Live data from Hacker News

Firefox OS devices officially released

hacks.mozilla.org

81–90 of 146 posts

Re: Firefox OS devices officially released

#81
post #58
post #17

Earlier quoted context omitted.

I'm not sure about asm.js/OdinMonkey. I believe that FirefoxOS 1.0 does not have it, but that FirefoxOS 1.1 (which should be available as a free over-the-air update) has it. More generally, FirefoxOS does not automatically get the latest version of Gecko, because the process for mobile operating systems is more complex. The operating system must first pass certification by (if I recall) both the constructors and the…

1.1 is still the b2g18 branch, so no asm.js. 1.2 will be the version that pulls from master (gecko 26 I think?) and will have asm.js.

Oh, it's 26? In that case, they have not missed my recent I/O optimizations :)

Re: Firefox OS devices officially released

#82
post #17

Earlier quoted context omitted.

I'm not sure about asm.js/OdinMonkey. I believe that FirefoxOS 1.0 does not have it, but that FirefoxOS 1.1 (which should be available as a free over-the-air update) has it. More generally, FirefoxOS does not automatically get the latest version of Gecko, because the process for mobile operating systems is more complex. The operating system must first pass certification by (if I recall) both the constructors and the…

Can you give an idea of how you expect that to work compared with Android? Given that Firefox OS devices seem to be cheap and low end, how likely are constructors and operators to perform a "long and costly" process to upgrade them regularly? Cheap and low end Android devices have terrible support from both, even for upgrades without new hardware requirements, is there any indication that they will better support sim…

I believe that in a previous thread, FirefoxOS devs were hopeful that the way the OS is "layered" will make this less painful -- that upgrading Gecko should be much easier than upgrading Android versions, because it wouldn't introduce driver issues etc.

Updating the underlying kernel would be more arduous.

Re: Firefox OS devices officially released

#83
post #55

Earlier quoted context omitted.

>Right, but in Firefox you also have "native" HTML5 apps, which give access to more hardware functionality than a regular web app. Which negates the sole benefit of a web app in the first place since those APIs are not standard and not available on other platforms, so those apps are not portable. Maybe the APIs will become standard and implemented on Android and iOS etc, but at that point you can write those kinds of…

Actually, almost all of them are available on Android and on Desktop. Interestingly, while you seem to consider that this is a weak point of Firefox, we (Mozilla) did this on purpose and we consider this one of the strongest points of the platform: _no vendor lock-in_. If you write an application for FirefoxOS, we do not place any obstacle that can prevent the developer from distributing the application on other devi…

>Interestingly, while you seem to consider that this is a weak point of Firefox, we (Mozilla) did this on purpose and we consider this one of the strongest points of the platform: _no vendor lock-in_.

There is already no vendor lock-in with html based apps on Android or iOS. The complaint people have about iOS is that it is locked down with regards to native apps. Mozilla themselves have made this exact complaint about iOS -- that they wanted to write a browser for iOS but could not because a) HTML/Javascript are not suitable for writing a browser so they have to go through the app store and b) native apps in the app store on iOS face restrictions that would limit performance of a Javascript VM, so Firefox would not have competitive performance compared to Safari.

FirefoxOS solves neither of those problems.

More generally, the idea of all user facing software being targeted to a single sandboxed standard API is very restrictive and worrying. It means developers giving up complete control to a small cabal of 4 or 5 large organisations that decide what gets standardised. Organisations that have conflicting interests and financial motivations that certainly don't align with my own. That's assuming that various competing web based platforms would even be sustainable; I don't see it lasting more than 5 mins before platform specific extensions are added to provide additional value over the competitors.

I want to be able to write software for my device and have full access to the hardware without some third party being able to decide what I can and can't do with my phone. FirefoxOS doesn't get me any closer to that goal, in fact it's a step backwards. Web platforms are great for SV startups where raw user numbers are the most important thing, but to be honest I don't give a shit about the ability of the latest photo sharing startup to 'exit'. I do care about people doing much more interesting things and all I can see is web based OSs being a restriction on that. Think about how you would feel living in a world of Gopher based operating systems.

>As you may recall, we (not just Mozilla) did this to the web a few years ago, and look where this brought us all

Yes, a world where I don't control my own data anymore, it lives on some server I can't access, subject to the whims of foreign governments. Being processed by code I can't even reverse engineer never mind read the source or modify. This is great freedom for developers and governments, but utterly disastrous for the freedom of users.

Re: Firefox OS devices officially released

#84
post #65

Earlier quoted context omitted.

Opennes's nice but is it single-person hackable or huge monolithic blob that nobody except the most skilled teams would dare to bite? My problem with Android is that I miserably failed to hack it alone (I wanted to mess with app permission system and make it work my way). The Behemoth seem to have quite a steep learning curve, so I quickly lost my way in its innards. However, I was able hack Maemo (except for proprie…

What's your opinion on Ubuntu-phone ?

Haven't tried one (don't have the readily compatible hardware), so can't really tell.

However, from what I've heard, it's Cyanogenmod 10 with Dalvik parts removed and some Ubuntu parts (no X11, Qt5 only GUIs) running in a sort of chroot environment. Doesn't sound much better than any other chrooted GNU install.

Re: Firefox OS devices officially released

#85

Earlier quoted context omitted.

However, it is VERY restrictive in its Live account interaction. I had some very ugly experiences setting up one for my cousin (I stupidly didn't lie and registered her as being from Uruguay and under 18), which made the phone behave like a brick ("sorry, no apps available in your country", "sorry, you have to be over 18 under laws from another country to use most of the features of this phone"). Changing the Live ac…

That doesn't surprise me. Regional restrictions on apps might not make much sense, but they're well known and exist on every platform. Same with age restrictions. You can't enter into a contract with Microsoft or a third party developer unless you're over 18. It's not just Microsoft that forces this.

"You can't enter into a contract with Microsoft or a third party developer unless you're over 18"

That only applies for the United States and I guess most European countries, but it's not universal.

http://en.wikipedia.org/wiki/Age_of_majority

For example, here in Uruguay, there's a distinction between minors up to 16 years, then they can enter some contracts (for example work contracts, I don't know where downloading apps falls), and they become full adults at 18.

In the U.S. there are some restrictions up to 21 years old, but they can drive at 16 (!). My cousin was 17 at the time I was setting up her phone, I'm not sure, but I guess she might have been legally able to enter a contract with Microsft at the time.

Google is more lax with its settings, or has better lawyers. Also, if you move to another country, you CAN change your settings (it's extremely ridiculous that Microsoft doesn't let you do that).

I can totally understand that Microsoft doesn't want to go over the minutiae of every small nation to see whether someone is eligible or not to enter a contract.

I wonder what Firefox OS will do with all these legal restrictions (offload to the app developers?)

Re: Firefox OS devices officially released

#87
post #55

Earlier quoted context omitted.

Actually, almost all of them are available on Android and on Desktop. Interestingly, while you seem to consider that this is a weak point of Firefox, we (Mozilla) did this on purpose and we consider this one of the strongest points of the platform: _no vendor lock-in_. If you write an application for FirefoxOS, we do not place any obstacle that can prevent the developer from distributing the application on other devi…

>Interestingly, while you seem to consider that this is a weak point of Firefox, we (Mozilla) did this on purpose and we consider this one of the strongest points of the platform: _no vendor lock-in_. There is already no vendor lock-in with html based apps on Android or iOS. The complaint people have about iOS is that it is locked down with regards to native apps. Mozilla themselves have made this exact complaint abo…

> There is already no vendor lock-in with html based apps on Android or iOS. The complaint people have about iOS is that it is locked down with regards to native apps [...]

We agree on this point.

The strategy chosen by Mozilla with FirefoxOS, for good or for ill, is to standardize upon the subset of applications that have no vendor lock-in, i.e. the html based apps, and then to progressively extend the set of capabilities granted to these applications, while still avoiding the lock-in. Other open mobile OSes make different choices (e.g. Ubuntu mobile), and I personally wish them success. The future will tell us which one, if any, of these attempts to make the world a little more open will succeed.

> I want to be able to write software for my device and have full access to the hardware without some third party being able to decide what I can and can't do with my phone. FirefoxOS doesn't get me any closer to that goal, in fact it's a step backwards. Web platforms are great for SV startups where raw user numbers are the most important thing, but to be honest I don't give a shit about the ability of the latest photo sharing startup to 'exit'. I do care about people doing much more interesting things and all I can see is web based OSs being a restriction on that. Think about how you would feel living in a world of Gopher based operating systems.

That is an interesting point. As all new generation, high-level platforms, FirefoxOS opens some doors and closes others.

Now, I don't know your personal interests, your experience, or exactly what low-level code you want to write, but that reminds me of something I experienced quite a few years ago. At that time, the newly released Windows 95 killed a number of low-level libraries that I had carefully hand crafted using assembly code and BIOS calls and that let me read/write floppy disks with custom formatting, hence guaranteeing that nobody could read my floppy disks without my consent (they could not be ported to Linux either, of course). It also killed my low-level asm-based libraries for direct access to the video card, the libraries that had taken me years to complete, and that I could use to write smooth 256 colors side-scrollers.

Well, I survived that loss. A few years have passed. Now, FirefoxOS is the first platform I have experienced in which I can write a small game prototype in about 2 hours. Some doors have been closed, others have opened.

It may be that FirefoxOS is not the platform of your dreams, if your dreams require you to go low-level. But I believe that, for most of us, FirefoxOS is pretty damn good.

> >As you may recall, we (not just Mozilla) did this to the web a few years ago, and look where this brought us all

> Yes, a world where I don't control my own data anymore, it lives on some server I can't access, subject to the whims of foreign governments. Being processed by code I can't even reverse engineer never mind read the source or modify. This is great freedom for developers and governments, but utterly disastrous for the freedom of users.

While the dangers you describe are very real, you seem to forget the situation we avoided – one in which every single communication on Earth had to go through Microsoft, using proprietary code and protocols, with contagious DRMs & authentication on every single one of your documents, with a single company deciding of the tools at your disposal both for office work and for development.

The fight is far from over – it may never be – and the Mozilla community is part of the frontline, but the open web was a considerable push in the right direction, one that at least gives us a fighting chance.

Re: Firefox OS devices officially released

#88
post #55

Earlier quoted context omitted.

>Right, but in Firefox you also have "native" HTML5 apps, which give access to more hardware functionality than a regular web app. Which negates the sole benefit of a web app in the first place since those APIs are not standard and not available on other platforms, so those apps are not portable. Maybe the APIs will become standard and implemented on Android and iOS etc, but at that point you can write those kinds of…

Actually, almost all of them are available on Android and on Desktop. Interestingly, while you seem to consider that this is a weak point of Firefox, we (Mozilla) did this on purpose and we consider this one of the strongest points of the platform: _no vendor lock-in_. If you write an application for FirefoxOS, we do not place any obstacle that can prevent the developer from distributing the application on other devi…

This point needs to be stressed more. Not so much that it is 'open' but that you can write an app and it will run everywhere: Mobile, WEB and DESKTOP. If there are good solutions for this now I am unaware of them. Desktop is still important.

Open is great, but is part of the larger, political issue(s). The marketing has to appeal to the 'what does it buy me'?

Re: Firefox OS devices officially released

#89
post #38

Earlier quoted context omitted.

Neither does iOS, for that matter.

It doesn't? All recent MacBooks I have seen insist rather hard to be paired with an account and a credit card number. I have never done the initial setup of an iOS device, but I would have assumed that it was the same or worse.

> All recent MacBooks I have seen insist rather hard to be paired with an account and a credit card number.

That is completely untrue, unless you are seeing some weird install of OS X. But you're not, you're making stuff up.

You can skip the product registration and you can skip signing in with your Apple ID and essentially have an "anonymous" install with little effort.

Re: Firefox OS devices officially released

#90
post #79
post #71

Earlier quoted context omitted.

Those devices still need to be built. So a 4th or 5th place might not generate enough revenue to pay them.

I'm not a manufacturer, but at a guess, I'd say that 4th/5th place is still plenty of billions of dollars.

That depends on the shareholders.
Post reply on HN