Hi Hans,

> The main raison d'etre of TSIK was always simplicity.  As it is now,
> TSIK runtime has only one single dependency, and that is on the 
> ubiquitous log4j. 
> 
> The freedom of a single jar, of a single download/compile to get
> XML/SOAP security related stuff going, without worrying about this 
> and that other jar is very refreshing.  It makes it difficult for 
> the developer to screw up inadvertently.
> 
> Again, simplicity is key, both for developers and deployers.  To 
> the greatest extent possible, I'd hate to lose that "instant 
> usability" of TSIK.  
> 
> Also, since TSIK is still in incubation, I don't know whether it 
> is prudent to depend on it too much. Could this not be a show-
> stopper?

I'd put it the other way .. in order to graduate from incubation, TSIK
needs to become a part of the Apache WS community of projects. In order
to do that, it needs to remove most of the overlapping coverage and
start becoming a piece of the puzzle, instead of a single holistic box. 

I fully understand and appreciate the simplicity argument, but there's
no way to build a community around TSIK without it becoming part of the
community of projects.

> Dims, do you have a simple sketch/outline how you see the various 
> wss4j/tsik packages broken out and working together?  That would 
> definitely help me grok your plan.

I'd say it should be able to deploy TSIK to get WS-Sec functionality for
Axis, at the bare minimum.

Sanjiva.



---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to