Canonical Demos Early Stage Android-On-Ubuntu 165
An anonymous reader notes Ars Technica's report from the Ubuntu Developer Summit in Barcelona, where Canonical has unveiled a prototype Android execution environment that will allow Android applications to run on Ubuntu and "potentially other conventional Linux distributions." "Android uses the Linux kernel, but it isn't really a Linux platform. It offers its own totally unique environment that is built on Google's custom Java runtime. There is no glide path for porting conventional desktop Linux applications to Android. Similarly, Java applications that are written for Android can't run in regular Java virtual machine implementations or in standard Java ME environments. This makes Android a somewhat insular platform. Canonical is creating a specialized Android execution environment that could make it possible for Android applications to run on Ubuntu desktops in Xorg alongside regular Linux applications. The execution environment would function like a simulator, providing the infrastructure that is needed to make the applications run. Some technical details about the Android execution environment were presented by Canonical developer Michael Casadevall... They successfully compiled it against Ubuntu's libc instead of Android's custom libc and they are running it on a regular Ubuntu kernel."
trying not to be a troll (Score:2, Insightful)
What's the point? Most apps use GPS, tilt, and camera that most computers don't have(except for the camera). Those that don't use them are boring calculators and notepads. And even then, for the apps GUI to look right the window is restricted to a 320x480 rectangle or else you wind up with stretched buttons and text boxes.
Re:important lesson (Score:5, Insightful)
Windows Mobile is a broken OS, even the die hard MS fans know that out of the box its broken, sure, you can add software to make it usable, but a vanilla WinMo device is unusable. The iPhone is restricted to one device and one carrier, Android can run on many and is or soon will be on many different networks. Palm OS is severely outdated, but Web OS which is their replacement already has a strong following and the Pre is set to be the next thing in phones.
If Android was marketed as a full desktop or placed on "real" (ie: x86, full keyboard, decent screen) hardware it wouldn't sell because people won't learn a new OS on what they think is a Windows platform and it won't run some applications.
Re:important lesson (Score:2, Insightful)
I don't think this proves anything. Android has a bit of buzz around it, but there are so few handsets commercially available using it that it's popularity is impossible to gauge. It's a bit like saying the iPhone proves that people prefer OS X, but only if you remove the dock.
Re:If I had the choice (Score:5, Insightful)
I'd rather run Ubuntu on my smart phone.
Yeah we know how well that went [openmoko.com].
Re:If I had the choice (Score:4, Insightful)
Maybe one day phones will become open platforms, but yeah, I'm not gunna hold my breath :)
Re:Speed? (Score:4, Insightful)
> According to the summary it seems like it will be emulating everything, that raises a
> real speed concern, not perhaps for newer desktops but for older hardware and netbooks.
Sounds more like a shim than a simulator.
> Wouldn't a better option be to have a second real kernel being launched within the real
> one and native libs, etc?
Not a kernel, no. It might be better to run in a chroot and use the Android libraries, though. Perhaps that is what they are doing.
Re:important lesson (Score:3, Insightful)
They don't have an app store
Great point! Instead there are many stores and places you can download whatever app you want and then install as you wish. Shame on Microsoft.
Re:And Google doesn't get Sued? (Score:5, Insightful)
They aren't marketing the Android environment as Java. That's the key difference.
Microsoft had their own implementation of Java (the VM and the language), which wasn't entirely compatible with Sun's, had additional features that were only present in Microsoft's implementation, and lots of Windows-only libraries. It implemented only a subset of Sun's Java specification, and didn't pass the test suites. They still called it Java, and encouraged developers to use their implementation instead of Sun's. Their development environment and documentation led you straight to using Microsoft's implementations of everything, rather than Sun's, and their made it very hard for developers to tell if their application could run on Sun's VM as well. So in effect, they created their own distinct version of Java, with applications written for one implementation being incompatible with the other, but still called it Java, and still tried to benefit from Sun's Java marketing (including the "Write Once, Run Anywhere" promise). Basically, they tried to usurp the platform, while still using Sun's trademarked Java name to market it. Sun really had no choice but to sue.
Microsoft's Java implementation lived on after that, under the name "J++", and later as "Visual J#". They no longer position it as "Java", but as a Java-language compiler for .NET.
Google, on the other hand, don't mention Java anywhere. You're not writing Java applications - you're writing Android applications. Those applications happen to be written using the Java programming language, and execute inside a Java VM, but that's just an implementation detail. Their main website doesn't mention Java. The first few pages of their developer site don't mention Java, until you get to the page detailing the requirements for running the SDK. They make no attempt to claim that their implementation is Java, or even compatible. They make it clear that Java applications don't work on Android, and Android applications don't work on standard Java.
Re:Speed? (Score:3, Insightful)
According to the summary it seems like it will be emulating everything, that raises a real speed concern, not perhaps for newer desktops but for older hardware and netbooks. Wouldn't a better option be to have a second real kernel being launched within the real one and native libs, etc? I know it might be hard to do and would have security problems, but it seems a lot faster that way.
I'm not sure if it is emulated or not, but even if it's slow to run an app, a desktop likely has a much faster processor than the phone the app was designed to run on, so it would probably be fine normally.
-Taylor
Re:This is where a subject should be (Score:3, Insightful)
Re:important lesson (Score:5, Insightful)
how WM is broken out of the box?
although did hard reset a few times and using vanilla WM for a while.
There is a clue right there.
This is a view I find among Windows users a lot. (I support Windows for a living, servers and Desktops - XP, Vista and even the odd Win2k box and the other day I actually came across an office PC running Win98SE - eek)
The view is this - "Windows is not broken - it runs fine after installing all this, and doing that, and tweaking this, oh and don't forget the Firewall and Antivirus... look, no more crashes!"
The moment Linux comes up in the conversation (they usually ask after watching me troubleshoot the network from my laptop - Ubuntu, or whatever takes my fancy) they have this idea that it takes a lot of tinkering to get it to run properly.
From personal experience I have found a modern Linux distro takes about the same effort to get running to a users liking as would a typical Windows install. The effort is usually expended in different areas over the lifetime of the install as opposed to Windows - but it takes some effort both ways.
On some hardware I might need drivers for Windows or Linux, on others no drivers needed for either. What seems a common theme with Windows installs are general slowing down over the life of the install, random Virus issues that needs to have an eye kept on it, MS updates that break stuff.
With Linux I find that the slowing down over time is not as obvious, if at all, and update related breakages are less common.
But Windows Users will happily spend hours to tinker with a Windows box, but the moment a Linux install needs some effort they throw their hands in the air and yell "This will never be ready!"
Re:Fault != problem (Score:3, Insightful)
And if the driver is for the network card how do you download it? What about if it is for the SATA controller? Or even the motherboard chipset? Sometimes you can not download the driver. But how you get the driver on the system isn't really the issue here.
You don't really understand Linux drivers. Even small kernel changes currently can break a driver. There is no way to create a binary driver and put it on a website or CD and besure that it will work with any given distro.
It isn't just about RPMs and Debs but that is also an issue it is about a lack of a stable binary interface. Right now you can not write a driver and release a binary driver and know with any certainty that it will work with the next 2.6 kernel that comes out.
Driver should work for every version of 2.6 that comes out. That is the real benfit to a stable binary interface. It is something that Windows does do well. XP drivers tend to work for all builds of XP. Many of them will work for Vista as well. Printers and Graphics cards are the big exceptions. As far as I know Vista drivers will work for Windows 7. That makes everybody but the OS developers life a lot simpler.