I would love to hear what you think about Qualcomm's BREW, (if Ramsey has not already asked you.) I find the BREW message difficult to understand, and I just don't see how BREW can become the Windows of cell phones as Paul Jacobs called it.
Kahn, I don’t know enough about Brew to comment deeply. At a specific level, it reminds me of how eSim uses a specialized kernel sitting on top any OS to convert a software model (finite state machine) to a simulator on the one hand or full-blown application system on the other. It also reminds me of Java. But I doubt it has any of the robustness of a programming language and execution environment like Java.
Should it become popular, then without doubt it is not good for Palm or Microsoft, because it defeats the main thing they bring to the party – applications compatible with their OS only. Thus, as wireless device functionality increases, with Brew there would be no reason to require an OS with more application features than any traditional RTOS. This would open the door even wider for WIND, since VxWorks and support tools and development environment has everything else that might be wanted, including portability with regard to all popular hardware platforms, hardware and software debugging, protocol stacks, etc. etc.
Even so, I am not entirely comfortable with the notion of Brew. I would rather see standard Java utilized more directly (i.e. implement the application support API in compiled code accessible by Java). Better yet, I still think an ASP-based approach, which may or may not extend itself with Java, is preferred in the wireless handheld space. However, I readily admit that my lack of comfort may simply reflect lack of detail about Brew and the hundred or so questions that come immediately to mind.
Allen |
