LEGO wrote:
Hi,
I believe than in very few weeks I'll release an almost complete
and usable version of MATE. As so, I'd like you to consider whether or
not would it be appropriate to apply the very small changes that I've
made to packet-frame.c in order to allow mate's to run if the users
install the plug-in.
Being MATE a plug-in, it can be installed afterwards by those who'd like
to try it. If it not installed there would be no impact on stability
and the impact
on the overall performance of ethereal would be minimal.
I think it would be much useful if users could start getting their
hands on it. I've being making some changes to mate based on the
feedback I've got from this list that I believe make MATE a very
useful tool to assist network troubleshooting.
To my experience, we might better get some *developer hands* on your new
feature first.
A good way might be:
- wait until next release is out and a few addtional days as the release
might be buggy for various reasons
- check feature into subversion, so the developers can have a look and
play a bit
- there might be some change requests and documentation issues coming up
which should be cleaned up then
- the next release will include the changes in a somewhat (more?) stable way
I would suggest doing it that way, as it's not a good idea to hurry up
things here (although you may like to see it in the next release, I
understand).
If things done too fast, you may:
- get a lot of user questions because of bugs and bad documentation
- get a buggy Ethereal release :-(
That's my experience about "best practice" of adding new features,
Regards, ULFL
