Re: [Unreal-users] Planned features for 3.3?

[tabris <ta...@ta...>]

On Sunday 19 June 2005 11:29 am, Bram Matthys (Syzop) wrote:
> tabris wrote:
> > 	What is the planned feature list for 3.3? The FAQ doesn't list it,
> > and there's nothing in google either.
>
> Yeah, I've though about publishing it, but.. I also don't want to put
> all kinds of ideas online that others can "grab" before we actually
> implemented it ;). Another reason is I (we?) don't want to expect
> everyone to have feature X implemented that is on the list. Some
> things might not be done.
	An idea that I'd like to see:
1) client-level zlib compression. I admit it would chew CPU, but it=20
doesn't have to be over level 6 to be effective. many would just go for=20
level 3 or so.
I also realize that it would require client-support, and that many=20
client authors do not give a damn.
2) More flexible per-channel message filtering. chmode +G isn't that=20
effective, and kick-for-swear scripts are rather pointless. By the time=20
someone has sweared, it's too late.
>
> In any case, among them are:
> - Completely different I/O engine, one that deals with
> kqueue/epoll/etc. While tests show that the performance advantage
> isn't as big as one would expect, it will be cleaner code (which
> means less bugs, easier to extend, etc.). - Besides the above,
> several other subsystems will be completely replaced and/or
> rewritten. Others will be highly improved (speed). Basically in all
> cases it would result in cleaner code, and lower cpu usage.
> - Much more will be modulized. This is especially for "hot swap"
> ("hotfix") purposes (patch the ircd for a vulnerability or normal
> bug, without requiring a restart).
> - Numerous new features (web interface, distributed spamfilter,
> internationalization of numerics/messages, and more)
Web interface I'm ambivalent on. I have too much problems with idiotic=20
admins already, this will just lower the bar and allow in new levels of=20
idiots.
Distributed spamfilter. what's that? we already have them in the files,=20
and we have them in TKL F.
I18N sounds good, tho I think at some point we might need to discuss a=20
new client-protocol for this. Unicode anyone?
> - Some of these things might be discussed on the forums to gather
> opinions (such as more customized oper levels, etc.)
Point. I just wish it would happen more on the ML. The web is good. it's=20
just not so good for many-to-many communication.
>
> The main goals (in this order) for 3.3* are:
> - Cleaner code
> - Speedup
> - New features
>
> So yes, new features isn't #1. Still, there will be many (major) new
> features of course ;). It's just that we will concentrate much more
> than usual on cleaning up the code and optimizing things, which is a
> good idea IMHO.
>
> > With 18 months to do the development, anything is possible, right?
>
> One would say so.
> Still, it depends on how many "anythings" there must be done.. ;).
>
> 	Syzop.

=2D-=20
Colvard's Logical Premises:
	All probabilities are 50%.
	Either a thing will happen or it won't.

Colvard's Unconscionable Commentary:
	This is especially true when dealing with someone you're attracted to.

Grelb's Commentary:
	Likelihoods, however, are 90% against you.