Operating Systems with Chinese characteristics – revisited

HTC is reportedly working on a new operating system (OS), one that is specifically tailored to a Chinese audience. This is a smart strategic move for them, software is they key value component of smartphones now.  Beyond that I  will not comment on their prospects. All I know is what has appeared in the blogs, and I do not want to delve into a specific OEM’s plans.

That being said, there is already a large number of similar efforts underway in China. Many companies are trying to build their own OS. For their part Apple has already integrated their iOS with many of the leading Chinese web properties, for example one-touch posts to Weibo instead of Twitter and the ability to set Baidu as default search rather than Google.

Most of the Chinese language OS to date have in reality just been re-skinned versions of Android. OEMs and their software partners have taken the completely free version of the Android kernel, which is available for download by all, and inserted their own user interface layer. Underneath the hood, they insert links to their own services or those of partners.

There are a few difficulties with this strategy:

  1.       Building an operating system from scratch is very expensive. Re-skinning Android is less so. But even more expensive than building the OS is maintaining it. Operating systems need to be updated and advanced. This work is not well rewarded, it is essentially just table stakes for being in the business.
  2.    An OS is a ‘platform’, meaning that other things are built on top of it. Mostly, that means apps, and for apps to be interesting they need to come from lots of developers. There are no clear rules here, but generally speaking 10,000 apps on an OS is a good place to start, but critical mass is probably closer to 100,000. No single OS vendor can build that many apps on their own. So the platform has to appeal to a lot of developers. Attracting those developers is a big complex process, about which we have written many times before. For the purpose of this post, suffice it to say this is not easy.
  3.    An OS needs hardware partners.  Microsoft had dozens of hardware partners, and the cost of building the OS could be amortized across all of them. Building on OS for a single hardware line is exceedingly rare. Apple being the exception that proves the rule. Google can do this because they have another, large revenue base to tap into that in turn feds of the OS. And they are also the exception, even Facebook with its billion users has not  been able to launch their own OS, and probably will never try.

So far, all of the China-based attempts to build an OS have stumbled on one or more of these. Even China Mobile, which has more subscribers than the US has citizens could not generate critical mass for its Android variant.

For their part, Google does not see worried by these clones. They know that they can keep updating Android. The clones then have to build their own updates. This process ensures the clones will always be a step behind.

That all being said, I still think there is room for more operating systems in the world. The odds do not favor any of the contenders, but the smartphone market is large and there are still some huge markets with almost no smartphones. So maybe, just maybe there is room for another OS, or two.

4 responses to “Operating Systems with Chinese characteristics – revisited

  1. IF you were willing to do the work to not just provide a China-specific OS at the UI level, BUT also throughout the entire OS, the game changes.
    Specifically suppose that you were willing to also create a Chinese based programming language and Chinese named versions of every API and SPI for the system?

    The first step in this is, of course, just having your OS (including EVERY dev tool) understand Unicode, not just ASCII. So that means not just your editor and your compiler, but your linker and loader, your debug tools, etc. Likewise your file system and partitioning tools (and the specialized file system exploration and repair tools). Likewise for whatever database type technology your OS is built on (eg MS’ Registry).
    This MAY be the case today for OSX/Win/Linux/Android, but it would not surprise is holes still remain in any of these OSs for a few edge cases.

    Once this is done, you now need a Chinese-based programming language. In theory, of course, you can simply translate every keyword of Java or Objective C to its Chinese equivalent, but I don’t know how well that would work — it may result in something that seems more artificial than Java and C feel because the word ordering is wrong. Likewise the need to translate every API/SPI to Chinese.

    IF you did all this, and did it well, you now have something pretty remarkable — a tool that feels a lot more pleasant/natural to use for 1 billion or so potential programmers, than anything available today… Throw in the patriotism factor, and I think it’s trivial for you to get to your 100,000 programs.

    It’s not all easy, of course. There’s a substantial risk here that this theory (the desirability of a Chinese language dev environment) is lower than I imagine because everyone with the skill to program in China wants to understand the Western dev systems so they have a hope of working for a Western company and even, one day, leaving China. I have no idea the extent to which this is true.
    This risk could be mitigated by partnering with the Chinese Government, which I think would be quite willing to bankroll this exercise. But that brings along the risk that this same government will demand even more backdoors and security loopholes in the OS than they do already, and that in time this will lead to widespread disaster via banking fraud, leaked emails and so on. (cf the godawful mess that is banking in South Korea, and the consequences of that we have already seen.)

    But if I were in the business of trying to keep my company alive, and desperate for alternatives, this is the route I would go — pitch this vision of a “Chinese” OS to the Chinese government (based essentially on Android/Linux, but with localization driven all the way down to the programming language and the APIs, and maybe even with the promise to “translate” all the current source code into the new Chinese based computer language) and get them to fund it.

    • Maynard
      Congratulations. You win the award for first comment on my blog that is longer than the original post.
      No mean achievement.
      You make a lot of sense. It would take a lot of work to do well, but its not an impossible task. And could make strategic sense along many angles.

  2. Pingback: ”No one at Google is Returning Our Calls” – DIGITS to DOLLARS·

  3. Pingback: Mobile OS Alternatives: Today is not Saint Jude’s day – DIGITS to DOLLARS·

Leave a Reply