sarnold | hmm I always thought dax was -r +r periodically as dax was able to deal with potential noise from it | 19:58 |
---|---|---|
sarnold | the data I've seen suggested that +r drastically reduces the channels use, but perhaps that's more due to decline of the ubuntu community itself | 19:59 |
el | yep -r should only happen if someone is absolutely 100% able to supervise it | 20:07 |
el | that's just how the internet is now | 20:07 |
hggdh | so, my take on it: if one ops available (as el points above) and willing, then set -r & monitor. At the end of availability, set +r. | 20:33 |
sarnold | that sounds good to me | 20:34 |
hggdh | a priori, our take is -- given internet & trolls -- +r is the *default* state for #u | 20:34 |
el | yeah and not too long a gap or you'll have the usual trollish people putting in sleeper clients | 20:34 |
hggdh | ^aye aye | 20:35 |
hggdh | sarnold, tomreyn: so go and experiment and keep at least one eye on the channel :-) | 20:36 |
sarnold | it's that last part that's a bit harder to do :) meetings come and go, etc.. so I haven't flipped it yet myself, even though it was suggested to me a week back or something | 20:37 |
hggdh | yeah, this is the rub. But if two or more commit to looking at #u frequently, the burden might be shared. I try to every time I can, but real work interferes... | 20:38 |
hggdh | even worse right now due to power outages in TX (mostly caused by brain outages in government, but this is besides the point) | 20:40 |
el | outages imply the existance of the service prior. | 20:57 |
hggdh | heh | 21:03 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!