[Embedlets-dev] RE:[Arch]New version 1.3 of the Architecture Discussion Document posted in CVS...

[Ted K. <tk...@ya...>]

James stated:

>So, in summary we take the XML Embedlet Configuration (
>Or wiring model) and autogenerate the code to build the 
>embedlet application this is 'included' in the build and 
>called by the container to construct the application, ie
>Instantiate the Embedlet tree from the XML into a 
>initialisation code. Ie Create a verbose code version of 
>serialisation? - pseudo Serialisation.

One thing that I have been wrestling with since I started experimenting with
the Elevator challenge was which of the following scenarios made more sense:

A) The Embedlet Container for any given device would be pre-installed on that
device and then the Embedlet application would be deployed into it as a
separate operation.

B) The workstation that was being used to build the application would start
with a default minimal functionality Embedlet Container, have JAPL peripherals
and Container Services plugged into it as needed, have the Embedlet application
assembled and installed into the container, have the system as a whole Unit
tested and then have the complete system (Embedlet container and all) packaged
into the target system's proprietary deployment format and then sent to the
target device.

For a TINI, the deployment file is called a .tini file and it contains an
application's compressed .class files.  The way that a .tini file is deployed
to a TINI is through FTP and one way to execute the application is through
telnet.


So, is the Embedlet Container pre-installed on the target device or does it get
installed with the application?  Are there compelling reasons to allow for both
scenarios?


Ted

__________________________________________________
Do you Yahoo!?
Yahoo! Mail Plus - Powerful. Affordable. Sign up now.
http://mailplus.yahoo.com