>I have a similar problem with the generic handling of the modules (Action,
Screen and friends). I guess this is what annotations were invented for in
the first place. My suggestion would be to do only the things absolutely
necessary to get the milestone 1 out of the door and postpone the "real"
solution to >milestone 2 where the security stuff is about to be replaced
anyway. Rather put your energy into tuning the fulcrum-security stuff. LDAP
needs some testing, for example.

The thing is...
Like I said in my introduction when i became a Turbine comitter, I run a
company where we develop a web-app that is now 2 years old based on turbine.
I intend to test Turbine4 through our web-app. If we don't generify turbines
security model before the release of M1 I would have to change a lot in our
code just to make it compile. I'm not sure that I want to put my time there.
I'd rather develop turbine one step further to test it. And as you say, I'd
probably focus on fulcrum security. I have no experience of LDAP however and
I would probably but all my energy into torque.

>> There is also a "trick" that I found in the book Effective Java by 
>> Joshua Bloch (which I very much recommend btw).
>> 
>> You can create a general method that looks like this:
>> 
>>     public <U> UserManager<U> getUserManager(Class<U> type)

> I like this one. That's how ArrayList.toArray(T[]) works. Maybe this can
help me with the AssemblerBrokerService.

If the team feels that this is a good solution I will start working on it as
fast as I can. 

If you still want to release M1 my vote will be [0] and I don't think I will
be giving feedback this round. I will on the next though.
/Ludwig


---------------------------------------------------------------------
To unsubscribe, e-mail: [email protected]
For additional commands, e-mail: [email protected]

Reply via email to