Back in March 2011, Brian Madden wrote an article about why Android thin clients will rule, which generated some good discussion over the prospect of using a powerful mobile operating
The idea was that those clients could serve as open source replacements for the Linux-based ones currently on the thin client market. That would be beneficial because you'd have:
- A large number of native apps
- Familiar OS interface
- Existing, comprehensive remoting clients
- The ability to place one OS on all your thin client hardware, no matter the manufacturer
I say "next to nothing" because there are a few Android thin client options out there, but let's face it: With marketing taglines like "Let network applications generally successful (sic)" and product images that look like they were edited in Microsoft Paint, they're not exactly taking the world by storm.
More on the thin client market
Choosing the best VDI thin clients
Vendors churn out thin client hardware
Testing an Android thin client
Due to the lack of options and an excess of curiosity, I recently bought a Logitech Revue Google TV device just to do my own testing. Figuring the Google TV was designed for high-resolution displays and Flash video, I thought it might have the horsepower to work with virtual desktops.
What I learned was that either the device is woefully underpowered (which is likely, because they're a few years old and all but forgotten), or the applications designed for remote desktop access aren't optimized for this device.
That, actually, could be the problem with an Android thin client. The successful companies in the thin client market have tight control over the hardware and software that comprise their solution. Much like Apple, which manages its entire supply chain, those vendors require everything to be tightly controlled for better manageability and cost savings. With Android in the mix -- and perhaps all the weird hardware platforms -- you never know what you're going to get.
The Logitech Revue as a potential Android thin client exemplifies that. Here, we have an Intel Atom-based device running an OS that was designed for ARM processors and ported to Intel, trying to run applications that, for one reason or another, were also ported to Intel, despite originally being made for ARM processors. So now we have a Wild West situation, where the hardware isn't meant to be running the OS and the apps aren't meant to be running on that OS.
That all adds up to a TV-watching solution that's just OK, with almost no potential to be anything more than that -- including an Android thin client.
I did manage to find a Remote Desktop Protocol client for the device, and it does run. You can watch blocks of pixels as they slowly draw on the screen at 1920x1080 resolution, and the application's interface was clearly written for a phone. In fact, it's a wonder that it even worked on this Intel-based box at all.
Although that experiment failed, it also highlighted some of the shortcomings of running Android as a thin client. The bulk of the thin client market's hardware is Intel-based, and adding Android to that is an uphill battle. The existing Linux-based OSes, despite limitations because they're tightly controlled, are much more predictable and stable. To me, that makes them better.
Perhaps as ARM-based processors expand from the mobile device market and to more traditional form factors, we'll see Android become more fashionable and break into the thin client market. That's probably the only way we'll ever see a successful Android thin client, because it looks like the Logitech/Google/Intel experiment has been put out to pasture.
This was first published in July 2012