Anyone interested to run apk apps on FFOS ?
Maybe is possible to add dalvik vm in b2g and run android apps.
Of course on “top devices” (with 512 MB - 1 GB ram, dual - quad core, etc.)
Waiting for more apps on marketplace run some existent android app 
Do you like the idea and want to help me ?
One key advantage of Firefox OS is the speed, which you get precisely because it doesn’t have to support the Dalvik and Java layers. That means it can run equally fast on lower-powered hardware. Given that, why would you want to do this? You’d end up with a bad Android phone.
Also, I strongly suspect they’ll have taken out a bunch of stuff in the lower layers that Dalvik and the Android APIs would need.
@gerv ok, the speed of native html, css, js apps. The web phone philosophy is not in discussion.
But … do you have all apps that you want in https://marketplace.firefox.com/ ?
The possibility to run many android apps (apk) is not useful ?
I strongly suspect they’ll have taken out a bunch of stuff in the lower layers that Dalvik and the Android APIs would need
So there!
Could be a good idea check the difference between gonk layer (> https://developer.mozilla.org/en-US/Firefox_OS/Platform/Architecture Gonk is the lower level operating system of the ffos platform, consisting of a linux kernel based on aosp and userspace hal)
and hal, open source libs used by dalvik http://edureka.in/blog/wp-content/uploads/2013/01/Android-Stack.jpg
core libs and app framework in android are java stuff
kernel and drivers should be in common with android aosp (jelly bean, kit kat)
Maybe could be necessary a “software layer” dalvik <–> gonk
Disregarding the utility and or convenience of Android apps on a FxOS, the problem is exactly the “parallel part” of your idea.
Think in the lower level. You have 2 engines fighting for the processor, that’s not how software works, to solve that, then, we would need another layer to manage these 2 to work well, so we are already increasing a layer, which would make the phone slower even if you are not running any android based apps.
Also, you need to think about the integration with the JS APIs and it’s utility as a technology in general. The idea about making technology better and more accessible is not about making you able to run Instagram and twitter, it’s about making a better environment and affordable phones, that’s why FxOS is focusing in emergent markets, where android phones are usually expensive besides the shitty ones.
Mozilla thought the web had all the ability to make this, but it was still short on certain APIs, so they are making new ones for the sake of being universal, so for example, you have a battery API that works not only on phones, but also on laptops, how amazing is that?
Then, you must consider: Is it interesting to make an API to communicate with Android Apps? Will it help evolve the web? Will it open new possibilities? For the user, yes, sure, they will be able to install Instagram, though as @gerv said, you will end up with just a bad android phone.
Now, will it evolve the web? I doubt it, Android programmers will just continue making android applications and giving no support for other plataforms, and when it stops working, it will not be the creator, but the platform who will take the blame. Do you think Mozilla should be blamed if Instagram stops working? Not only that, would developers want to do that? Blackberry tried to force a bunch of converted android apps on the blackberry store and they got sued because they had no permission, BB did all the “hard” job, but most developers just didn’t want it anyway.
Where would a normal user get an APK if he can’t enter the Play Store because he doesn’t have an Android phone, would developers release their apks on a secondary story? Amazon got their own store, and they only met success thanks to the kindle fire (which does not have/can’t enter the Play Store), so developers reached them, so unless you make some cheat or illegal stuff, there would be no way to get these APKs legally.
Of course, a solution to that is a dual boot phone, you boot either Dalvik or Gonk, but not both at the same time.
Well … mozilla vision, philosophy (open web, evolution, etc.) is important.
But user needs are very important 
A wide choice with many applications installable on the phone is normal condition for a smartphone user today.
Like on ios and android.
And maybe in the next months these apps will not appear in ffos marketplace.
“Focus on emerging markets where android phones are usually expensive” was right yesterday.
Now there are android one phones http://www.android.com/one/
Cheap and good, with last release of kitkat and frequent updates.
On the technical side: yes, 2 app full stack that run in parallel (app framework + core libs + dalvik) and (gaia + gecko + gonk) may not be an ideal condition.
Also in a top device quad core with 1 GB RAM
As you said could be required low level software layer (for gonk, gecko, dalvik optimization) and integration layer between gecko - gaia (js) and android apps (java)
A possible solution is a dual boot android - ffos.
Or maybe an lxc container with fast switch ffos - android (like https://www.youtube.com/watch?v=UpIFByNLM5U)
I’m not sure if you actually know how it works in third world countries. What I mean is that a pretty bad phone costs half the minimum wage, unless of course you buy it on contract, but since most people end up preferring pre-paid, then there is no contract and you pay a lot.
I’m from Brazil, a third world and emerging country where FxOS is already available in the market (I already saw one person using an FxOS phone in my limited traveling time from work to home and vice-versa). Here, the minimum wage is the equivalent to $350, and the cheapest up to date (I’m not considering phones that should be out of the market already) usable phones costs around $150. Of course, there are smaller countries that have it worse.
But even these cheaper phones can get somewhat quite expensive since you end up multiplying the price by 3, and that includes FxOS phones. We are the land of the $2K PS4, and while the iPhone 6 still hasn’t been officially released here, it’s expected for the price to go over $1500 in the simplest versions, with the most complete and bigger ones going over $2K, and that is still considering a lower minimum wage than first and second world countries. So even Android One, which could be something like a fever in the US, will get heavily taxed outside it, so don’t expect it to disregard all economic problems just like that.
And then, about user needs, sure, they can be important, but for these people who are still on a Nokia feature phone, getting a cheap smartphone to be able to navigate the net is probably more of a priority than running Instagram, because there are a lot of people who are not up to these things that richer people buy.
I give myself as an example, I went from Nokia feature phone to the Geeksphone Peak, I wanted to try it and even with all taxes making it quite expensive, I was still going to use it as a main phone, but I admit I felt a little bored by the lack of advanced stuff, I could understand it, but since I’m a tech person, while seeing the phone growing and getting better was pretty cool, so was the frustration at those moments where I just needed it to work well. I now have a Sony Xperia Z1 and don’t expect to see myself going back to FxOS for a good while (Until it gets a lot more stable or more easily customizable [Though that generates security problems…]), but I still support it and it’s technology, it’s just a matter that it’s not made for me, which means I’m not a target for this product right now. I know it’s hard to understand this concept, but after you accept it, you get to appreciate even things that aren’t made for a consumer like yourself.
@ssj4maiko there are many possible targets for ffos phone (in the world)
Not only the minimum wage people that now use feature phones.
Are called customer segments.
From ultra low cost to low cost to middle cost (I don’t think ffos will be in the future on high segments like android with Samsung Galaxy S5, HTC One, etc…)
And now you in Brazil got a top device 4G - LTE http://www.gsmarena.com/sony_xperia_z1-5596.php
I suppose with kitkat 4.4.4. But you are not the target (because as a dev when possible you buy a last technology powerful device - quad core, 8 core, LTE, with 2-3 GB RAM, 16-32 GB flash mem, 20 Mpx camera …)
Having said that (not so important … seems that http://www.android.com/one/india/ is selling many devices in India.
Could be a good success in the next weeks - months
If we consider “possible targets”, you can think of every target as possible, such as a High-end FxOS phone, is it possible? Yeah. Is it commercially viable? I don’t think so.
And about my phone, my point was that me, as a Tech Person, which I believe you are too, will always prefer a better phone, we are always seeing new technologies, we are actually up to date with the latest trends, and if we can afford those (I could afford this damn expensive phone), it’s hard to have a low end phone which can’t do stuff that already exists on another platform, yeah, but that’s how it is, just like there a lot of apps that specialize first in iOS keeping Android as a second plan because it doesn’t generate as much money, and this will also happen with Android and FxOS, where people are still not sure about the web against native, but it sure has an advantage, as making webapps are cheaper and way more open than the other alternatives.
I’m trying to say that FxOS not running apps available on Android is not a “problem”, but that it’s probably you that has requirements over the ones FxOS are now supporting because your life style asks for more, which is not a problem, it’s just that the phone is still too young for you, just as it is for me.
So basically, my point to the title point is that allowing to run Android apps on FxOS is not only a software problem , but it’s also not a solution, it’s actually just a patch to allow developers to continue doing things wrong.
But sure, as an OS it has a lot to grow up, and I can’t wait to see it as a powerful OS, but that is something that requires time and experience, and it won’t be by adding an option like running Android apps that will suddenly turn FxOS phones better.
@ssj4maiko ok, end of 3d (and discussion … not productive)
ultra low cost is your customer segment (in your head by mozilla marketing)
but there are low cost, middle cost customer segments possible for ffos (I have market research data)
of course customer segments are commercial, business related
and for low cost, middle cost customers (in europe, usa too … not only south america, africa, india, asia) few applications in the marketplace is a great problem
said that, the technical overhead of 2 parallel full stack app execution remains
Actually, there are a lot of people that don’t even know what they are buying. I know someone (a non-tech girl) who bought a samsung phone and was not getting to install Instagram. When I looked at it, it was a Bada OS phone, which of course, doesn’t has Instagram. she bought it because it was cheaper than the Galaxy Y by just a few bucks (A few months later I would find out what was the Galaxy Y, and damn that’s a real piece of crap). Anyway, she was pretty mad after I told her about it, but she knew she was at fault.
So, does it work as a phone? Can you open the internet and check stuff? Can you access facebook and post photos? That’s the basic idea, everything else are just things that you find out later, for bad or for worse. Those who actually know more, like us, sure, we look for the best market, and we help our friends and relatives with these decisions.
@ssj4maiko of course when I say “for low cost, middle cost customers few applications in the marketplace is a great problem” I have customer interviews, surveys, data.
I don’t think to my needs (as advanced - tech user)
I think to customers needs (their requests)
Fortunately or unfortunately not all buy a phone knowing nothing, as any other appliance.
And sell incomplete stuff to people, knowing in advance that after, using the device, they will complain (with you it’s a bad trade policy.
Interesting discussion, I think. The dual boot gets my vote as the easiest way to do this.
I imagine there is still a long way to go to see the potential of FFOS, and other technology working together like webRCT for example as well.
This discussion does surface for me that issue of targetting the right segment of users and their needs how FFOS has a specific potential for apps that exploit the open architecture of the web. I can see that this does require that creative developers think a bit differently than when working on a platform like Android.
Good discussion, it helped me learn quite a lot. Thanks
Yep, dual boot is a solution.
But maybe a Firefox OS running on top of android could be a good solution.
FFOS quick diffusion and all apps for the user
With virtual environment (lxc, kvm, etc.) or a software layer like gecko on android
@sergio_berisso I think we had a talk about linux containers about this same thing a month or two ago in a Facebook community
Firefox OS is borned to create an ecosystem free, secure and open for a Mobile system.
The apk tecnology is not the same.
Firefox os is ultra light, an android system not.
Firefox os don’t have a X server for rendering the OS, is use Gecko so for create a system that support apk is technologically impossible and useless.
Use a free android like Cyanogenmod with Firefox for android like me is better and simple.
A dual boot phone with Android and Firefox OS? Geeksphone Revolution that is official and the Nexus 5.
Why a mobile OS with above another SO? I think that is useless, before we need a complete and functionally working with amazing feature Firefox OS after this i think that a solution can be thought.
