Why Google needs to put Chrome OS in developer's hands - Techies Updates

Breaking News

Monday, October 15, 2018

Why Google needs to put Chrome OS in developer's hands

Google wants to disrupt with its stateless, cloud-focused desktop operating system. But it needs to become much more viral with exploitative apps if it expects it to succeed beyond just being a niche platform for education and light computing tasks.

Apparently, I wasn't the only one who noticed that, during Google's presser unveiling of the Pixel 3 and Pixel Slate, the word "Android" wasn't mentioned a single time.

Who's Google turning into now? Amazon? The open-source OS that shall not be named?


There may be a number of reasons for this: As a brand name, Google already appears to be phasing out Android, as it pertains to platforms (with Android Wear becoming Wear OS), and dropping the moniker from a number of its own apps, such as with Pay and Messages, preferring to just use the name Google as the main brand for many of its applications and services.

There are also a number of legal issues being worked out, and it is a possible major liability for the company in its litigation with Oracle.

We also know Google is working on a successor to Android, called Fuchsia. But it could be several years away before we even see functional beta versions of it.

But, for now, we have Android and its various APIs as the main programmatic target for apps running on its two main platforms: Android Pie and Chrome OS.


Android has a fairly robust and very mature SDK, Android Studio, which provides a comprehensive IDE and a device emulator on Windows, Mac, and Linux so that apps can be tested and debugged before deploying to a target device.

No such SDK equivalent or installable runtime environment exists for Chrome OS -- which runs Android applications. If you want to develop for Chrome OS, you need to build Chromium OS from source and try to install it on generic PC hardware or a VM.

Like AOSP, it does not include the Google Play Store or Android runtime pieces, as they have to be licensed, so it's kind of useless -- unless you are a hardware OEM like Asus or Samsung and are looking to produce your own Chromebooks.

Or, alternatively, you need to buy Chrome OS hardware: a Chromebook, Chromebox, or Chromebit (which was launched by ASUS in 2015 and hasn't seen any hardware updates). If you want the most up-to-date Chrome OS builds first, as with Android, you will ideally want to run it on Pixel hardware. Which is more expensive.

That's still a significant expense for an application ecosystem that isn't entirely proven or optimized well for that platform. While Android apps all pretty much universally "run" on Chrome OS, the majority of them are not tablet or desktop-optimized, which makes for an overall poor user experience.

This problem needs to be solved if Pixel Slate and similar Chrome OS convertible systems have any chance of gaining significant market share against iPad and Windows tablet convertible systems like Surface.

While $300 (or even $80 for a Chrome bit assuming it gets updated) doesn't seem like a lot of money, it's not inconsequential for programmers and students and one-man shops in developing countries, which have been able to leverage the Android SDK in the past to put their apps in the Google Play Store. Some of the very best apps for Android have come out of that talent pool.


Also, not being able to run the software in a virtual environment on a Mac or a PC on Windows or Linux -- and taking advantage of that isolation and ability to prototype on multiple instances -- is a severe limitation.

Over the years, there have been a few third-party solutions for installing Chrome OS on native hardware or VMs, such as Neverware CloudReady, which is a free download for non-commercial use and is a great way to re-use older PC equipment and provide a nice secure environment for students and other use cases where a full Windows, Mac or Linux desktop environment isn't needed. The company also recently purchased Flint OS, another Chrome OS "distribution," based in the UK.

I've used CloudReady to re-use old laptops and breathe new life into them. Prior to buying Chromebooks for older family members, it's how I got Chrome OS up and running. I highly recommend it, it's an extremely polished product.

But Neverware doesn't have the Google Play Store installed, and the company has no intention of licensing it. It also doesn't behave well in a VM, as the pseudo-device support for higher-resolution displays doesn't exist in it.

It would behoove Google to create an installable runtime environment for Chrome OS that runs without hassle on PCs and Macs. It should be able to be installed on the metal, and ideally, its primary use case should be in a VM or a container, and that virtualization/containerization support should come with the distribution click-to-run as a self-enclosed application.

The container/VM use case would be beneficial because it would provide an isolated runtime environment provides that extra layer of protection to the end-user. On native Chromebook hardware, Chrome OS uses a combination of specialized firmware and sandboxing to prevent malware hijacking.

On legacy PCs, it's not necessarily possible with BIOS to do that, and even EFI has its limitations, as well. But, in a VM or a container instance, that functionality can be replicated.


There is no reason why, when this is implemented, an installable Chrome OS virtual desktop with an integrated Google Play Store cannot replace the regular Chrome browser -- because it simply adds Android application support and additional security.

Once Chrome OS is in the hands of tens of millions of existing users in this manner, it becomes viral. Because then everyone using it is now a potential Chrome OS developer, and it puts the Android API stack everywhere.

Does Google need to make Chrome OS viral and installable everywhere? Talk Back and Let Me Know.


No comments:

Post a Comment