware accelerated
• Multiple Raster Threads: Enabled
• Native GpuMemoryBuffers: Software only. Hardware acceleration disabled
• Rasterization: Software only. Hardware acceleration disabled
• Video Decode: Software only, hardware acceleration unavailable
• Video Encode: Hardware accelerated
• VPx Video Decode: Software only, hardware acceleration unavailable
• WebGL: Hardware accelerated
• WebGL2: Hardware accelerated
  
  I'm not so sure.

[Vistaus]

Firefox os wasn't really a move away from their core, imho. If it did nothing else, it forced someone to look really hard at the idea of browser-as-a-platform and see what it was missing.

That "someone" must've slept under a rock for years then 'cause that person missed the whole webOS era (and ChromeOS, if you will, but webOS was a mobile platform rather than laptop platform so more in line with FFOS). So it had been tried before and they could've easily looked at it rather than looking at FFOS for that. (not hating on FFOS though, the whole UI was pretty neat IMHO, esp. with the never-released 3.0)

[Vistaus]

I like Elementary OS and I run it. They picked Midori first and then Epiphany later because they could more easily re-theme them to work with the Elementary desktop themes, title bar behavior, etc...

I think the team's focus on beautiful aesthetics is admirable. But everything else about Elementary's desktop environment is fast and stable as well as pretty. Midori and Epiphany aren't. The first thing I do on Elementary is install Firefox and Chrome.

Web (Epiphany) is getting a hell of a lot of better though now with 3.23.x It has seen tremendous amount of progress. There's more customization (though admittedly, some of those things can only be customized through dconf-editor right now, but at least it's there!), there's more privacy focus (AdBlock and web tracker blocking by default), it's way faster, more stable and there's even experimental support for libhttpseverywhere and Firefox Sync. They also have plans to support WebExtensions in the future. So don't write of Web just yet; it's becoming way, way, way better now with 3.23 and will get better after that as well.

[nanonyme]

firefox has to improve hardly providing reliable hardware acceleration as well as to manage effectively both system ram VRam, cpu cache and buffers because it is very slow compared to chrome. It is not able to detect all the system abilities by which to take benefits. On the same hardware chrome is able to feature very well hardware acceleration where firefox fails.

Meh, Chrome blacklists accelerated hardware decode on Linux largely for political reasons. It's not even flag-enablable because Google thinks Linux GPU driver situation is awful. The feature is enabled in ChromeOS

[Delgarde]

Mozilla isn't the only one here to blame, Google dropped the ball earlier too. But the message is the same: if you are a hardware manufacturer, you cannot thrust then to base your product features better than Microsoft, and to me this is a very bad message from the opensource guys.

This is not something that's happening without warning - Mozilla, and all the other browser projects, have been talking about dropping support for binary plugins for a long time now... I remember this being discussed more than ten years ago. So while there hasn't necessarily been a good alternative (e.g. WebRTC) for some things until recently, TP-Link and others can't say they didn't see this coming.

[Delgarde]

crap!!, many corporate apps still require npapi java :/ I wonder how this will end up. Glad that I'm on MacOS with Safari and I'm safe at least until next version of OS

Corporate apps just have to deal with it, and invest the money to ditch that requirement. Because it's not just Firefox dropping plugin support - Chrome doesn't support Java, Edge doesn't support Java... Oracle is deprecating the plugin in Java 9, and presumably dropping it in 10.

Basically, anyone still using Java applets has until IE11 reaches end-of-life. That's going to be the last browser that still supports them, and it's already an obsolete platform that's being displaced by Edge.

[liam]

CNET dedicated a whole article to the layoffs where they blared "Firefox Fail". If it were a fail, it's because they didn't do it sooner. It's the right move for them. They need to concentrate their limited resources on core technologies - making them best in class. They finally realized that. A strong/competitive Firefox is good for everyone. I just switched back to using FF from Chrome. It's going to be a bit bumpy while they get webextensions working - but I'm pleased with the overall experience. As far as the RAM, etc. there have been countless articles that all agree that FF uses less RAM - but for most people it doesn't really make that big of difference. All the major browsers these days are perfectly capable of providing a good user experience. What matters though is making sure there are choices and commitment to user privacy and security. Mozilla keeps those issue in the forefront.

Firefox os wasn't really a move away from their core, imho. If it did nothing else, it forced someone to look really hard at the idea of browser-as-a-platform and see what it was missing. Part of the issue was the lack of standard low level apis. That is continually being addressed. The other part is performance. With this you butt up against an intrinsic issue with the web (well, html/js) that has to be either worked around or fixed in order to get native-like responsiveness. Servo (the layout engine) will go some ways as a workaround, and had it been working, fxos would've been much nicer, and, imho, it will provide enough of a realisable performance difference that the responsiveness issue will be all but gone.
When that out of the way, i agree that having Mozilla focus on improving their product is a good thing (though, keep in mind that they still have an eye on IoT). Clearly ff hasn't been the best browser for a very long time.... almost a decade? Has it been 10 years since the iPhone, and eight since chrome? Ugh, well, regardless, Mozilla scraping some of the stuff that was diverting their engineers from making changes to the core product has already been advantageous for the desktop.
Lastly, we need to keep in mind were ff sure relative to their competitors. The web is no longer a place that make software companies can simply ignore. This means that the giants, like ms, apple and google can't afford to short their web offerings. All what being equal, that implies they can simply outspend Mozilla. Unfortunately, this is pretty much the current situation. Even with a stripped down Mozilla org, they don't have the resources to keep up, let alone pull ahead, with these other companies. So, with that in mind, you can see why they've made attempts to acquire alternate revenue streams, and with that failing, trying desperately to find a niche whereby Mozilla can become a sustainable endeavor.