> Why not provide two different impls?
> 
> Same API, different Bundle-RequiredExecutionEnvironment.
> 
> With no difference in the external APIs you're not breaking the 
> portability of user's code.
> 
> Obviously concurrency is far less of a problem in embedded. But on 
> the server side actually the opposite is true?
> 
> Maintenance is certainly a hassle, 

I think you answered the opening question right here. We also have the 
issue of being able to test both impls.

Since the tracker package is Apache licensed, people can tweak as they see 
fit. But since the package is normally packaged with the framework

> but it's an issue where serving 
> two masters means neither can be completely satisfied.

Until someone can demonstrate a significant improvement in real work 
usage, I don't see that the hassle of this is worth the effort. Especially 
when problems like your can be better solved by using your own 
customizer/subclass to use your own collection to hold the tracked objects 
with even better performance for your needs.
-- 

BJ Hargrave
Senior Technical Staff Member, IBM
OSGi Fellow and CTO of the OSGi Alliance
[email protected]

office: +1 386 848 1781
mobile: +1 386 848 3788
_______________________________________________
OSGi Developer Mail List
[email protected]
https://mail.osgi.org/mailman/listinfo/osgi-dev

Reply via email to