Tuesday, May 17, 2022

Superb example of SearchResearch... in Algeria

 I have a wonderful video I want to recommend to you.  It's about what this mysterious circle of circles is... 

In this video from Vox, they explore the limits of what you can find by internet searching, and go beyond the limit when they realize that they'll have to visit this site in person.  It's not exactly around the corner: it's at 27.270161, 4.322245, which is, you'll quickly realize, in the middle of the formidable Algerian desert.  

The key question is: When / what / who made these circles in the (literal) middle of nowhere? 

I didn't expect to watch this entire video, but it is well worth the time.  In it, the researchers do all of the things you'd expect from a SearchResearch Challenge (finding original sources, locating experts, contacting them, etc).  It's a wonderfully produced video that lays out their research process step-by-step.  Check it out, and tell me if you're not pulled into the mystery after the first 30 seconds.  

Bravo, Vox!  Bravo!  

Vox's video: Who made these circles in the desert?  

Search on!  


  1. This comment has been removed by the author.

  2. who knew ground truth could be so explosive… and there would be nothing fishy about sardines showing
    the French crepes aren't always what they appear… or that colonialism would leave such a bad taste…
    thanks Dan, Christophe, Samir & team and can collectors everywhere… & Google, in its many forms.
    fwiw, I was struck by how the many experts were so certain of their views…
    (also saw that Bellingcat made CBS's 60 Minutes in the U.S.)
    impressive storyteller
    TOTAL related - started 1956
    is information akin to oil?
    SERP snippet mentions 1958…
    post France, but the tentacles remained

  3. I watched the video, and thought it was really interesting. So, thanks for posting it.

    As well, I evaluated it using the CRAAP test (see below for sources), an acronym developed by a librarian at CSU Chico which stands for Currency-Relevance-Authority-Accuracy-Purpose, and which I've found to be useful when evaluating different websites. Also, due to its length, I had to break this comment into two parts. So, here's how this video stacks up with it:

    * Currency (when it was made) - It was made this year (and was posted on May 10), meaning it's very recent, so it passes.
    * Relevance (is it relevant to the topic at hand?) - Given it was posted on a blog about web search, and given that it deals with a web search query, it is relevant and passes.
    * Authority (who posted it, and what are their credentials) - It was posted by a reporter at Vox, which is a news agency. Also, he explained each step he followed for answering this question (questioning, investigating, finding hypothesis, testing it, finding another one and testing it), which adds to his credibility, to say nothing of how he followed the scientific method with his query. Third, he showed his sources, which boosts his own credibility. Fourth, he was able to speak French fluently, which enabled him to talk directly to some people without need of a translator. Fifth, he knew when to stop. Finally, Dan Russell, a senior search engineer at Google, vouched for it, which further speaks to the reporter's authority. So, it passes.

    (end of part 1)

  4. Part 2

    * Accuracy (how accurate is it? And, does it seem right, based on what you know?) - Again, Dan Russell gave it as an example of how to do web search, which speaks to its accuracy. Also, the method he used for finding it seems like a good one (though he did more asking of others who were knowledgeable about the topic, while I would've relied more on using lots of search queries [though both are merely different ways of solving this type of problem]). Third, the people whom he asked (someone at Vox who knew about maps, someone with the company that took Google's satellite photos, an expert on seismic imaging, someone at Vox who was an expert in videos, a Tunisian archaeologist, people who had studied qanats, a farmer, an Algerian journalist who was able to arrange a visit to the site, a Algerian archaeologists, a guide who knew that part of Algeria, a French sardine can collector, a retired desert guide who remembered the seismic surveys) knew what they were talking about.

    Fourth, he gave needed context at different parts of his search query (e.g. oil drilling in Algeria, tombs in Algeria, 2,500-year-old irrigation systems). Fifth, he refused to stop when he had reasonable-sounding hypotheses (oil exploration, wells and tumuli, ancient irrigation systems [qanats]) but, rather, kept on investigating (and throwing out ones that didn't hold up) until he got a the answer (seismic surveys for oil that used dynamite and which were carried out by CREPS [a joint venture between Shell and the French government] in 1957 or 1958, though they didn't find oil), which further speaks to his credibility. Sixth, he showed various sources he used (maps of seismic surveys in Libya, an article from 1885 about wells and tumuli is Algeria, pictures of wells, database of Saharan oases, better-quality videos, footage from a long trip into the Sahara and to the sites [though it proved the Algerian journalist had gone there], seismic surveys that used dynamite, the company that made the sardines, information on CREPS, Algeria's war for independence), which improves this video's accuracy.

    And, finally, one person whom they'd communicated with earlier who understood seismic surveys commended them on a job well done. So, it passes.
    * Purpose (what was the point or purpose of the video?) - It relies on facts, it's objective, and it's an investigation that relied on credible sources, so it's about fact-based, in-depth reporting. Thus, it passes.

    1. https://commons.emich.edu/cgi/viewcontent.cgi?article=1009&context=loexquarterly (paper that introduces it)
    2. https://library.csuchico.edu/sites/default/files/craap-test.pdf - the test itself, and questions to ask for each part.

  5. also in the Sahara, Mauritania
    more mystery:
    in Mauritania

  6. I figured out early on that someone was going to have to visit the site. We live in a three (or more) dimensional world and it can be difficult to interpret correctly a two-dimensional image. Perhaps the remoteness of the circles prevented this from being the first step.

  7. Thoroughly enjoyed this video with the tenacity of a searchresearch searcher, the logical steps, the tirangulating of information, the use of experts, and open minds. I can see a whole new type of travel/tours, Dr. Dan please lead the way.