<Deep, smoky narrator voice>: Ladies and gentlemen, may I present to you ... based on piles of newly released secret intelligence files, with stunning, never-seen-before 1960s spy photographs ... the Active Measure NEPTUN https://www.wired.com/story/uncovering-operation-neptun-the-cold-wars-most-daring-disinformation-campaign

This narrative of NEPTUN is based on newly released sources from the Archiv bezpečnostních složek. The project files are about 2,800 pages, with more than 500 photographs—which makes NEPTUN the best-documented AM I know of.

Full file @internetarchive: https://archive.org/details/stb-neptun-90039

Btw I’m very excited about my ACTIVE MEASURES collection up on the @internetarchive. Almost every obscure source has a URL in the endnotes. The book basically comes with a ~33,000 page appendix of primary and very-hard-to-find sources.

Example:

Ah, while I'm at it, let me take you down a rabbit-hole, about the making-of of ACTIVE MEASURES. I’ll reveal a few dirty research secrets along the way.

In late March 2017 I had a drink in DC with @JohnHultquist, I think at Bar Pilar. We talked about my upcoming SSCI testimony. I was a little nervous. At one point John showed me a Facebook account then still online, of “Melvin Redick,” patient zero for DCLeaks, a GRU leak site.

The find was exciting. I made a screenshot, and filed it away. The following weeks were a bit hectic. Then, at some point, I recalled the screenshot and “Redick,” but I had forgotten where I filed it. So I went to my Google Drive search bar. I keep all my work files in a G-Drive.

I typed “Redick.” In a split second, the result below showed up. And I was like: Oh. My. God.

Let me spell that out. The result was a PNG file. And “Redick” was not in the file name. Google, it dawned on me, was text-recognizing *images* in my drive, and at a very impressive quality. “Redick” was rather small in the screenshot. https://archive.org/details/20160608-dcleaks

Wait, I thought. Just two months earlier CIA had put its entire archive online. ~930,000 files. More than 12 million pages. The archive was a goldmine for historians. But the OCR quality wasn’t great. Worse, the search function on the CIA website was so bad it was unusable.

So there I had my summer-side-project: download the entire CIA archive—and upload it into my Google Drive.

wget can't handle that load, of course. So we made a Python script for the grab. I discovered that the CIA site would block me if I requested more than one file per second. But not for a long time.

Downloading took a loooooong time. More than two months. Very long logs. Many.

Then, finally

Then came the upload to Google Drive. It turned out that the regular client, “Backup and Sync,” was the best way to lift this giant load into the cloud. And it wasn’t very good. At all. Crashed all the time. And I saw more “Processing …” than I care to recall. Took months.

The outcome was magical. I was able to make cross-connections that would likely be impossible to uncover with conventional techniques. I could read results faster, and anywhere, by swiping left and right. This drastically increased my intake. Then there was the language support.

One day, too close to my manuscript deadline, I received a release from the StB archive in Prague. More than 33,000 pages of files. From Bittman I knew that at least one or two KGB memos should be in that release. I assumed KGB would write in Russian, not Czech. I typed “КГБ.”