Firefox OS Pivot to Connected Devices | Hacker News Reader

Firefox OS Pivot to Connected Devices

blog.mozilla.org

62 pointsbpierre78 comments

So, at first, they've told us that they don't have enough resources to pour into Thunderbird because they were working on something else. That something else was Firefox OS. Okay, sure. I would love to have a Mozilla developed OS on my phone. Then, months have passed and, even though they have an agreement with the country right next to mine, I was not able to have one without having someone from that country who will sign a contract with the carrier. Then they have decided to give Thunderbird away to some other project because they wanted to focus on Firefox. Sure, why not? But, a day or two later, they have released an app for iOS that completely does not belong in their model of service since it has nothing to do with Firefox and it won't actually work with Firefox. And now they're telling us that they're moving Firefox OS to "IoT devices", killing Firefox OS smartphone idea before I had a chance to actually get my hands on one of them. Great work Mozilla!

+1 for the point about Thunderbird. We need good open-source mail clients. I'm not sure we need a Firefox OS, and regardless they just don't have the market power to make it work.

There's Mailpile: https://www.mailpile.is/

That looks neat but doesn't fufill the "low resource client on my desktop with notifications" that Thunderbird does.

Maybe re-package the Firefox OS Mail Client? ;) It's "low resource" ;) -> https://github.com/mozilla-b2g/gaia/tree/master/apps/email

Low resource? Thunderbird?

You might want to check out claws-mail. You get notification via a plugin (on debian/ubuntu, the plugin is in another package, but it's very easy to get).

>We need good open-source mail clients

Take a look at https://nylas.com/N1

It's not the traditional mail client, it requires server (https://github.com/nylas/N1/blob/master/CONTRIBUTING.md#runn...)

You can host a server yourself or use a version hosted by the company. If you are not concerned with privacy issues, it works like a traditional mail client.

He said he wanted a mail client, not a mail client that requires a self hosted server infrastructure or giving the email out.

> it works like a traditional mail client.

Except it's not just a client.

Wait, last time I checked, the API/Server wasn't free software? Did I miss something, or has the licensing model changed?

It seems that sync engine was AGPL licensed from the start https://github.com/nylas/sync-engine/blob/master/LICENSE

I didn't realize the sync engine was available and possible to use for a 100% self-host solution -- but evidently it is :-)

FossaMail is an Open Source, Mozilla Thunderbird-based mail, news and chat client for Windows and Linux, brought to you by the Pale Moon developers (and therefore not affiliated with the Mozilla Corporation). It is an alternative version of the Mozilla Thunderbird mail&news client, and based on the Pale Moon browser core.

http://www.fossamail.org/

Thunderbird is / was probably the last comprehensive desktop mail client that enabled ordinary users to decouple themselves from e-mail providers.

Unfortunately Mozilla [ com | org ] have shackled themselves to Googlemail internally, so seem to have abandoned any aspirations to fight for openness in the domain of e-mail.

They're also removing tab groups (panorama) in the next version. Apparently due to low usage -- which is surprising considering how much they've advertised the feature. /s

> So, at first, they've told us that they don't have enough resources to pour into Thunderbird because they were working on something else. That something else was Firefox OS

I'm not really familiar with Mozilla's internal org - so I have just had a look at Mozilla Corp's leadership page[1]: there is an SVP for Firefox, and an SVP for "Connected Devices" - who authored the linked article. There are no other SVPs: which suggests MozCorp consider connected devices to be as important as Firefox . It looks like Thunderbird is orphaned and has no champion within Mozilla (I don't think the Firefox division wants it).

So it looks like it boils down to org structure & politics (for the lack of a better word): Thunderbird is dying because it's an unwanted step-child, and FirefoxOS is revived because it's the biggest thing going on for an entire division/department ("Connected Devices")

1. https://www.mozilla.org/en-US/about/leadership/

It looks like Mozilla is chasing tech buzzwords, but are always a few steps too far behind to actually do any impact. Why not focus on what they are doing best (i.e. browser, email)? There's still plenty to do, Firefox keeps crashing both on my desktop PC and Android phone, now Thunderbird is abandoned. IoT market is full of big players with vast amount of resources and saturated not less (if not more) than smartphone market, I don't see how they gonna compete there. As for OSS, Linux is doing great and keeps doing better in IoT scene[0], why not merge efforts? I appreciate Mozilla's efforts, but lately it lacks focus and clear goals. Maybe some fresh blood would help?

We will explore and prototype new use cases in the world of connected devices as an open source project with a clear focus on the user benefit and experience.

..yada yada, sounds like a buzzword filled marketing crap with no substance whatsoever[1].

0. https://www.linux.com/news/embedded-mobile/mobile-linux/8180...

1. https://media.licdn.com/mpr/mpr/p/5/005/04f/263/04c1eb8.jpg

They focus a ton on the browser. I use chrome a lot less than Firefox and I still see chrome tab crashes more often. Spidermonkey performs amazingly today, and even beats V8 on its own benchmarks. FF used to be resource hungry, but now is a lot lighter on memory usage than chrome. Beyond current tech, they're investing heavily into Rust and Servo for the next generation, and their ES6/7 support has been steadily growing. By all measures, it's clear that they have goals on the browser and they take those goals seriously. What else can they do?

Third party desktop mail clients are just not a killer feature for the vast majority of users. IoT may not be a winner for various reasons, but I'm sure it can't be easy trying to figure out where to go from here. (let's face it, IoT is going to be the same locked-down, vendor-specific experience that Internet tv has become; and without manufacturing a must-have device or buying out a manufacturer of one, why would Mozilla win in that field?)

Rust and Servo are easily the most important things they're working on IMHO. And WebAssembly, which combined with Servo I hope will have a big impact on web UI development and performance.

> Spidermonkey performs amazingly today, and even beats V8 on its own benchmarks.

And yet somehow Firefox manages to run WebGL a lot worse than Chrome and Opera (Blink/V8 based browsers)...

Because what they're doing good - desktop browser and desktop mail client - are becoming more and more irrelevant.

Browsing is quickly becoming a thing most people do on (only) their phone and email is for more and more something you do on the web.

If Mozilla keeps focusing on these two aspects they will start to cater to a niche market and lose the ability to influence the Internet at large.

> [...] email is for more and more something you do on the web.

Yeah, I would say that I agree with that, but about a month ago, I have joined a tech team of an NGO that works with a lot of sensitive data.

Convincing them to use Enigmail with Thunderbird is hard. Convincing them to use Enigmail with Thunderbird even though you are not sure what will be the future of Thunderbird is just plain frustrating.

And no, there's absolutely no alternative that could handle all three desktop OS's (Windows, Linux, OS X), attachments, and other things that are completely necessary to have a safe, encrypted communication.

There's absolutely no alternative what so ever. I really wish that some cool project (like LibreOffice) takes over Thunderbird and makes it great, but the odds for that are practically non-existing.

I'm not sure where you got the idea that Thunderbird was sunset to keep resources for Firefox OS, but that's not the case. There's only one former TB dev that works on Firefox OS, and guess what, he's in charge of the email app.

If you want to try Firefox OS, even with no carrier support, take a look at https://firefoxos.mozilla.community/devices/ which lists supported devices.

From Mitchell Baker's blog post, it appears that it is Firefox that has diverging needs from Thunderbird.

"Thunderbird and Firefox are interconnected in a few different ways. They are connected through our technical infrastructure. Both use Mozilla build and release systems. They’re connected through the sharing of some pieces of technology. The time has come when it is no longer effective for Thunderbird and Firefox to keep sharing the same technical infrastructure. Firefox and Thunderbird have diverging needs."

https://blog.lizardwrangler.com/2015/12/03/thunderbird-updat...

The irony of this pivot is that Andreas Gal, one of the creators of Firefox OS,[1] has already tried to do IoT with Firefox OS[2] and failed. His startup is now doing IoT with NodeJS.[3]

Li Gong, former executive in charge of Firefox OS,[4] also has a startup focusing on Firefox OS smartphones.[5] Having raised $100 million and poached dozens of Mozilla employees, he is looking for even more funding.[6]

[1] http://www.cnet.com/news/mozillas-new-cto-javascript-firefox...

[2] http://www.cnet.com/news/firefox-os-in-flux-as-mozilla-loses...

[3] http://www.cnet.com/news/startup-aims-to-make-home-devices-s...

[4] https://www.linkedin.com/in/ligongcn

[5] http://www.cnet.com/news/startup-lands-100-million-to-challe...

[6] http://www.cnet.com/news/startup-acadine-picks-up-the-torch-...

Whoa at the last two points. I wonder if Acadine Technologies means that the reports on the death of the Firefox OS smartphone are premature?

http://www.fiercemobileit.com/story/acadine-undeterred-effor...

So the OS was slow and difficult to use because of the silly idea of making an entire OS's user interface run in a browser engine (and to be honest, not even the fastest browser engine out there), and then they put it on low end phones and it predictably failed. So they decide to "pivot" and put it on even lower power IoT devices. Come ON, Mozilla.

It sounds like Oracle's netbook which was original chrome book in 96.Every body can now absorb pretty clearly the bandwidth and stability is not there.(even today.let alone 96).

I think companies should understand if you don't have money for just spending on products like these.then don't start them.because it will make you look like idiot. Maybe 25 year from now you can run Gecko(what was their engines name?) On low powered chip.but today?no way.

FxOS 1.0 on ZTE Open was/is plenty fast despite the very low end hardware.

It boggles the mind. At the same time, their browser market share is eaten by Chrome now and probably MS Edge starting next year or so. But it typical enterprise lingo, they "pivot" a dead project into something that continues eating resources for nothing.

Mozilla, get your act together, you are burning through a lot of good-will lately.

Here comes the hope that they might focus on Rust for this. (And drop the idea of browser stuff)

> (And drop the idea of browser stuff)

"Dropping the idea of browser stuff" was never a goal of Rust.

I'd disagree about slow and difficult. On early devices, the Peak for example, it was slow but the more recent Flame with the latest FxOS is fine. Easy to use as well - it's been fine for our child's first phone - but too restrictive for a power user. The browser is ironically a prime example in that things are missing such as extension support, about: pages.

"they decide to put it on even lower power IoT devices" is your own interpretation, but absolutely not what we plan to do.

Might have that on my headstone.

"Here lies Martin. Not dead, just pivoting to connected devices."

Why would anyone pick this instead of Qt/QML, Java/JavaFX, Android/Java, Win10/XAML which are already being used in production?

EDIT: Forgot to mention Tizen as well.

Something not related.why Oracle discontinued javafx scene builder?.it send signal to programmer javafx is not serious.

Yeah, I was also pissed off with it.

However there are lots of IoT vendors that offer first class support for Java on their SDKs.

There were lots of talks at JavaONE 2015.

Also there are car makers like VW exploring JavaFX for their infotainment systems.

http://www.bredex.de/blog_article_en/research-into-the-use-o...

Gluon is anyway taking care of it.

EDIT: Changed a bit the content.

Why is Mozilla wasting time on this useless stuff?

What they need to do is work on their browsers, and specifically:

1. Ship Electrolysis ASAP, so Firefox is once again as technically good as Chrome and others

2. Add support for syncing with Chrome and IE if possible, to smooth transition for mainstream users

3. Fund Servo as much as possible, and get it ready to ship it in both desktop and mobile browsers as soon as possible

4. Get the marketing people to work on a strategy to market Servo to get the #1 browser market share spot again

5. Stop eroding their reputation with dumb stuff like the Pocket and Telefonica integrations. If they do something to make money, clearly state so and make sure it's harmless to users and trivially disabled. Then, market their reputation as a trustworthy non-profit much better.

If they REALLY want to do a mobile OS, just work together with Cyanogen on an Android distribution with added/better HTML app support, instead of inventing their own crappy thing.

Working with Cyanogen isn't a bad idea. They really should just create a united "open mobile software" group along with Canonical to pool together their efforts.

Sigh.. Just let it die and move on.

Something like http://janos.io/ ?

I wonder what would be an example of a "connected device" in this context for Firefox OS, as it feels really vague for the moment.

"Smart" TVs is one of the few things I can think of that seem to fit somewhat: Have a display, apps make sense, ... A lot of IoT stuff is little boxes without flexible UI, that does seem to miss the mark with FirefoxOS.

They were going to attempt that with http://www.matchstick.tv/ but that project was cancelled.

Which was terribly disappointing. I was really hoping for an open-protocol equivalent to the ChromeCast.

Just to clarify, “They” here is not Mozilla. Matchstick was a separate company.

By the way, it is bizarre that they do not mention their death on the website (http://www.matchstick.tv/overview/company-news/). On their Kickstarter article, their wording is a bit unclear, but it does seem like they have stopped working on Matchstick (https://www.kickstarter.com/projects/matchstick/matchstick-t...).

Yep, they refunded my money.

They provide the "smart tv" system for Panasonic.

"How to use OpenData, Firefox OS and a Raspberry Pi to build a better Washing Machine" http://www.slideshare.net/andrefiedler1/how-to-use-opendata-...

As much as people here aren't a fan of the choice, I saw Firefox OS running on TVs and other IoT devices at this year's MozFest and it was really impressive. The developer APIs were easy to grasp and the UI was intuitive.

No one (serious) thinks that Mozilla cannot deliver impressing feats of engineering, at the very least because they've got quite a bit of tail left on their goodwill for breaking the IE monopoly.

What people are implicating is their strategy. As others have mentioned, they don't seem to lead entry into any space, and therefore their presence in the space is constantly in the vein of catching up. Thus far, the only clear reason I've ever heard them mention is the same "open web standards" argument. Obviously this approach is valid, but with their browser success they should know better than anyone that you cannot overtake Goliath with open standards if no one is using your product.

Lo and behold, they've apparently shown a product at MozCon that Google and Apple have already bet a ton on and released compelling consumer offerings for. Regardless of how impressive the APIs and UI may be, history repeating itself here.

The Mozilla organization is disassembling itself.

I'm sure that if they rename a few more things that have "bro" in their title, everything will change for the better.

Aaaaany day now.

Sounds like Mozilla wants to jump on the Internet of things bandwagon.

Noooo. I don't want any Firefox on my light bulbs. So when will this die? 1 year or in 2 years? And who is gonna patch my Firefox light bulbs then?! ;)

What advantages would Firefox OS have over other Linux-based options for IoT?

A consistent platform is the most obvious benefit.

Could it also be a fairly static non vendor customized and hence upgradable OS also? Most Smart TVs are never updated after all

I don't see Smart TVs as a good market for Firefox OS, Android is the obvious choice in this market. Generally, you'd want to target IoT devices where battery life/power draw is important but you still want a rich UI.

Consistent platform could be a decent benefit for the markets where Firefox OS makes sense though.

Are there any Smart TVs shipping with Android? I don't think it is the obvious choice, there are tons of different systems used right now.

Current gen SmartTVs use Tizen (Samsung) and WebOS (LG), Chromecast (ChromeOS), Panasonic (FirefoxOS), Sony (Android), etc.

Fair enough.

> Most Smart TVs are never updated after all

Along with the usual problems of "smart" (spyware) devices, the non-updatability of IoT devices is a time bomb.

We already have a problem with firmware based on very-old Linux kernels; shipping devices with known kernel exploits should be automatic liability for any damage the device does. It's willful negligence.

As Dan Geer says[1], embedded devices need to either be updated or have a limited lifespan. We cannot allow a situation where common consumer IoT devices are both immortal and unfixable.

[1] http://geer.tinho.net/geer.blackhat.6viii14.txt

Can't any platform claim that as a benefit? For a consistent platform to have any value you'd like it to stick around for some time... Unlike FirefoxOS.

Webkit is eating the world

Firefox isn't based on Webkit (apart from the iOS version).

Which I think was the original poster's point.

Dropping Firefox OS on smartphones is nothing to do with Gecko vs. Webkit.

Yes it is, Mozilla is fighting to keep it relevant.

The decision to move away from a smartphone OS and towards IoT devices is not based on what serves Gecko. Gecko is better suited to smartphones.

Besides, Gecko is still relevant, and will continue to be for quite some time. Mozilla and Google appear to have a good working relationship when it comes to web standards, and new standards are only signed off when at least two browser vendors (or rather, browser engine vendors) implement the proposed functionality.

Gecko isn't relevant on mobile. Nothing apart from WebKit and Blink are: look no further than the fact that everyone else supports webkit prefixed CSS properties on mobile.

Firefox on Android uses Gecko. All iOS browsers are based on Webkit due to Apple's policies.

> "everyone else supports webkit prefixed CSS properties on mobile"

The question is, how often are these directly used? Isn't the common pattern to target modern features without universal support through polyfills and solutions like Babel nowadays?

> Firefox on Android uses Gecko.

Which has < 1% marketshare, and is largely ignored by web developers. I didn't say it didn't exist, I said it wasn't relevant. :)

> The question is, how often are these directly used? Isn't the common pattern to target modern features without universal support through polyfills and solutions like Babel nowadays?

Plenty often enough that users refused to use browsers without support for webkit prefixed properties. People who have a clue may well use polyfills and Babel and autoprefixer, but plenty just do copy/paste coding.

The last Palm OS was based on WebKit and it had similar issues to Firefox OS (very slow, resource hungry).

It got renamed to WebOS, and nowadays it runs in LG SmartTVs.

There is also ChromeOS and its similar to WebOS and FirefoxOS and runs on low spec notebooks just fine.

There is little performance difference between Android and FirefoxOS, for both you need at least 512MB RAM (memory), so that it works without lags.

Worked so well for Beos.

s/Firefox OS/WebOS/g

Reply on news.ycombinator.com