Google Doesn’t Want Acer To Fork Android?

Acer recently canceled a press conference where it was about to announce a smartphone running an Android fork created by Chinese Alibaba.

Alibaba released a statement, “A Sept. 13  news conference announcing the China launch of a high-end Acer smartphone running a cloud operating system made by Alibaba Group was abruptly canceled after Google, owner of the Android OS, threatened to cancel Acer’s license to use Android for its other phones if the launch went ahead.”

Alibaba’s statement is surprising as well as confusing. Android is an open source project and no one has to buy any licence to use it, as there is none.

Android is released under a permissive Apache open source license. Like any free software license, the Apache License allows the user of the software the freedom to use the software for any purpose, to distribute it, to modify it, and to distribute modified versions of the software, under the terms of the license. Which means anyone is free to take the code and use it. No one needs to get or give a license to use Android.

Since Android is not licenced in traditional terms, it’s not possible that Google could have threatened Acer with license revoke. So, Alibaba’s accusation seems to be unfounded and needs some clarification from the company.

Something did happen. What made Acer cancel the conference? Since there is no official word from Google I don’t want to assume anything. And if Google did step in, what could have been the reason?

Google released a statement to the respected blogger Danny Sullivan, “All members of the Open Handset Alliance have committed to building one Android platform and to not ship non-compatible Android devices.”

This statement shows what Google might have communicated to Acer – “to not ship non-compatible Android devices.”

However, in the same statement Google also clarified that joining OHA doesn’t meant members can’t work with competing ecosystems. In fact many major Android players such as Samsung and HTC also manufacture Windows phones.

The question arises why Google has issues with Acer manufacturing Aliyun phones when they have no issues with a Kindle fork or Nook fork? Is trademark or branding at play here? Is compatibility at play here?

Acer is a noted Android player so I am curious how will it position its Aliyun powered devices in China? Will these Aliyun devices be sold as Android devices along with Acer’s ‘Android’ devices? Will it not be too confusing for Acer users — which Android to pick? How will Acer differentiate its devices running ‘different’ Androids? Confusing?

It doesn’t seem Alibaba is making any effort to separate their fork with Android as they call their fork the ‘Android of China’. Which to me would sound like official Android localized for China.

Result? Customer confusion. Again.

The question is – can a fork use the name of the upstream project?

I can see Red Hat cringing if Oracle starts pushing OEL as Red Hat of databases.

I can see Red Hat cringing if Oracle starts pushing OEL as Red Hat of databases. Mozilla will be fussing a lot if someones starts pushing a fork of its browser as Firefox of Enterprises. Under no circumstances Oracle will allow Monty AB to ship MariaDB as MySQL for servers. I don’t see a smiling face of Mark Shuttleworth if Clement starts providing Linux Mint as Ubuntu for Gnome users.

Similarly Google may have issues with an OHA partner to fork Android and push it as Android. I can clearly see customer confusion. Google’s issue could be about brand identity or trademark.

Why is there so much fuss about compatibility?
As we noticed in the statement Google has issues with compatibility. Why is there so much fuss about compatibility?

Those who deal with Linux are aware of compatibility issues. We have seen this in the Linux world before. There are gazillions of Linux distributions and even if they use the same Linux kernel, same applications, same packages they all are incompatible with each other. You can’t just grab a binary for Linux and install it on your distro, you need that distro specific binaries.

Forget about the wider Linux community, if you look at the rpm-based distribution (Fedora, openSUSE, Mandriva or Mageia) they don’t work well with each other either. This leads to user frustration and head-aches for developers because they have to support their apps on gazillions of distributions practically using the same code-bases. There is so much duplication and wastage of resources in the Linux world.

This is also one of the reasons why Linux is not popular on desktop as there are so many incompatible distributions to support. If Adode does want to bring Photoshop to Linux, which Linux should it support? Ubuntu or Debian or Red Hat or openSUSE or Arch or Gentoo?

Google may not want it’s OHA partners to create a mess that we see in the Linux world. In order to ensure compatibility (which the Linux world is suffering from) Google had created an Open Handset Alliance. All major hardware vendors, mobile operators, semiconductor companies, software companies and other players with stake in the mobile space joined the OHA (apart from Nokia). There are 84 active members of the alliance who have committed to support Android. All these players committed to work on the code-base so that there is no incompatibility issue. I don’t know why Amazon or B&N needed to fork Android when they could have been part of OHA and benefitted from the large Android ecosystem — and at the same time helped it get better. So, it is understandable why Google may not want OHA members to create incompatible forks of Android. It just creates a big mess.

It’s easy to be Android compatible, the OHA supplies all the tools and details on how to do it.

I can sense this in the statement by Andy Rubin, the creator of Android, “As the lead developer and shepherd of the open platform, we realize that we have a responsibility to app developers — those who invested in the platform by adopting it and building applications specifically for Android. These developers each contribute to making the platform better — because when developers support a platform with their applications, the platform becomes better and more attractive to consumers. As more developers build great apps for Android, more consumers are likely to buy Android devices because of the availability of great software content (app titles like Fruit Ninja or Google Maps). As more delighted consumers adopt Android phones and tablets, it creates a larger audience for app developers to sell more apps. The result is a strategy that is good for developers (they sell more apps), good for device manufacturers (they sell more devices) and good for consumers (they get more features and innovation).”

Andy Rubin wrote on his Google Plus page:

We were surprised to read Alibaba Group’s chief strategy officer Zeng Ming’s quote “We want to be the Android of China” when in fact the Aliyun OS incorporates the Android runtime and was apparently derived from Android.

Based on our analysis of the apps available at, the platform tries to, but does not succeed in being compatible.

It’s easy to be Android compatible, the OHA supplies all the tools and details on how to do it.

Android Needs A New Name?
Danny Sullivan suggests that the Android Open Source Project should pick a new name. “As for the Android Open Source Project, it needs a new non-Android name. After all, if devices can be built off that code but can’t necessarily use the Android logo, why continue with this confusion of some (including those at Google) referring to them as Android? Pick a new name.”

I tend to disagree. There are major open source projects such as Debian, MySQL, openOffice, RHEL and many more. There are many forks or derivatives of these projects notably Ubuntu, Linux Mint, Maria DB, LibreOffice, etc. These derivatives don’t call themselves Debian, or MySQL or OpenOffice; they maintain a clarity in the name and branding.

Just because there is Ubuntu or Linux Mint doesn’t mean Debian should re-brand itself. The onus is on derivatives to create their own identity and not on the main project.

So, why should Google drop the term that it worked so hard to make a household name? On the contrary derivatives/forks should refrain from calling themselves Android.

Acer is a known Android player and by working on an incompatible/fork of Android it might create some confusion when people may buy Aliyun phones thinking they were buying Android devices only to find it did not work the way an  Android devices works.

However, Acer is not giving up on Alibaba’s offer as China is a huge market, WSJ quoted an Acer official, “Acer will continue to communicate with Google and the company still wants to launch the new smartphone based on Alibaba’s software.”

The open source development model makes it very tempting for others to take the code and create their own fork, instead of contributing to it and making it better. The strength of open source is contribution. I wish companies like Amazon, B&N or Alibaba would have worked on the same code-base and make Android better instead of creating their own forks.

Slideshow Image:

About Swapnil Bhartiya

A free software fund-a-mental-ist and Charles Bukowski fan, Swapnil also writes fiction and tries to find cracks in the paper armours of proprietary companies. Swapnil has been covering Linux and Free Software/Open Source since 2005.

One thought on “Google Doesn’t Want Acer To Fork Android?

  1. Okay, first off, the writer of this article is entirely correct about the branding situation. However they are entirely incorrect with most of their other off-topic conjecture.

    When the author says, “The open source development model makes it very tempting for others to take
    the code and create their own fork, instead of contributing to it and
    making it better. The strength of open source is contribution. I wish
    companies like Amazon, B&N or Alibaba would have worked on the same
    code-base and make Android better instead of creating their own forks.” – they show how completely ignorant they are of the Open Source development model.

    It isn’t “tempting” for others to create a fork. It’s MANDATORY. You cannot make a change to a code tree, build it with your changes, test it, prove it, and contribute your changes back, without forking it, PERIOD. The author doesn’t make any attempt to uncover whether Amazon, B&N, or Alibaba made pull requests to Google to have their improvements / changes merged with Android. If they did, they would most likely be turned down because their changes were not generally useful enough to all the people who use the Android code base.

    If Alibaba had merely localized Android and contributed that set of localizations to be merged, then yes I’m sure such changes would be accepted. However the changes performed for Kindle were primarily a custom reduction of features. Basically they reconfigured the build process to skip some things, and added their own branding and default apps. That is exactly the correct thing to do.

    The author ranted about “… so much duplication and wastage of resources in the Linux world.” Well here we go again, complete ignorance. The reason things are incompatible in the Linux world is primarily to prevent duplication and wastage of resources – one copy of glibc in use at a time is plenty thank you very much, so if you build a package against an older, incompatible version then the distro’s installer is going to refuse to unpackage it into the system, precisely to prevent wastage of resources. But if you want to you can –force it. Now on the other hand if you talk about “duplication and wastage of resources” from a human perspective, ie so many people wasting their time building these incompatible packages, that’s why there’s the Linux Standard Base (LSB) to help ensure that the lowest common denominator is available to developers and package maintainers. The Linux world is full of freedom, including the freedom to ignore the LSB, and in fact only a miniscule share of packages depend on the LSB. A default install of Debian with Gnome or KDE does not bring in LSB, for example. The reason for this is because there’s a lot of Not Invented Here syndrome in the Distro camps. Each has its own beliefs of the best way to do things, and using the LSB is not a cornerstone of those beliefs… in fact the LSB brings in pieces of other distro’s ideology into their environment and so they studiously avoid it. But having said that, the proprietary world (ie RT/iOS if we’re talking about mobile) have absolutely massive duplication of effort on the part of their developers, because they don’t share their code. So everyone has to reinvent the wheel over and over. I don’t see an inherent benefit to either system as far as efficiency is concerned. I only see that with the proprietary model there has historically been more success for developers to make money from their work, while with the open model there has been a lot of excellent opportunities for education and less need to reinvent that wheel… but it hasn’t stopped projects like KDE and Gnome from discarding their entire code tree at each major iteration, starting over from scratch with a completely unstable mess, and then overzealous distro maintainers trying to get glory by being early adopters and subjecting their users to such a mess that it made Linus swear at Fedora and Gnome, lol.

    Alibaba proabably took their fork too far before attempting to sync it with upstream. The phrase “We want to be the Android of China” says it all. They should have aspired to be the company who Brings Android to China, but they’re merchants… they have a mentality to eliminate the middleman… they don’t seem to recognise that they are that middleman, and have successfully cut themselves out by not playing ball.

Leave a Reply