I'm interested in seeing Ubuntu getting a fair shot at the market. I don't know what to think about web-apps based operating systems like Firefox OS yet, but I think it will help Firefox, the browser, become more competitive from what they learn making the OS.
I don't care as much about Tizen and Sailfish, but I'd rather see strong open source competitors standing up against Android and iOS, than more closed ones like WP8 and BB10.
I think, if not Samsung, at least more desperate companies like HTC, LG and Sony should definitely give Ubuntu a fair try. Asus might do it anyway, as they are already making Ubuntu laptops, and they have that whole Padfone thing going on, but they haven't been very successful in smartphones yet.
> I don't know what to think about web-apps based operating systems like Firefox OS yet,
I think it is a mistake. A modern operating system, particularly on mobile where power management and security are concerns, needs a mature well-optimized managed code environment to write middleware and more complicated applications in - hence Objective-C with automatic reference counting on iOS and Java/Dalvik on Android.
HTML5 for all uses under the sun is going to either underperform users' expectations or drain power, or both; and letting unmanaged C applications in is a recipe for disaster.
I would like to be excited about this, but I do not think the architecture as presented makes any sense.
Objective-C is not "managed". iOS apps are sandboxed, but obj-c is very near to C (and most of the Foundation API are pure C). Also, ARC is not mandatory and there is no garbage collector on iOS.
Nonsense. If my Objective-C app is sandboxed, it is "managed". Saying ARC "is not mandatory" is another way of saying it is "enabled by default". There is little formal computational difference between ARC and a conservative garbage collector.
Pros:
- A real linux OS phone, more open than Android
- Ubuntu getting more popularity
- Gives Apple, Google, MS less control/power, users more diversity
Cons:
- Canonical likes making big mistakes (mandatory Unity interface, Amazon search/spyware integration)
This is the first headline I've ever seen that violates Betteridge's law of headlines[0], which states that any headline that ends in a question mark can be answered by "no". But in the case of this headline, the answer is a resounding "yes".
Why not? I don't see a reason to use it myself, but hey, a diverse software ecosystem is a healthy one, and good ideas can come from anywhere. So, "yes".
Of course we need competition. Android is a nice platform - I use it every day. But I also would very much like one or more vanilla GNU/Linux phone platforms. The Linux ecosystem has benefited a great deal from having multiple distros(fedora,debian,gentoo...) with each their own focus. As long as they will agree on some common standards like freedesktop.org - diversity is a win.
I would absolutely love to see Ubuntu and others (Jolla) succeed. Unfortunately I'm also highly skeptical they will. The reason is that currently their marketing story seems so far removed from actual consumers: think about the fact that they're even trying to sell the idea of an operating system, not a complete product. This seems really backwards compared to the evidently successful Apple model. Granted, they are probably having discussions with manufacturers right now, but we'll have to wait for the results. I'd be happy to hear somebody refute this argument!
Ubuntu for phones is not (yet) a choice available to consumers. I assume their marketing is currently targeted at OEMs because that's who they need to convince first.
Hopefully when phones running Ubuntu are available on the market, they'll come up with a more compelling story for consumers.
This, and also Android's overall userspace design is kind of idiosyncratic. I'm sure there are other legitimate designs of a Linux-based mobile OS, which don't rely so heavily on Java and a virtual machine. Ubuntu will be Qt-based, for example. That throws out the possibility to reuse the Android application framework.
Ok, but there's nothing that prevents them from implementing QT on Android using the NDK. You don't have to throw out all application compatibility to achieve something like this.
If I remember correctly then it can also run Android apps in the same WM together with other apps. It actually shouldn't be too hard to replace Android's WindowManager with some other implementation (say, QT-based) due to the nature of Binder (Android's IPC mechanism).
Android.com only lists 4.1 as available. Either way, by only making source code available on the day of release it is by definition already old. That means development happens in the dark.
Wouldn't a real Linux OS on a smartphone be every developer's dream? I will definitely get a Ubuntu phone. Think about the the variety of options you'll have...
However I am not sure if the mainstream market will be as interested.
You are correct that there's a VM -- Dalvik is the process virtual machine in Android, and that a common use case is to use Java bytecode and then convert it to Dalvik, but this is a far cry from being "trapped in a Java VM".
In fact, at even the fundamental level, the Java VM is a stack-based machine, while the Dalvik VM is a register-based machine (the merits or criticisms of doing so are beyond the scope of this post; that's a different discussion).
And if the notion of a VM is still utterly offensive, you are welcome to write portions of your Android applications in native code, such as through C or C++, although for many applications there's really no or marginal benefit to doing so.
Android does use the linux kernel and does have some of the GNU stack, but it's not directly comparable to Ubuntu.
I believe that we are going to be using our phones as our desktop computers in the not-too-distant future, and what Ubuntu is doing is one of the first steps to achieve this. Android has tried to do this in the past (Atrix), but the desktop experience was abysmal because the experiences were inconsistent.
I'm very excited to see how this develops, but Canonical is going to need some good endurance to see this to a point where it can compete commercially.
The real Linux aspect is definitely something that many developers find appealing. I still meet a lot of hackers who couldn't think of using anything else than the ageing Nokia N900. Hardware keyboard, Debian packaging, root out of the box, ...
Developers are a small percentage of the overall market. Sure, we might think it's cool because it's running Ubuntu. But the vast majority of the population is going to look at it and say "How is this going to make my mobile experience better in a way that I don't already have from IOS or Android?"
I wish Canonical luck, but I just don't see my non-technical friends caring one whit about this phone unless it does something extraordinary that they can't get elsewhere. And I just don't see that happening.
hmm, meego was 'real' linux. I had it on a tablet. It was horrible :). Intel gave every developer one in an event. I think the next day everyone had installed a different OS :)
The phrase "A real linux OS phone" is pretty frustrating considering Android exists and is supposed to be "real Linux". What are the main reasons Android is not "real Linux"? (Its hard not to be upset with Google for this situation if the reason is what I think it is).
When people say Linux, they usually mean more than just the kernel. Stallman insists we call this GNU/Linux since a good portion of userspace comes for the GNU project. When people say 'real linux' I believe they are referring to the full userspace that usually accompanies the Linux kernel.
The userspace in Android is very different from the userspace in GNU/Linux. In GNU/Linux, it is very easy to program using your preferred programming language. This isn't the case in Android, everything is very tightly wrapped up around Dalvik. Before you mention the NDK, let me say that it's only a subset of C++ (no exceptions among other things). Ubuntu on the cell phone also opens up the possibility of things like shell scripting, local webservers, etc.
What this really comes down to is programming freedom that you can find in GNU/Linux that isn't found in Android. You will now have a full fledged computer in your pocket instead of a limited device.
It says something about the fashion sense of a reporter if they approach the question this way. You should ask, "What capabilities are we going to want in our phone going forward?" and work from there, but fretting over brand awareness seems, well, a bit shallow.
Do you only buy a car because it has a "Hemi" V8 engine ? And wonder, "Do we really need another engine for cars?" Of course you don't. I would love a smart phone that was less susceptible to malware (the tension of programmability meets actual exploits against my bank account). If the current OS makers can do that, fine, if there is a new OS that does this better, that is ok too.
43 comments
[ 3.4 ms ] story [ 97.9 ms ] threadI don't care as much about Tizen and Sailfish, but I'd rather see strong open source competitors standing up against Android and iOS, than more closed ones like WP8 and BB10.
I think, if not Samsung, at least more desperate companies like HTC, LG and Sony should definitely give Ubuntu a fair try. Asus might do it anyway, as they are already making Ubuntu laptops, and they have that whole Padfone thing going on, but they haven't been very successful in smartphones yet.
I think it is a mistake. A modern operating system, particularly on mobile where power management and security are concerns, needs a mature well-optimized managed code environment to write middleware and more complicated applications in - hence Objective-C with automatic reference counting on iOS and Java/Dalvik on Android.
HTML5 for all uses under the sun is going to either underperform users' expectations or drain power, or both; and letting unmanaged C applications in is a recipe for disaster.
I would like to be excited about this, but I do not think the architecture as presented makes any sense.
Fact: Objective-C is "managed".
Cons: - Canonical likes making big mistakes (mandatory Unity interface, Amazon search/spyware integration)
Mandatory?
./call 01
[0]http://en.wikipedia.org/wiki/Betteridge%27s_law_of_headlines
Hopefully when phones running Ubuntu are available on the market, they'll come up with a more compelling story for consumers.
Android is open source. Fork it. Improve it. Why build something completely new. Is it just to keep a brand name like "Ubuntu" and "Firefox"?
android was also deemed the least open of a bunch of high-profile open-source projects: http://www.visionmobile.com/rsc/researchreports/Open%20Gover...
However I am not sure if the mainstream market will be as interested.
You are correct that there's a VM -- Dalvik is the process virtual machine in Android, and that a common use case is to use Java bytecode and then convert it to Dalvik, but this is a far cry from being "trapped in a Java VM".
In fact, at even the fundamental level, the Java VM is a stack-based machine, while the Dalvik VM is a register-based machine (the merits or criticisms of doing so are beyond the scope of this post; that's a different discussion).
And if the notion of a VM is still utterly offensive, you are welcome to write portions of your Android applications in native code, such as through C or C++, although for many applications there's really no or marginal benefit to doing so.
I believe that we are going to be using our phones as our desktop computers in the not-too-distant future, and what Ubuntu is doing is one of the first steps to achieve this. Android has tried to do this in the past (Atrix), but the desktop experience was abysmal because the experiences were inconsistent.
I'm very excited to see how this develops, but Canonical is going to need some good endurance to see this to a point where it can compete commercially.
For this group something like Jolla's Sailfish (see http://bergie.iki.fi/blog/jolla-sailfish/) or Ubuntu Mobile will be very appealing options.
Disclaimer: I've worked with some of the people behind both Sailfish and Ubuntu Mobile back in the good old Maemo/MeeGo days.
I wish Canonical luck, but I just don't see my non-technical friends caring one whit about this phone unless it does something extraordinary that they can't get elsewhere. And I just don't see that happening.
Do we really need more screen sizes?
Do we really need more apps?
That's an inane question, and the article doesn't even try to answer it.
The userspace in Android is very different from the userspace in GNU/Linux. In GNU/Linux, it is very easy to program using your preferred programming language. This isn't the case in Android, everything is very tightly wrapped up around Dalvik. Before you mention the NDK, let me say that it's only a subset of C++ (no exceptions among other things). Ubuntu on the cell phone also opens up the possibility of things like shell scripting, local webservers, etc.
What this really comes down to is programming freedom that you can find in GNU/Linux that isn't found in Android. You will now have a full fledged computer in your pocket instead of a limited device.
It says something about the fashion sense of a reporter if they approach the question this way. You should ask, "What capabilities are we going to want in our phone going forward?" and work from there, but fretting over brand awareness seems, well, a bit shallow.
Do you only buy a car because it has a "Hemi" V8 engine ? And wonder, "Do we really need another engine for cars?" Of course you don't. I would love a smart phone that was less susceptible to malware (the tension of programmability meets actual exploits against my bank account). If the current OS makers can do that, fine, if there is a new OS that does this better, that is ok too.
Why wired? Join the discuss here.