Short-Term Plans

The following are my short term plans about how to evolve µCode:

Provide security

Right now, µCode does not provide any feature that is especially designed to provide security, although at least I strived to achieve a design that tries not to be inherently insecure. The simple defininition of the unit of mobility---the group, and the fact that any other mobility primitive can be implemented on top of it should facilitate the task of providing a secure version of µCode. However, security of mobile code and mobile agents is a big topic per se, and a one that is not among my immediate and main interests. I would love to see some of the guys working in this field to use and hack µCode for experimenting with their schemes and mechanisms. Any takers?

Experiences with PDAs, phones, and other resource-constrained devices

The small size of µCode is one of its assets, that make it amenable for use on resource-constrained platforms. A student of mine just started investigating how to port µCode on Symbian's EPOC and other platforms. 

Take performance measurements

There is a lack of studies characterizing the performance of mobile code systems (and mobile code applications, for that matter). Thus, I would like to get some data, and hopefully make some comparison, between µCode and other systems.

Port existing packages and applications to the new version

The packages providing resource bindings and communication need to be adapted in order to run with this last release. Same holds for a number of projects developed by my former students, that will constitute a new set of demos for µCode. 

Minor improvements

There a number of nitty-gritty things I would like to take care of, mostly small improvements here and there in the way things are carried out. Among these, I have to complete the implementation of synchronous operations.