>>9112861 (LB)
>If they are there are messages hidden in twitter file names that Q wants us to see.
No. They're base64 encoded timestamps + some (probable) randomness to prevent naming collision. I noticed the pattern after the famous "DOITQ" filename. More recently I finally bothered to workout their system in detail. Here is some Python code:
https://pastebin.com/yJMjGfKB
>Are you positive?
Yup. I sampled Twitter's image filenames going back several years, decoded the first 6 chars like it's standard URL-safe base64, and them put in a scatter plot vs the known posting time. The resulting graph is a flat as a laser.
I wasn't sure what I'd get. If they are simply sequential then I'd expect some sort of curve representing varying activity over time. If it was encoded date/time stings thing I'd expect to see steps in the integer representation. And it it was a unixtime-like system based on the seconds since some epoch them I'd expect to see a flat graph. I got a flat graph. After that it was a matter of applying that y=mx+b equation we all learned in high school (but never needed) to workout the slope and epoch (Nov 3, 2010, not a coincidence that this is about the time the the new Twitter system came online). There are some errors because of my neglecting timezone and daylight-savings issues in the Know Posting Time samples. But it's good enough to confirm my hypothesis.
The Badguys have infected pretty much everything. So it's not a surprise that you found spoopyness under whatever random rock you kicked over.