Transparency Watch Releases Searchable Database Of 27,000 US Intelligence Workers

from the publicly-posted-information,-searchable-by-the-public dept

Intelligence gathering on intelligence gatherers. Watching the watchers. Whatever you want to call it, Transparency Toolkit is doing it. It has gathered 27,000 publicly-posted resumes from members of the "intelligence community" and turned them into a searchable database.

The database -- ICWatch -- was put together using software specifically constructed by Transparency Watch (and posted at Github). Not only can the database be searched through TW's front end, but the data is also available in raw form for data-mining purposes.

Some may find this searchable database to be a form of doxxing, but TW says that isn't the intent. Instead, it's meant to give the public additional insight into the inner workings of the intelligence community, as well as allowing researchers and journalists to sniff out information on still-unrevealed surveillance programs.

"These resumes include many details about the names and functions of secret surveillance programs, including previously unknown secret codewords," Transparency Toolkit said.

"We are releasing these resumes in searchable form with the hopes that people can use them to better understand mass surveillance programs and research trends in the intelligence community."

What Transparency Watch has done is simplified a task anyone could have performed prior to the compilation of the ICWatch database. In fact, nearly two years ago, the ACLU's Chris Soghoian pointed out that public LinkedIn profiles were coughing up classified program names posted by intelligence community members in their listed skills and work history.

This is all Transparency Watch has done -- only in aggregate and accessible to those without a LinkedIn account.

The data was collected from LinkedIn public profiles using search terms like known codewords, intelligence agencies and departments, intelligence contractors, and industry terms, the group said.

What Soghoian noted back in 2013 remains true. Searches for known NSA programs frequently bring up other program names, all posted publicly by employees and contractors with an apparent disregard for the agency's "everything is a secret" policies.

A search for "PINWALE" brings up a profile listing the following:

Cultweave, UIS, Nucleon, CREST, Pinwale, Anchory, Association, Dishfire, SharkFinn, GistQueue, GoldPoint, Mainway

And another listing these terms:

Snort, TRAFFICTHIEF, PINWALE, BOUNDLESS INFORMANT, BLARNEY, BULLRUN, CARNIVORE

You can also find out who's involved in Predator drone flights. Or who's participated in the NSA's Tailored Access Operations.

Some may argue that this algorithmic collection of resumes and LinkedIn profiles may be dragging some people under the "intelligence community" umbrella that shouldn't really be there. That's likely true, but this is one of those inescapable outcomes of dragnet operations. They may also argue that turning over this information to the public may cause some of those listed to be subjected to harassment or put them in danger. Also, this may unfortunately be true as well.

But there's a simple solution, albeit one that can't be applied retroactively.

As the government so frequently points out to us, publicly-posted information carries no expectation of privacy. The same goes for government employees and government contractors in sensitive positions who choose to disclose information about their skills and employment publicly. If any danger to these people exists, it has always existed. ICWatch may make the job simpler, but it's done nothing any person can't do on their own, using simple search tools.

Filed Under: code names, database, intelligence, intelligence community, resumes
Companies: linkedin, transparency watch

Details Revealed On Old NSA Intelligence Database: ANCHORY

from the from-way-back-when dept

You may remember that, back in June, we pointed out that if you plugged in a few of the "code names" for various NSA programs (as revealed by Ed Snowden's leaks), you could find a few resumes of NSA employees, listing out other such code names. Jason Gulledge apparently saw that post, and used the list of code names that we posted to file a Freedom of Information Act (FOIA) request on NSA documents concerning those programs. Amazingly, they actually sent back some info -- though, just about the very first program, ANCHORY, and the info sent is from 1993 (and some from 2000), describing the text search system that the government has to search through various intelligence reports. From 1993, long before "search engines" were a thing:

ANCHORY is a system that provides timely retrieval of textual data by keyword and other relevant information on a 24-hour per day, 7 day per week basis. [REDACTED] is the ANCHORY hardware platform and BRS/Search, a commercial software product, is the engine behind ANCHORY text retrieval and database management functions.

[....] ANCHORY contains three years of full text reports written by NSA, CIA, DIA, State and Foreign Broadcast Information System, as well as Reuters News Service, Cryptologic Intelligence Reports and precis of hard copy reports.

The documents are interesting, mainly for the historical interest aspect. Though, at the same time, it does suggest that the NSA is (once again) completely full of crap when it claims that it doesn't have the technology to search its own emails. If it had text search twenty years ago for all reports, it has the capability to do email search today (not that this is a surprise).

Unfortunately, the FOIA person at the NSA rejected Gulledge's request to waive the FOIA fees, which it will do for media operations when there is a strong public interest basis for the request. Muckrock, the FOIA service Gulledge is using (and which, of course, we've used many times as well) is now appealing that decision. However, because of that, the NSA has not looked into whether it can send the documents on any of the other code named programs listed.

Filed Under: anchory, code names, foia, nsa, search