The Matt Porter's Android Mythbuster
Android is plateform that I strongly support as a business and I believe in it (but I'm not a fan yet of the current UI's). Gameloft is not really agree with that and it's perfectly understandable as Google is playing the "I've done my part of the job, now it's your (the dev) turn" game and seems not having done enough to “entice customers to actually buy products” on the google's marketplace. The lack of a desktop store (à la iTunes, Do you think it can be made by a third party ?) is one of many example (even though most of the App on the iPhone are bought directly from the phone itself).This talk examines the myths and realities of the Android Operating System. Android has achieved an amazing level of anticipation and media coverage largely due to Google's backing. However, much of the technical and mainstream media coverage has led people to misunderstand the technology behind Android. In this session, we will explore the Android Open Source Project (AOSP), examine the AOSP developer community, and learn why Android is not "Linux". The discussion will cover the challenges involved in developing Android- based devices and porting Android to the MIPS architecture.
Matt Porter started contributing to Linux as an early Debian GNU/Linux developer. He has developed and maintained support for embedded systems in the mainline kernel including PowerPC VME/cPCI systems, PowerPC 4xx, and the RapidIO subsystem.
Matt led the Linux effort at Motorola Computer Group and later joined MontaVista Software as the founding member of the Arizona engineering center. He joined with other embedded Linux veterans to found Embedded Alley and now works on embedded Linux and Android product architecture.
And as Harald Welte says in his blog about Matt Porter and the presentation:
As you may know, Matt Porter was heavily involved in the MIPS and PPC ports of Android, so he and his team have seen the lowest levels of Android, more and deeper than even cellphone manufacturers ever have to look into it.
The presentation shows how Google has simply thrown 5-10 years of Linux userspace evolution into the trashcan and re-implemented it partially for no reason. Things like hard-coded device lists/permissions in object code rather than config files, the lack of support for hot-plugging devices (udev), the lack of kernel headers. A libc that throws away System V IPC that every unix/Linux software developer takes for granted. The lack of complete POSIX threads. I could continue this list, but hey, you should read those slides. now!
0 comments
Leave a comment...
