You are viewing a plain text version of this content. The canonical link for it is here.
Posted to dev@commons.apache.org by Ceki Gülcü <ce...@qos.ch> on 2002/10/15 19:40:19 UTC

Re: [Logging] How can I submit a concrete Log implementation (WLSLogger)?

At 10:31 15.10.2002 -0700, Costin Manolache wrote:
>Maybe we can do it at a later stage or in a LogFactory2 - so
>the code would work even if discovery is not available/used.
>I think the current hardcoded discovery, even if duplicated and
>not as complete as commons-discovery - does work reasonably well.
>
>--
>Costin

I happen to think that the commons-logging discovery method is too
smart for its own good but I am biased. :-)

--
Ceki

TCP implementations will follow a general principle of robustness: be
conservative in what you do, be liberal in what you accept from
others. -- Jon Postel, RFC 793



--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>


Re: [Logging] How can I submit a concrete Log implementation (WLSLogger)?

Posted by Ceki Gülcü <ce...@qos.ch>.
At 10:55 15.10.2002 -0700, you wrote:
>Ceki Gülcü wrote:
>
> > At 10:31 15.10.2002 -0700, Costin Manolache wrote:
> >>Maybe we can do it at a later stage or in a LogFactory2 - so
> >>the code would work even if discovery is not available/used.
> >>I think the current hardcoded discovery, even if duplicated and
> >>not as complete as commons-discovery - does work reasonably well.
> >>
> >>--
> >>Costin
> >
> > I happen to think that the commons-logging discovery method is too
> > smart for its own good but I am biased. :-)
>
>As long as it accepts that the user is smarter ( and accept explicit config
>overrides ) I think it's nice to be smart.

True, allowing the user to override automatic "smart" behavior is much
better than just assuming that the code always knows better. However,
forcing the user to manually choose an implementation would have been
both simpler and safer.

Logging is about identifying problems. It follows that the logging
component should be reliable, that means simplicity and predictability.
For the novice user, c-l may seem to behave at random.

Anyway, this is way off the WLSLogger topic.

>--
>Costin

--
Ceki

TCP implementations will follow a general principle of robustness: be
conservative in what you do, be liberal in what you accept from
others. -- Jon Postel, RFC 793



--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>


Re: [Logging] How can I submit a concrete Log implementation (WLSLogger)?

Posted by Costin Manolache <cm...@yahoo.com>.
Ceki Gülcü wrote:

> At 10:31 15.10.2002 -0700, Costin Manolache wrote:
>>Maybe we can do it at a later stage or in a LogFactory2 - so
>>the code would work even if discovery is not available/used.
>>I think the current hardcoded discovery, even if duplicated and
>>not as complete as commons-discovery - does work reasonably well.
>>
>>--
>>Costin
> 
> I happen to think that the commons-logging discovery method is too
> smart for its own good but I am biased. :-)

As long as it accepts that the user is smarter ( and accept explicit config 
overrides ) I think it's nice to be smart.  

-- 
Costin



--
To unsubscribe, e-mail:   <ma...@jakarta.apache.org>
For additional commands, e-mail: <ma...@jakarta.apache.org>