Meeting in #wikipedia-en-accounts-meeting on August 10, 2013.

acc_meeting_2013-08-10.txt

[2013-08-10 19:28:27] <Izmeetingz> -Now logging starts-
[2013-08-10 19:28:37] * JohnLewis waits for the irony to settle in for rschen7754 :P
[2013-08-10 19:28:58] <harlot> Fast dinner
[2013-08-10 19:29:01] <Izmeetingz> So this meeting is surrounding [Accounts-enwiki-l] Rechecking and acting on requests in current CU-needed list
[2013-08-10 19:29:42] <JohnLewis> ALl but rschen7754 will known of that email.
[2013-08-10 19:29:56] <JohnLewis> *all *know
[2013-08-10 19:30:16] <Izmeetingz> As CUs have been noticing, the CU backlog is getting insane, and is often a deturant (SP?) to actually clearing it out.
[2013-08-10 19:30:25] <Izmeetingz> rschen7754: get ACC :P
[2013-08-10 19:30:31] =-= rschen7754 is now known as rschen7754|boat
[2013-08-10 19:30:33] <Izmeetingz> anyway
[2013-08-10 19:30:40] <JohnLewis> Boat?
[2013-08-10 19:31:22] <Izmeetingz> The current procedures limit users from mistakenly sending through requests that should be checked
[2013-08-10 19:31:46] <Izmeetingz> and we are constantly getting requests where it's either inconclusive or safe to go anyway.
[2013-08-10 19:32:04] <Izmeetingz> These appear mostly from Vandalism and {{anonblocks}}
[2013-08-10 19:32:28] <JohnLewis> Yep.
[2013-08-10 19:32:40] <Izmeetingz> CU data in most cases is not any good on these types of blocks more than 2 weeks, unless it is a static IP
[2013-08-10 19:32:51] <JohnLewis> Izmeetingz: In rare cases, people also report non existent blocks.
[2013-08-10 19:33:09] <Izmeetingz> JohnLewis: I think those are mostly expired rangeblocks
[2013-08-10 19:33:11] <FunPika> JohnLewis: Non existent, or expired by the time they were actually looked at?
[2013-08-10 19:33:42] <JohnLewis> A few probably but there have been cases were no block was ever made.
[2013-08-10 19:34:57] <Izmeetingz> So, other than anon/vandalism blocks, are there any other block types that you guys think don't need to be deferred to CU?
[2013-08-10 19:35:37] <JohnLewis> I'd say proxy. We need a seperate queue or have them left in the normal queue.
[2013-08-10 19:35:39] <harlot> That's pretty much all schoolblocks
[2013-08-10 19:35:59] <harlot> And even school rangeblocks
[2013-08-10 19:36:08] <FunPika> proxy blocks, they should go into their own queue (https://github.com/enwikipedia-acc/waca/issues/43). I added a comment noting what would have to be added to the tool's configuration to add a new queue
[2013-08-10 19:36:13] <Izmeetingz> JohnLewis: yep, i'm going to be working with FunPika for a seperate queue
[2013-08-10 19:36:15] <JohnLewis> Proxy mainly because either a) Other get to them before CheckUsers or b) CheckUsers defer back as no check needed
[2013-08-10 19:36:25] <Mlpearc> or they're a school or single IP, I've deferred back a few of those
[2013-08-10 19:36:29] <Mlpearc> lol
[2013-08-10 19:36:34] <Mlpearc> <lag>
[2013-08-10 19:36:41] <JohnLewis> Mlpearc: :P
[2013-08-10 19:37:00] <Mlpearc> I'm all for the proxy section on the interface
[2013-08-10 19:37:06] <Mlpearc> sounds good
[2013-08-10 19:37:51] <JohnLewis> With that, we also need proxy as a default close/quick link. I get sick of c&ping from the guide >.<
[2013-08-10 19:38:47] <Izmeetingz> So Proposal 1: All vandalism/{{schoolblocks}}/anonblocks should only be deferred if they are under a week since the block. If people feel the IP is static, they can defer to CUs
[2013-08-10 19:39:13] <Izmeetingz> JohnLewis: send it to github issues. :P
[2013-08-10 19:39:37] <FunPika> https://github.com/enwikipedia-acc/waca/issues/11
[2013-08-10 19:39:40] <FunPika> already there :P
[2013-08-10 19:39:45] <Izmeetingz> Comments on Proposal 1?
[2013-08-10 19:39:48] <JohnLewis> Izmeetingz: Probably will :P I I could be bothered atm and weren't busy, I'd use my commit access to do stuff :p
[2013-08-10 19:39:52] <JohnLewis> *if I
[2013-08-10 19:39:57] <FunPika> or more specifically the task of me trying to make it so tool admins can add whatever close buttons they want
[2013-08-10 19:40:45] <Izmeetingz> Ok, better question, anyone object to proposal 1?
[2013-08-10 19:41:24] <FunPika> no objections here
[2013-08-10 19:41:26] <JohnLewis> Why under a week specifically Izmeetingz?
[2013-08-10 19:41:30] <Mlpearc> 1 agree
[2013-08-10 19:42:05] <Izmeetingz> JohnLewis: honestly even after a week, I can't tell the difference between one person or another on shared IPs
[2013-08-10 19:42:21] <Izmeetingz> like school IPs
[2013-08-10 19:42:28] <Izmeetingz> except if it's static
[2013-08-10 19:42:44] <JohnLewis> Hm. Well if it makes it better for CheckUsers in the current harsh environment inline with the difficulties of analysis, I agree.
[2013-08-10 19:43:33] <Izmeetingz> Ok.
[2013-08-10 19:44:18] <Izmeetingz> So obviously we all agree on a proxy queue also. Hell, it's going to help more people getting involved with that queue also
[2013-08-10 19:44:40] <harlot> I can/will deal with the proxy queue
[2013-08-10 19:44:58] <JohnLewis> Ditto with what harlot said.
[2013-08-10 19:45:17] <JohnLewis> Except it will be 'I will' :p Not a 'I can'
[2013-08-10 19:45:39] <Izmeetingz> Now, as for conditions in which some people should defer back to users without needing a CU, I think we have the following points:
[2013-08-10 19:45:49] <Izmeetingz> 1) No block in the first place
[2013-08-10 19:45:54] <Izmeetingz> 2) CU has already commented
[2013-08-10 19:46:15] <Izmeetingz> 3) Inappropriate to defer to CU in the first place
[2013-08-10 19:46:38] <Izmeetingz> 4) Block has expired, unless it was noted it was a CU block
[2013-08-10 19:46:46] <Izmeetingz> Any more to add or ammend?
[2013-08-10 19:47:09] <JohnLewis> Comment on 2) Only when the CheckUser has an answer or response. Some checkusers do 'Ask DQ about this/have DQ comment on this' In such cases, they should not be deferred.
[2013-08-10 19:47:32] <Izmeetingz> yes true
[2013-08-10 19:47:38] <harlot> 5) Cases where checkuser will be meaningless (a la class B Materialscientist rangeblocks)
[2013-08-10 19:47:55] <Mlpearc> lol
[2013-08-10 19:48:03] <Izmeetingz> *class B india Materialscientist rangeblocks
[2013-08-10 19:48:26] <Mlpearc> ^ likes that one
[2013-08-10 19:48:36] <Mlpearc> jk
[2013-08-10 19:48:52] <FunPika> wouldn't that fall under 3? ;)
[2013-08-10 19:49:37] <Mlpearc> No, don't say that :P
[2013-08-10 19:50:04] * Mlpearc well it had his name on it 9_9
[2013-08-10 19:50:10] <Mlpearc> lol
[2013-08-10 19:50:20] <Izmeetingz> FunPika: not really. Inappropriate to defer to CU is something like "I think this one edit of 20 million could be a sock edit, so check this pls"
[2013-08-10 19:50:46] <FunPika> all right
[2013-08-10 19:50:46] <Izmeetingz> excuse my extremes
[2013-08-10 19:50:54] -->| Logan_ (~Logan@ubuntu/member/logan) has joined #wikipedia-en-accounts-meeting
[2013-08-10 19:51:01] <Logan_> Is there some kind of stupid meeting here?
[2013-08-10 19:51:07] <Logan_> I blame DeltaQuad.
[2013-08-10 19:51:07] <Izmeetingz> so any more to add?
[2013-08-10 19:51:14] <Izmeetingz> Logan_: yes, and it's logged...
[2013-08-10 19:51:18] <JohnLewis> xD
[2013-08-10 19:51:21] * Mlpearc some kind
[2013-08-10 19:51:34] <Logan_> Someone brief me, please.
[2013-08-10 19:51:45] <Izmeetingz> Brief in PM pls
[2013-08-10 19:51:48] <Logan_> And public logging is illegal when I'm in the channel, FYI.
[2013-08-10 19:52:08] <Izmeetingz> Logan_: shut up or I shall quiet you so we can have this meeting :P
[2013-08-10 19:52:08] <JohnLewis> Logan_: Not public :P Private but AOR
[2013-08-10 19:52:31] <harlot> Enjoy saying "public logging is illegal". I log everything.
[2013-08-10 19:52:38] <Logan_> You're such a harlot.
[2013-08-10 19:52:40] <Mlpearc> anyway, i !vote yes
[2013-08-10 19:52:55] <Izmeetingz> ...anyway. as I said, anything else to add on those?
[2013-08-10 19:53:00] <Logan_> On what?
[2013-08-10 19:53:16] <Izmeetingz> Logan_: conditions in which some people should defer back to users without needing a CU
[2013-08-10 19:53:20] <Mlpearc> No flooding either :P
[2013-08-10 19:53:23] <Mlpearc> lol
[2013-08-10 19:53:51] <JohnLewis> Nothing to add.
[2013-08-10 19:53:52] <Logan_> What is the consensus so far?
[2013-08-10 19:54:05] <Logan_> The guide currently states school blocks older than six months, so that's one.
[2013-08-10 19:54:06] <FunPika> Logan_: so far we have agreed on vandalism/{{schoolblocks}}/{{anonblocks}} not needing to be deferred if it has been more than a week since the block was placed and on adding a separate queue for open proxies...now we are discussing a few more points (wait a moment for those)
[2013-08-10 19:54:34] <Logan_> Excuse me, two weeks. My bad.
[2013-08-10 19:54:49] <JohnLewis> Old than 6 months Logan_? Data went stale three months before ;P
[2013-08-10 19:54:55] <FunPika> [19:45:49]  Izmeetingz	1) No block in the first place
[2013-08-10 19:54:55] <FunPika> [19:45:54]	Izmeetingz	2) CU has already commented
[2013-08-10 19:54:55] <FunPika> [19:46:15]	Izmeetingz	3) Inappropriate to defer to CU in the first place
[2013-08-10 19:54:55] <FunPika> [19:46:38]	Izmeetingz	4) Block has expired, unless it was noted it was a CU block
[2013-08-10 19:54:55] <FunPika> [19:47:09]	JohnLewis	Comment on 2) Only when the CheckUser has an answer or response. Some checkusers do 'Ask DQ about this/have DQ comment on this' In such cases, they should not be deferred.
[2013-08-10 19:54:55] <FunPika> [19:47:38]	harlot	5) Cases where checkuser will be meaningless (a la class B Materialscientist rangeblocks)
[2013-08-10 19:54:57] <Logan_> FunPika: I see.
[2013-08-10 19:55:22] <Izmeetingz> Logan_: we are looking at a month per this comment:
[2013-08-10 19:55:24] <Mlpearc> I'll be back in a few :P
[2013-08-10 19:55:27] <Izmeetingz> *a week
[2013-08-10 19:55:38] <FunPika> those are conditions when something should be deferred back to users
[2013-08-10 19:55:41] <Izmeetingz> 19:41:31 <JohnLewis> Why under a week specifically Izmeetingz?
[2013-08-10 19:55:41] <Izmeetingz> 19:41:36 <Mlpearc> 1 agree
[2013-08-10 19:55:41] <Izmeetingz> 19:42:10 <•Izmeetingz> JohnLewis: honestly even after a week, I can't tell the difference between one person or another on shared IPs
[2013-08-10 19:55:41] <Izmeetingz> 19:42:26 <•Izmeetingz> like school IPs
[2013-08-10 19:55:42] <Izmeetingz> 19:42:33 <•Izmeetingz> except if it's static
[2013-08-10 19:55:49] <Izmeetingz> Mlpearc: k
[2013-08-10 19:56:01] <Logan_> [19:54]  <FunPika> [19:46:15] Izmeetingz 3) Inappropriate to defer to CU in the first place
[2013-08-10 19:56:05] <Logan_> I feel like that could be clarified.
[2013-08-10 19:56:11] <Logan_> What is considered "inappropriate?"
[2013-08-10 19:56:25] <FunPika> [19:50:20]	Izmeetingz	FunPika: not really. Inappropriate to defer to CU is something like "I think this one edit of 20 million could be a sock edit, so check this pls"
[2013-08-10 19:56:31] <harlot> No reason to CU
[2013-08-10 19:56:36] <Logan_> Okay.
[2013-08-10 19:56:46] <harlot> ugh, storm rolling in
[2013-08-10 19:56:56] <harlot> I could start rapidly connecting and disconnecting
[2013-08-10 19:57:08] <Izmeetingz> harlot: send it rschen7754|boat's way :P
[2013-08-10 19:57:23] <Logan_> I think I agree with these.
[2013-08-10 19:57:26] <Izmeetingz> anyway. Logan_ anything else to add?
[2013-08-10 19:57:36] -->| Galadrial (~Reaper@wikipedia/Reaper-Eternal) has joined #wikipedia-en-accounts-meeting
[2013-08-10 19:58:02] <Logan_> I feel like a lot of the discretion is left up to CUs still, though.
[2013-08-10 19:58:08] <Logan_> So I don't see how this will really ease the backlog.
[2013-08-10 19:58:20] <Logan_> I could be mistaken.
[2013-08-10 19:58:37] <Izmeetingz> Logan_: instead of just school blocks, it's now anon/vandalism blocks also
[2013-08-10 19:58:44] <Logan_> On local or on range?
[2013-08-10 19:58:51] <Izmeetingz> either or
[2013-08-10 19:58:57] <Logan_> I feel like the latter is risky.
[2013-08-10 19:59:02] <harlot> why?
[2013-08-10 19:59:35] <FunPika> with anon/schoolblocks, it is rare for CUs to have useful data from a shared IP after a week
[2013-08-10 19:59:39] <Logan_> Because it denotes widespread vandalism that is likely to come from any IP on that range.
[2013-08-10 19:59:41] <JohnLewis> Logan_: As harlot said, It is easier to block a user and revert them than wait 1 month for a CheckUser to finally come and check it.
[2013-08-10 19:59:43] <Izmeetingz> Logan_: I also stated above that if users feel it is more static/there is a higher possibility that socking is occuring, they can defer
[2013-08-10 20:00:36] <Izmeetingz> Logan_: not really. some vandalism blocks have a LOT of collaterial
[2013-08-10 20:00:54] <Galadrial> Logan_: No. It's not likely to come from any IP
[2013-08-10 20:00:58] <Logan_> Izmeetingz: How can you "feel it is more static?" The guide already advises against using sites like http://whatismyipaddress.com/.
[2013-08-10 20:01:46] <Galadrial> Some ISPs tend to give very static IPs, even if you have access to a massive range
[2013-08-10 20:01:59] <Galadrial> I cover multiple noncontiguous /16s
[2013-08-10 20:02:07] <Galadrial> However, my IP rarely changes
[2013-08-10 20:02:10] <Logan_> Then we need to clarify how to tell this in the guide if we're giving regular ACC users this type of discretion.
[2013-08-10 20:02:12] <Izmeetingz> Galadrial: maybe we can create a general guide on which ISPs are dynamic/static
[2013-08-10 20:02:30] =-= harlot is now known as KittyReaper
[2013-08-10 20:02:55] <KittyReaper> Mobile ISPs: Extremely dynamic
[2013-08-10 20:03:09] <Izmeetingz> yes very much so
[2013-08-10 20:03:11] <FunPika> same with me, while with Comcast I technically have a dynamic IP, I'd say it actually changes one or two times a year at most
[2013-08-10 20:03:25] <JohnLewis> Oh god. Britains dynamic systems... :(
[2013-08-10 20:03:37] <KittyReaper> BT Internet: "Fuck it. Let's give em the whole class A"
[2013-08-10 20:03:39] <Izmeetingz> JohnLewis: ikr...story of my life...
[2013-08-10 20:03:42] <Logan_> Izmeetingz: I'd suggest making a sandbox for this subguide so that anyone with this type of knowledge can contribute.
[2013-08-10 20:03:49] <JohnLewis> I probably get a new IP every two days.
[2013-08-10 20:03:58] <Izmeetingz> Logan_: i'd agree
[2013-08-10 20:04:06] <JohnLewis> On occasions, I have three new IPs in the same 24 hours period >.<
[2013-08-10 20:05:03] <Izmeetingz> anyway, other than that, that should be a good kick into the CU backlog. Any other concerns/comments or questions people would like to bring up?
[2013-08-10 20:05:09] <Logan_> Yes.
[2013-08-10 20:05:14] <Izmeetingz> go ahead
[2013-08-10 20:05:16] <Logan_> "(a la class B Materialscientist rangeblocks)"
[2013-08-10 20:05:34] <Izmeetingz> what about them?
[2013-08-10 20:05:43] <Logan_> We need to give examples in the guide of what people should look for in the rangeblock summaries in order to determine whether or not they're meaningless in the context of ACC.
[2013-08-10 20:06:08] <Izmeetingz> Logan_: those specific ones, I always change the block telling ACC to ignore them
[2013-08-10 20:06:27] <Logan_> But how do you know?
[2013-08-10 20:06:44] <Izmeetingz> 1) they are from india 2) they are a rangeblock
[2013-08-10 20:06:49] <JohnLewis> DoRD (Or is it DQ, I think DoRD) usually reblocks users with '<!-- acc please ignore this block -->' which more range blocks (specifically CUblocks when appropriate) should probably have.
[2013-08-10 20:06:52] <Izmeetingz> rangeblocks + india = CU hell
[2013-08-10 20:06:54] <Logan_> If the goal is to reduce stress on CUs, then you adding that comment to every insignificant rangeblock is certainly not doing that?
[2013-08-10 20:07:33] <Logan_> s/?/./
[2013-08-10 20:07:35] <Izmeetingz> Logan_: but sometimes it's the price. We can't say in a concreate way which rangeblocks should be ignored and which shouldn't
[2013-08-10 20:07:44] <Logan_> Gotcha. No further questions/comments from me.
[2013-08-10 20:08:10] <Izmeetingz> JohnLewis: ya that's normally me. but DoRD might of started doing it
[2013-08-10 20:08:44] <Izmeetingz> So anymore general questions about ACC and CU?
[2013-08-10 20:09:10] <JohnLewis> Well if CheckUsers could do that more, it should probably lighten the queue up when appropriate.
[2013-08-10 20:09:13] <JohnLewis> Nope
[2013-08-10 20:09:43] <Izmeetingz> JohnLewis: mark the rangeblocks? ya. I'll try and pass the message on to active ACC CUs
[2013-08-10 20:09:53] <JohnLewis> Ok
[2013-08-10 20:10:05] <Izmeetingz> FunPika: Galadrial Mlpearc rschen7754|boat any other questions or concerns?
[2013-08-10 20:10:17] <FunPika> nope
[2013-08-10 20:10:21] <KittyReaper> nah
[2013-08-10 20:10:29] <Galadrial> I don't
[2013-08-10 20:10:43] <Izmeetingz> lol votestacking ^^
[2013-08-10 20:10:59] * KittyReaper is screwing with people's minds....
[2013-08-10 20:11:16] <Izmeetingz> --End of meeting and official logging--