Research talk:Wikipedia article creation/Work log/Thursday, December 19th

Thursday, December 19th

Back to hacking today on survival of pages by their origin. I bad a plot that is obviously wrong, but still communicates some cool stuff, so I wanted to post about it.

Article page survival (enwiki). The monthly proportion of surviving articles is plotted for "article pages" by the namespace from which they originated and by the tenure of the editor at time of draft creation.

So, I didn't plot the survival proportion of any (original_ns, tenure, month) set of articles unless there were at least 100 of them. So, we don't see AfC (ns=5) appear very often for newcomers since they rarely publish via AfC.

Another thing you'll notice is that there are no pages moved from non-NS=0 pages after 2012. It appears that is due to a change in the structured comments I was parsing. Newer move comments look like this:

> select rev_comment from revision where rev_id = 586811213;
+--------------------------------------------------------------------------------------------------------------------------+
| rev_comment                                                                                                              |
+--------------------------------------------------------------------------------------------------------------------------+
| Epicgenius moved page [[Talk:The Bronx/Name]] to [[Talk:The Bronx/Name and capitalization]]: capitalization is here, too |
+--------------------------------------------------------------------------------------------------------------------------+
1 row in set (0.03 sec)

So, now I have to go update my move detection SQL to capture that style of comment too. --Halfak (WMF) (talk) 17:30, 19 December 2013 (UTC)