Data Export for the research project of http://twitter.com/se_statistics
Q&A about the site for operations research and analytics professionals, educators, and students
SELECT top 100 p.Id as antwort_id, q.Id as frage_id, p.Body, p.Score, p.OwnerUserId, u.Reputation, q.AnswerCount, q.ViewCount, v.VoteTypeId, p.CreationDate , p.Id as [Post Link] FROM Posts as p JOIN Posts as q on p.Id = (select top 1 Id from posts where parentId = q.Id and CreationDate < '2011-09-01 00:00:00' and CreationDate >= '2010-09-01 00:00:00' -- from Year 2010 and LastEditDate < DATEADD(month, -3, GETDATE()) -- at least 3 months untouched and LEN(Body) > 550 -- more than 450 chars used (TODO: striptags) and Score >= 0 -- Score not negative ORDER BY NEWID()) -- only answers JOIN Users as u on p.OwnerUserId = u.Id LEFT JOIN Votes as v on p.Id = v.PostId WHERE (v.VoteTypeId = 3 OR v.VoteTypeId IS NULL) and q.ClosedDate IS NULL and p.CommunityOwnedDate IS NULL -- not closed, not community wiki and q.CreationDate < '2011-09-01 00:00:00' and q.CreationDate >= '2010-09-01 00:00:00' -- from Year 2010 and q.AnswerCount <= 10 -- not more than 30 answers (pagination) ORDER BY NEWID()