On Thu, 21 Jun 2001 02:50, Stephen McConnell wrote:
> Pete ... this is the first of two points I would appreciate additional
> information on. Can you summarise the specific reasons why Log4j is
> unsuitable for our use?
Unsecure design is a sufficient enough reason in my opinion, added complexity
is another good reason.
> > > What I would like to see in the near future is a Logger
> > > interface that both projects can aggree to so that people who develop
> > > in Avalon can use their logger of choice. If JDK 1.4 supplies such an
> > > interface it might be worth investigating.
> >
> > Never happen - not technically viable.
>
> If this is really the case, then what we are saying is that anyone using
> Avalon *must* live with possibly multiple logging frameworks
yep. Just like if you use JNDI you live with multiple Component directorys,
or if you use properties you live with multiple flat property inputs, or if
you use JDOM you work with multiple configuration formats etc
> and that it
> is not possible to abstract out a common interface.
given sufficient thrust pigs fly. Possibility and viability are different -
it may be possible but it adds no value - see below.
> Which leads to my
> second request .. can you summarise the reasons behind your "not
> technically viable" conclusion?
See archives of ant-dev. Essentially it comes down to - What features do we
need? And inevitably - Whats the simplest interface to provide them?
Essentially the features needed are exactly the Logger class (minus one
method - getCategory).
Now you could create an interface with all those methods. Now you create
wrapper interfaces for Appenders/LogTargets, maybe a instance of
AbstractFactory pattern, interfaces for Filters etc. This wrapper system is
now more complex than LogKit, Log4j or Logging JSR. It also adds zero value
as you have just created a *new* interface/API that offers nothing better
than already existing.
Cheers,
Pete
*-----------------------------------------------------*
| "Faced with the choice between changing one's mind, |
| and proving that there is no need to do so - almost |
| everyone gets busy on the proof." |
| - John Kenneth Galbraith |
*-----------------------------------------------------*
---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]
