It is easy to build a toy Android on another architecture, but can you make it production worthy?
With OpenSource Freedom Comes Responsibility
The Android codebase is 2+GByte of source code, and the many man years of effort that has gone into Android is not just writing the code, but also validating it on real devices using the ARM architecture. This validation effort is backed up by the major handset OEMs and their years of experience of brining handsets into production. If you take a complex codebase such as Android, and then port it to another architecture, it is relatively easy to get it to a demo state, but to validate it and make it production worthy is a completely different challenge. Do you want to take the risk of using code that has never shipped on a production device?
Be careful and ensure that your Android project does not turn into a monster.
Plan for the Future, Not for the Past
When you commit to an OpenSource platform such as Android you are not just benefiting from the work that has already been done, but the work that will be done in the future. Already there are two more versions of Android to be released this year: Donut and Éclair and Flan has been announced for next year. Of course there is no public roadmap for Android, but in addition to bringing great user experience enhancements, I’m sure you are going to see further ARM optimizations in these, or future releases, of Android. This has already started with the release of the ARM Native Development Kit that allows developers to write native ARM applications. Further ARM optimizations could be taking advantage of ARM Cortex Thumb 2 technology for Webkit optimizations, or it could be a wide range of features that are discussed on the Android platform groups - for example a Dalvik JIT. So if you are planning to take advantage of the latest and greatest Android releases on another architecture who will port all this ARM optimized code for you and ensure application compatibility?
Golden Rules of Using OpenSource
If you are planning to use a large OpenSource codebase for your project, I would suggest that you follow the following rules:
- Ensure the codebase supported by major players such as Google, Mozilla Foundation or Tier 1 OEM’s?
- PowerPoint’s and hacked demos are cheap; taking the codebase into production, on your processor architecture is the real proof point.
- Follow the future; make sure you can follow the main branch of the code, and not end up with the OpenSource equivalent of the duck- billed platypus.
- Don’t reinvent the wheel; the whole point of OpenSource is to take advantage of the community’s work, so use the architecture that the community is working on.
By following the above rules you can ensure that your OpenSource project stays that way and does not become ClosedSource were you lose the support of the OpenSource community.
What Golden Rules of OpenSource would you add?
More info on Android on ARM can be found here.
James Bruce, Mobile Marketing, ARM, is based in Silicon Valley. James is without doubt a gadget guy who is continuously looking at the latest devices and services on them. Working for ARM allows James to see what technology will be on your mobile device in 3 to 5 years time. This view of the future combined with being based in Silicon Valley and having worked on mobile for the last 9 years allows James to have a unique view of mobile technology. At the moment James is deciding which Cortex-A8 phone he will buy this year, and which dual core Cortex-A9 phone he will buy next year.
Shortlink to this post: http://tinyurl.com/l479a6
0 Comments On This Entry
Please log in above to add a comment or register for an account
Fortune Brainstorm Green
on May 13 2013 10:58 AM
Moonshot - a shot in the ARM for the 21st century data center
on Apr 09 2013 01:22 PM
Bringing the Benefits of the Smartphone to Pay-TV
on Mar 14 2013 05:34 PM
2013 - A Lucky Year For All Smartphone Consumers
on Mar 13 2013 06:58 PM
Internet of You at Mobile World Congress with M2M, Sensors and LTE
on Mar 12 2013 02:44 PM