Re: GConf vs. bonobo-config

[Dietmar Maurer <dietmar ximian com>]

Soeren Sandmann wrote:

> Does not require        yes     no              yes
> reimplementation of
> large piece of code

This is a two-sided argument. Yes, you do not need to reimplement the code for
the wrapper. On the other side I consider the implementation of GConf as
unnecessary complex and clumsy. This is due to the fact that GConf was written
2 year ago, at a time when we had not working bonobo/Orbit environment.

Havoc already pointed out that GConf has some nice features, like connection
recovery after a crash. The implementation of that feature is quite complex,
and it is a problem we have to solve. But GConf is not the only application
which needs such features, and I would be ways better to have a generic
solution to the problem. It is then possible to reuse that code somewhere else.

Another example is the event notification. Havoc claimed that his
implementation has a better performance, and is more scalable. Well, wouldn't
it be nice to share that code so that everyone can use his high-performance,
super scalable EventSource.

I consider bonobo-config as a chance to cleanup the GConf implementation and
simplify the internals. But it seems that most people think this is worthless,
because Havoc is maintaining his own code anyway. They don't see that code
sharing is a real advantage, and that we can gain much from a clean
implementation. Instead they assume I am totally braindead, sitting here and
unnecessarily duplicate code - a quite ignorant view of things :-(

-Dietmar