Robert

+1 LICENSE (or COPYING) files are the standard used in almost every piece of open source software. rilts has very little uptake or mindshare. The lack of an obvious LICENSE...

Almost every one i look at seems wrong when I look at the BigQuery data. here's another one: ```sql SELECT id, CONCAT( "https://github.com/",repo.name,"/commit/",JSON_EXTRACT_SCALAR(payload, "$.commits[0].sha"),".patch") AS url, JSON_EXTRACT_SCALAR(payload, "$.commits[0].author.email") AS email,...

I need to run out now, but I quickly looked at the two pull requests from `githubarchive.month.201601` (randomly chosen as "a while ago") and they (at least the first commit...

(hit wrong button)

I spent some time tonight trying to get more data on when the issue started... ``` WITH commits AS ( SELECT JSON_EXTRACT_SCALAR(payload, "$.commits[0].author.email") AS email, JSON_EXTRACT_SCALAR(payload, "$.commits[0].sha") AS commit, repo.name...

It seems very unlikely (but not totally impossible) that ruby's sha1 is producing the wrong output.

Consistently wrong would definitely be a step forward. I can't see anything about those MurderMysteryMansion commits that looks at all suspicious. Could it be something related to EM (eventmachine) or...

Major props to your teammate for that find. It looks like it started happening at 2016-05-27 19:54:28.000 UTC (or around that time, filtered to @gmail.com addresses for simplicity.) ``` WITH...

You're welcome. It was a fun one to figure out.

> treat and report separately mere clues such as this one: they could be an interesting insight in some cases, but alone they are too weak to be considered a...