Danny Angus wrote:

>>Currently we're stuck with this problem, so we have to pick a global
>>compromise between verbosity and lose of critical information.  Until this
>>problem can be fixed, at the least exceptions thrown by a mailet should be
>>considered a warning, if not an error.  True, some of them
>>aren't, but we're
>>still stuck with a compromise, and better to assume that an exception is
>>important than to discard it.
>>    
>>
>
>As far as I'm concerned the logging of mailets should be an issue for mailet
>developers, developers are free to use commons logging or anything else in
>their mailets, as is the case in the servlet API.
>  
>

Danny:

Have to jump in and disagree with you on this (partially :-) ).  The 
issue of logging is firtly a management issue - not a developer issue. 
 Logging information crossing boundaries of multiple application and 
frameworks, coordinated under a single management context is absolutely 
important.  Take for example the execution of James, some mailest, an 
ORB, and throw in a few Servlets - if the logging management is not 
unified it becomes impossible to seperate operational management from 
the developer maindset.  Today I can run all of the above using a single 
logging strategy and keep logging tools and targets seperate from the 
individual implementations.

Will this be true for Mailets?

Cheers, Steve.

>d.
>
>
>--
>To unsubscribe, e-mail:   <mailto:[EMAIL PROTECTED]>
>For additional commands, e-mail: <mailto:[EMAIL PROTECTED]>
>
>
>
>  
>

-- 

Stephen J. McConnell

OSM SARL
digital products for a global economy
mailto:[EMAIL PROTECTED]
http://www.osm.net




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

Reply via email to