Content deleted Content added
Line 158: #::Let's start with all the reverts of edits of railways stations on 1 Sep 2015 which then seem to be repeated the following day, e.g. [https://en.wikipedia.org/w/index.php?title=Talk:Reading_railway_station&diff=prev&oldid=678992686 Reading]; [https://en.wikipedia.org/w/index.php?title=Talk:Reading_railway_station&diff=prev&oldid=679111282 repeat]. [[user:Andrew Davidson|Andrew D.]] ([[user talk:Andrew Davidson|talk]]) 00:08, 5 January 2017 (UTC) #:::Let's start with asking you if you do it deliberately, {{U|Andrew Davidson|Andrew}} - it's beginning to raise a concern. [[User:Kudpung|Kudpung กุดผึ้ง]] ([[User talk:Kudpung|talk]]) 01:42, 5 January 2017 (UTC) #::::I wouldn't call Andrew D's participation at RfA trolling, but I would call it [[observationally equivalent]] to trolling. We're rapidly reaching the point where the community will need to evaluate whether Andrew D's continued participation at RfA is a net positive. Andrew has always leaned heavily toward opposing, but until recently, the percentages seemed reasonable. Up until the end of June 2016, 14 supports and 40 opposes. After that, it goes off the rails with a single support and 14 opposes, mostly with extremely reaching rationales that are well removed from what the community requires of an administrator. Such a serious departure from past practices of supporting a reasonable percentage of candidates suggests ''something'' has changed. ~ [[User:BU Rob13|<b>Rob</b><small><sub>13</sub></small>]]<sup style="margin-left:-1.0ex;">[[User talk:BU Rob13|Talk]]</sup> 11:22, 5 January 2017 (UTC) #'''Oppose''' I share Andrew D's concerns about code quality, and therefore if admin bot creation is going to be bundled up in this nom, I have to oppose. Cyberpower is a fine contributor that will not do harm to the project through his own direct actions. I do not have the same confidence of his bot work. [[User:Samsara|Samsara]] 02:52, 5 January 2017 (UTC) #:Overall, Cyberpower's bots have 1,824,280 edits. Andrew D has demonstrated a couple hundred reverted test edits. Giving Andrew a very generous 500 failed edit count, that amounts to a 0.02% failure rate. What level of accuracy do you require to support? PLEASE tell me you do not work in information technology if 0.02% scares you. By comparison, NASA has a 1.64% fail rate for manned spaceflights (0.79% in the last 20 years) and a 8.08% fail rate for unmanned flights (6.68% in the last 20 years).--v/r - [[User:TParis|T]][[User_talk:TParis|P]] 03:02, 5 January 2017 (UTC) | |||