> <quote> > It's an interesting idea--but it requires that the bot identify itself. > I highly doubt we could convince botnet owners to make their bots > identify themselves as bots. It's unfortunate--it would be such a > useful thing :( > </quote> > > > I don't see why we couldn't convince them... If you have nothing to hide > and are not abusing, you shouldn't fear this feature right? If you don't > use it, you are a potential kill target, and servers might be able to > refuse bot connections from unknown bots. That should be enough reason to > use this feature I think!
My point is that abusers never pay attention to the rules anyway, so it wouldn't have any effect on bot nets--status quo would be preserved. (See a botnet, it's probably flooding someone, G-line it bot by bot--no benefit.) As for the other features, sure, we may be able to get legitimate bots to use the feature by making the advantages juicy enough, but any restrictions are going to make some bot owners decide not to use them. Take the PRIVMSG restriction, for example--many bots respond to commands issued on the channel. On one channel I'm on, we have a bot that'll report the weather. Many game channels have bots that send questions to the channel, then report the results for each round. If we make that part of the feature set, we either have to permit PRIVMSG to channels the bot(s) are on or find that those bot runners will be unable to run their bots with this feature. But every exception like this that we set up is going to make the ircd code that much more complicated and more difficult to maintain. Now these are just the thoughts that I have. I have no stance on this; if someone submits a patch, we'll take a look at it--probably have to run it by -admins because of the extent and type of the change--then accept it. I point out that .11 is still in feature freeze, however, so it would not be accepted for this release. -- Kevin L. Mitchell <[EMAIL PROTECTED]>
