We've been impressed with Google's ongoing transparency report
efforts, and it appears that the folks at Twitter agree. The company recently released its own Transparency Report
(and gave a hat tip to Google in the process). Like Google's report, Twitter's looks at how many requests for information it has received, as well as how many copyright takedown requests. The first chart it showed highlights how nearly all of the government requests for user info it receives comes from the US:
As you can see, the US had 679 requests for info, dwarfing second-place Japan, who had 98 requests. After that, the highest you got was the UK and Canada with 11 requests each. There may be a lot more American Twitter users, but still, that's quite a spread. Perhaps more interesting is that Twitter appears to have resisted an awful lot of the requests (especially coming from other countries). However, in the US it did end up abiding by 75% of requests. This isn't a huge surprise, as there have been a few court cases that have shown that Twitter can be compelled to cough up such info here. At the very least, Twitter has been very good
about letting its users know that their information is being sought, whenever it can (though there may be cases where it's legally restricted from doing so).
Next up, Twitter showed how often it gets requests not to provide info, but to flat out remove
Tweets (outside of copyright takedowns, apparently):
Here you see that there aren't really that many requests -- and thankfully none from the US. What's most interesting here, of course, is the fact that Twitter has refused to remove such info
each and every time. Considering how many companies just roll over when they receive an official government correspondence or court order, this is really nice to see (and, yes, a bit surprising). As the company notes, it will
remove content for "rules violations" (and for copyright infringement, as we're about to see), but these requests fall into a different category -- likely just information that governments didn't want to exist.
And, finally, they discuss copyright takedowns. This is actually an area where we've felt that Twitter doesn't always
do a very good job, as there have been multiple stories
of questionable DMCA-based takedowns.
Given our earlier stories, I'll admit that I'm a bit surprised to see that Twitter seems to actively resist questionable DMCA takedowns -- to the point that only 38% of requests resulted in material being removed. The company admits
that it gets a ton of bogus complaints and is careful to review them:
We also receive a large number of misfiled, non-copyright complaints through our web form. We carefully review each report received, and follow-up with the reporter as appropriate.
There may be room for improvement here, but it does appear that (unlike many companies), Twitter doesn't just roll over on every DMCA request.
Either way, it's great to see these kinds of transparency reports that really reveal some detailed info on just how often companies like Google and Twitter are asked to remove content or reveal info on users.