By Jonathan Corbet
February 21, 2012
As many observers have pointed out, the phone handsets that many of us
carry now exceed the power of the laptops we were carrying not all that
long ago. The much-hyped Galaxy Nexus, for example, includes a 1280x720
display, 32GB of flash storage, 1GB of RAM, a 1.2GHz dual-core
processor, and a number of interesting peripherals never found on that old
laptop.
And, of course, there is a Linux kernel running the whole thing. Given
that, one might well wonder why one should still bother carrying a laptop
around. Canonical, it seems, believes a number of people are wondering
that; thus the
announcement of
Ubuntu for Android, an
interesting attempt to move laptop-based activities onto the handset.
Ubuntu for Android is intended for handsets that can be docked and will,
thus, have a keyboard, mouse, and display available. In that setting, it
will provide the usual, Unity-based Ubuntu experience on that external
display; the Ubuntu system essentially runs inside its own container on top
of the Android kernel. The interface on the handset itself, meanwhile,
remains pure Android. So
Ubuntu for Android can be thought of as providing two distinct
personalities for the device. There is some data sharing between the two -
the contacts database, for example - but they remain mostly separate from
each other. Rather than create a single integrated interface to the
handset, Canonical has made something closer to a dual-boot system - except
that the two can run simultaneously on their respective displays.
According to Canonical, the split system is the best solution:
Android is a mobile solution, designed for a touch interface on a
handheld device. On the desktop, where users expect a
pointer-driven experience, a PC operating system is
essential. Several vendors have tried to bring Android-based
desktops or laptops to market, with no success; Android was
designed for touch only, and has its hands full winning the tablet
wars.
Even a well-equipped phone does not have vast amounts of storage by
contemporary standards. But even with more storage, it seems likely that
users of Ubuntu for Android would want to have their files available
outside the handset as well. So it is not surprising that this system is
cloud-heavy. So there is no LibreOffice by default; instead, the system
expects to use the Google Docs service. It does provide
Thunderbird, though one might imagine that its storage-intensive indexing
has been disabled by default. For good measure, Ubuntu TV has also been
built into the system.
The hardware requirements (found on the features
page) rule out a lot of devices, but are certainly not out of line for
a current high-end device. Ubuntu for Android wants a dual-core CPU
(clocked at 1GHz or higher), video acceleration and the ability to produce
HDMI output from a secondary frame buffer device, and 512MB of RAM. The
need for a dock for the phone to provide HDMI and USB ports is implied; few
devices have the requisite connectors without a dock. As Canonical points
out, the hardware requirements are easily satisfied by devices that are in
development now.
So Ubuntu for Android seems like a useful and feasible development. The
unfortunate part is that it is not available for users or developers to
play with. Canonical is clearly hoping to sell this offering to device
manufacturers and carriers; as this page
makes clear, shipping it will involve per-unit royalties. Canonical
clearly believes that vendors may find those royalties worthwhile, though,
as a way to sell more high-end devices:
Ubuntu for Android gives mobile workers a compelling reason to
upgrade to multi-core handsets with more RAM, more storage, faster
GPUs and CPUs. It’s not just a phone they are buying, it’s a
desktop too. While mid-range phones can deliver a perfect Android
experience, it takes high-end horsepower to drive a phone and a
desktop at the same time. Newer multi-core processors are up to the
job, and Ubuntu is the killer app for that hot hardware. It’s the
must-have feature for late-2012 high-end Android phones.
Canonical also pitches the idea that a bundled Ubuntu desktop will drive
demand for fast broadband offerings (LTE, for example) from the carriers.
And they claim that it could be especially attractive in parts of the
developing world where high-end handsets are being sold to customers who
have never owned a computer before. Such people, Canonical says, have
"no legacy attachment to the desktop" and will find a combined
offering compelling.
This reasoning may make some sense; it is possible that hybrid, handheld
Linux-based systems will bring about the year of the Linux desktop after
all. But there are a couple of concerns worthy of note. One is that users
may quickly tire of having two different interfaces to the same system,
leaving Ubuntu for Android vulnerable to a competing system with a more
integrated experience. One can imagine, after all, that, if this idea goes
anywhere at all, there will be Windows- and Mac OS-based variants
available in short order - and, perhaps, other Linux-based implementations
as well. Some of these systems may look like less of a hybrid and, as a
result, be more successful.
The other concern is that Canonical appears to be taking a step toward
proprietary systems. If there are plans to offer this functionality
directly to users, or to enable it to be bundled with a distribution like
CyanogenMod, Canonical has not disclosed them yet. Instead, we have a
system that, by all appearances, will only be available in binary form from
manufacturers or carriers. Source for GPL-licensed components will
naturally be available, but it is far from clear that Ubuntu for Android
will be all free software; vendors like Citrix and Adobe feature
prominently on the product's page. It is also not clear that
device owners will be able to modify the distribution to their own liking
and run the result on their devices.
A handset or tablet that can run a full Ubuntu system has some appeal; one
running a locked-down Ubuntu system would be rather less exciting.
Ubuntu for Android is clearly an important step in the evolution of the
"desktop" away from traditional personal computer systems. It has a lot of
potential as a practical replacement for bulkier systems. But, to be
commercially successful, Canonical will have to convince a lot of people
that the Unity-based desktop is what customers want. And to be successful
as free software, it will have to result in free systems under the control
of their owners. It will be a sad day if the Ubuntu community of the
future is focused on the creation and propagation of tools to jailbreak
their Ubuntu systems.
(
Log in to post comments)