Monday, August 21, 2017

The search power of a social network

People working together... 

... can sometimes discover amazing things.  

This is true in news reporting (think of the Spotlight investigative team at the Boston Globe, or the Watergate scandal investigative team at the Washington Post--reporters Bob Woodward and Carl Bernstein).  It's true in our multi-week Challenges here at SRS, and it seems generally true:  Teams often beat individual researchers, and not just because they can dedicate more time to the task.  

A small group of researchers that are focused on a common task bring multiple perspectives and multiple ways of searching for the same target.  This is really one of those times when diversity really is MUCH better than a monoculture.  

In the April 3, 2017 issue of the New Yorker magazine, an article by Andrew Marantz tells a rather remarkable story is told of how a small group of dedicated people were able to repeatedly find a flag located somewhere in the world.  (The story is actually a bit sad: it's really about 4CHAN trolls searching for the flag in order to vandalize it, but the lesson for team research is still striking.)  
Could you locate this flag?

The thing that impressed me is that this team was able to find it even when the location clues were remarkably small.  

In one instance, the flag was shown flying in a live webcam view against just an open sky.  But by pulling together several streams of information (e.g., the last known tweet location of the flag installer), they were able to generally locate the flag. Then, since it was shown on a live webcam, planes flying through the view could be tracked down by using one of the live flight data streams (e.g., FlightRadar24).  It only takes a couple of known planes to triangulate the spot.

Then, one of the searchers (who happened to live nearby), drove around the area honking his car horn until one of the search team heard his horn... 

This search process had a fairly trollish goal, but the story illustrates the power of a search team working together.  They had to pull together bits of data from different places in order to find the target.  

Three important observations:  

(1) There wasn't a master plan to use those data streams made ahead of time--the plan was cooked up in the moment using whatever data was available.  Being able to improvise is an important skill.  

(2) The diversity of the team was powerful.  They had different people with different expertise.  Some knew about different data sources, some were able to sit and just watch the stream until something happened that they could use. 

(3) The team took direct action to probe the likely location (by honking their horn and listening in on the stream).  Sometimes research takes more than just looking up stuff online.  Every so often, you have to go out into the field and collect the data directly.

Search on... but be prosocial in your research!  

Friday, August 18, 2017

Changes in SearchResearch. (Simpler Challenges and an upcoming book)


... as you might have noticed, after a few years of pretty punctual posts, the past few months have had a series of slightly delayed posts, or posts that ran over multiple weeks.  

You might wonder why. 

Part of it has been all of my travels interfering with my posting.  (It's really hard to connect to wifi from a boat bobbing in the ocean way offshore.)  

But there are two other things going on:  First, as I mentioned in this week's Challenge, sometimes the questions turn out to be tricky, and take a lot longer than I have budgeted for my SRS writing.  Generally, that's great, as it means there's really something interesting going on, and it's worth dedicating the time to writing it up. 

My goal in writing SRS is to help readers understand more about how to do online research.  Sure, I love to show you the inner workings of Google search, but as you've probably noticed, we end up touching on lots of different topics.  (There are times when I've recommended Bing, Wolfram Alpha, and many, many different databases and data sets.)  

As I've said before, one of my ultimate goals in setting up and writing this blog has been to figure out if there's a need for a book about "how to do online research."  And if so, what should be in such a book.  

Not actually me, but close.  Original work by Rembrandt. 
Second:  The second big thing that's been going on is that I'm trying to finally actually write that book.  

To nobody's surprise, the book will be based on what we've learned in SRS, and will be heavily inspired by what we've learned here.  

So now I'm writing this book.  It's tentatively called "The Joy of Finding Out," (affectionately known as JOFO). I bet it'll get a subtitle later on, perhaps something like "How to do fast, effective, and accurate research on the internet."  Or something.  

I've already sketched out the entire book, and managed to write big chunks of it in-between writing the Challenges and Answers for SRS.  But it's been slow going since the average SRS week-of-writing is around 5 hours.  

What does this mean?  Well, first off, I need to cut back on my SRS writing time by a bit.  Going forward, I'm going to simplify the Challenges a bit.  I need to free up some time to work on the book.  I'll keep doing roughly 1 Challenge / Answer each week (although I reserve the right to stretch them out to a 2-week issue if I'm behind on writing).  

Also, I'd love to have a person or two act as a first-pass reader of the text.  If you're interested in (and have the time), let me know if I can send you a chapter or two for your comments.  Click here to let me know if you'd like to either (a) be an early reader, or (b) sign up to be notified when the book is complete.  

I'm still around... and will be working even harder on SRS... but much of that work will be on the book, and it won't come out for a while.  I'm hoping for mid-2018... but stay tuned. 

Keep searching... and keep reading.  There's MUCH more to come! 

--- Dan 

Thursday, August 17, 2017

Answer: Questions about the Yucatán? (#1000)

The Yucatán!  

Foreword:  Once again, a simple sounding SRS Challenge ended up taking me a LONG time to work through.  I'll explain why it took so long below.  But see the tomorrow's post for what this means...

The cenotes of the Yucatán are impressive places.  They are underground and full of water--but gorgeous.  

They're part of a large network of caverns that are mostly filled with water and marked by sinkholes.  The water is incredibly clear and a fascinating place to explore.    

Naturally, my short visit to the cenotes at Dos Ojos (near Cancun) made me very curious about cenotes, the Yucatán, and the whole place in general.  What's up with this place?  

1.  Cenotes seem to appear all over the Yucatán peninsula.  If you look at a map of the area, it seems they all line up just inland from the Riviera Maya coastline.  But I know there are cenotes in the north of the peninsula as well. Is there a larger pattern of cenotes at work here?  If so, what caused that particular pattern of cenotes to form? 
When thinking about "larger patterns" of cenotes, my first inclination is to search for a map.  I did a general search:

     [ map Yucatan cenotes ]

and found lots of tourist maps showing cenotes you could visit, but that's NOT what I'm looking for.  What I need is a map of ALL the cenotes in the peninsula.

To get a fast overview, I switched to Images mode, and saw this SERP:

If you look through this result set, there are really only a couple of maps that show ALL (or at least most) of the cenotes in the entire Yucatán peninsula. As you can see, there are thousands of them!

The original version of this map was by Jake Bailey and David Kring for the NASA/Univ. Arizona Space Imagery Center. 
It was recently modified by the Lunar Planetary Institute for teaching purposes. 

But there are two striking features of this map.  First, there's a pretty dramatic ring of cenotes with the center just off-shore near Merida.  

The other remarkable map of Yucatán cenotes is below, showing the cenotes in white.  It too has the same ring.
A dramatic ring of cenotes (marked by white dots) is associated with the largest peripheral gravity-gradient feature. The origin of the cenote ring remains uncertain, although the link to the underlying buried crater seems clear. (Gravity map adapted from Chicxulub Crater, Mexico, and the Cretaceous - Tertiary boundary, Canadian Space Agency.)

What's going on here?  I didn't expect a giant ring of cenotes.  Luckily, the source of this image gives us a big clue about what's going on here.  As the USGS points out, this is the edge of the impact crater formed by the Chicxulub meteor from 66M years ago.   (BTW, this USGS web page has an old, out-of-date URL link to the source page.  To find the new location of this page, I had to do this search:  [ inurl:chicxulub.htm] -- this limits the search to the site and looks for webpages with the name of the page in the URL, chicxulub.htm - the actual page with the report is LINK.  True as of August 16, 2017)  

But the second major feature of this map is that there seems to be a suspiciously straight-edge where cenotes seem to be missing (see below).  Is this a real feature (e.g., caused by a change in geology at that point), or could it be caused by just missing data?

Answering this question: Are there cenotes in here?  took me a couple of hours.  

In general, it's really hard to find missing data.  As a researcher, you're looking for something that might not exist!  

So I started by searching for other maps of this region.  I'll spare you all of the details, but one of the reasons that this "possibly missing data" is so difficult to track down is that this is not in Quintana Roo--it's mostly in the state of Yucatán and Campeche.  

And, as I found out, there are a LOT of studies of the cenotes in Quintana Roo, a few in Yucatán, and even fewer in Campeche!  I keep getting hints that there ARE cenotes in Campeche, (e.g. this article, Campeche, a Land of Still Unexplored Cenotes) but it’s hard to find a map showing them.  It’s hard even to find a master list of all the cenotes in this state or in Yucatán! Luckily, that article tells us that there are cenotes near the town of Miguel Colorado.  After just a few minutes of zooming around this area on Maps, it's really easy to see that there are a lot of small lakes, many in an arc and round in shape.  (You'd want to verify that these are cenotes, but I'm willing to bet you a dollar that's what they are.)  

Probable cenotes near Miguel Colorado, Campeche, MX.

But I kept searching, doing all kinds of things that didn't work out, and eventually I tried the query: 

      [ cenotes map peninsula ] 

in Google Scholar.  And this query took me to a fascinating article Review: The Yucatan Peninsula karst aquifer, Mexico (published in Hydrogeology Journal, 19.3, 2011).  And THIS article has the following map in it: 

As you can see, yes, there are cenotes in the center of the peninsula--they're just not quite as well organized as the ones in the Holbox or Rio Hondo areas.  (A little more reading tells us that these are fracture zones in limestone where lots of cenotes have formed.) 

Naturally, I want to know HOW this ring of cenotes formed.  

My query: 

     [ origin of the ring of cenotes ] 

took me to the Chicxulub Crater Wikipedia page which tells us that "A team of California researchers including Kevin Pope, Adriana Ocampo, and Charles Duller, surveying regional satellite images in 1996, found a cenote (sinkhole) ring centered on Chicxulub ... the sinkholes were thought to be caused by subsidence of the impact crater wall.  More recent evidence suggests the actual crater is 300 km (190 mi) wide, and the 180 km ring is in fact an inner wall of it..."  (reference: "The Cretaceous-Tertiary Impact Crater and the Cosmic Projectile that Produced It" Annals of the New York Academy of Sciences. 822: 353–80) 

That makes a great deal of sense.  A giant meteor strikes just off the coast of Yucután and causing an impact crater in a heavily limestone area. And by the way, killing off all of the dinosaurs as a side-effect!)  That could lead to caves forming, and ultimately their roof collapse, forming cenotes.  

So, yes, there are multiple patterns in the cenotes.  On the east coast of Quintana Roo, you see them lining up along the fracture lines of the Holbox and Rio Hondo fracture zones.  And amazingly, in the west, you see a giant 300 km wide ring of cenotes, caused (in some way) by the Chicxulub meteor.  

My friend, playwright Ross Nelson, happened to post this remarkable article about the Chicxulub meteor as I was writing this blog post.  

In it, the author argues that the meteor wiped out the dinosaurs not just because it was a large meteor, but because it hit the largely limestone (calcium carbonate)  and gypsum (calcium sulphate) rock layers of the Yucatán peninsula.  This mixture, when heated by a large amount of energy (such as from a meteor impact), would cause the creation of a vast amount of sulfur dioxide (with the sulfur coming from the melted gypsum).  

Sulphur dioxide then slowly reacts in the atmosphere to form a haze of sulphuric acid droplets, scattering sunlight and cooling the Earth. That haze, rather than simply dust, is how the Mt Pinatubo eruption affected climate.

The estimate is that the haze produced by the Chicxulub impact would have blocked enough sunlight to reduce temperatures worldwide by more than 10 degrees C (18oF).  

The surprising claim of the article is that if the meteor were 10 minutes earlier, or 10 minutes later, it would have fallen into the open ocean, boiling a lot of fish (and probably causing other effects), but perhaps without creating all of that sulfur dioxide haze, and consequently all of that cooling.  

Hard to know what the effects of a Chicxulub meteor would have had in the open ocean, but maybe not quite like this... 

2.  As you know, Cancun is on the eastern side of the Yucatán, in the state of Quintana Roo.  That name--Quintana Roo--has always struck me as slightly odd. Where's this name from, and why does it sound so non-Spanish?  (Extra credit:  How do you pronounce "Quintana Roo"?  While there, I learned I've been saying it wrong all these years!)  
Finding this out isn't terribly hard--the query:  [ Quintana Roo ] gives multiple sources telling us that Quintana Roo was made a territory of Mexico by decree of President Porfirio Díaz in 1902, and was named after an early patriot, lawyer and author, Andrés Eligio Quintana Roo.  He ran the Constitutional Assembly that drafted the Mexican Declaration of Independence in 1813.  

Tracking down the history of this name is a bit more complex.  Andrés Eligio Quintana Roo's father was Matías Quintana and his mother was María Ana Roo.  Searching for just her name:  [ María Ana Roo ] leads immediately to a genealogy site that gives a lot of background information.    Here we learn that María Ana Luisa Roo y Rodriguez de la Gala
was born in 1768 in Campeche (in the Yucatán), and that her parents were Antonio Roo y Fonte and Leonarda Rodriguez de la Gala y Torres.  Her father is from San Cristobal de la Laguna in the Canary Islands, and her mother is also from Campeche.  HOWEVER..  Antonio Roo y Fonte's father was Manuel de Roo Villareal.  

That name, "De Roo" is a Dutch or Flemish name, and looks like it was truncated to just Roo along the way.  We also know that there was significant immigration to the Canary Islands from both Holland and Belgium, so it looks like that's where the name comes from!  

How do you pronounce "Quintana Roo"?  The easiest way to find out is to search on YouTube for:  

     [ pronunciation of Quintana Roo ] 

which leads you to a nice YouTube video with someone saying the name.  Here, the title of the video is in Spanish ("Quintana Roo - Pronunciación en español"), which makes me feel more comfortable that it's correct:  

Reader Contribution:  Luckily, we also have a recording by Regular Reader Ramón, who sends us this MP3 with his pronunciation followed by his Mom's pronunciation.  (Thanks, Ramón!  I've been saying it wrong all these years!)  

3.  Speaking of Quintana Roo, when did it become a full-fledged state of the United Mexican States (Estados Unidos Mexicanos)?  
My query: 

     [ history of Quintana Roo ] 

... and then a lot of reading from multiple sources.  (Esp. the Wikipedia pages (both in English and Spanish--luckily, they mostly agree, although as usual, the Spanish language version has much more detail. I also drew from the Nations Encyclopedia.)    

The area that makes up modern Quintana Roo has a shared history with the Yucatán peninsula.  In the 1840s,  the Caste War of Yucatán, all non-natives were driven out of the Yucatán, leading to the foundation of the independent Maya nation of Chan Santa Cruz. For decades it was a separate country, with its own trade and foreign policy.  

Quintana Roo was declared the territory of Quintana Roo by decree of President Porfirio Díaz on November 24, 1902. After several uprisings, the Mexican army succeeded in defeating most of the Maya population of the region during the 1910s (during the time of the Mexican Revolution). In 1915 the area was again declared to be legally part of the state of Yucatán, although still as a separate entity.  

Quintana Roo was granted statehood within the United Mexican States on October 8, 1974, and is the youngest state in Mexico.

4.  While walking around, I found a tree (apparently native) that is said to have been the basis for chewing gum.  Really?  What kind of tree is this? What's the story here?   
Working from what we have, 

     [ chewing gum tree Yucatan ] 

which leads to multiple articles about chicle and the Sapodilla tree. 

Manilkara zapota, commonly known as the sapodilla is a long-lived, evergreen tree native to southern Mexico, Central America and the Caribbean.

An interesting story (and high quality source) was an NPR story about the history of chicle, chewing gum, and sapodilla trees.  In there, we learned that chicle is the latex (sap) of the sapodilla tree, and that Quintana Roo had the largest number of sapodilla trees, effectively the world's production of the basis of chewing gum.  

It's a great story, well worth a read (or a listen).  It has great characters like Antonio Lopez de Santa Anna, the 11-time president of Mexico, who worked with a New Yorker to develop chicle as a replacement for rubber.  He did this while he was exiled in the US trying to raise money to fund an army for a return to power. But using chicle as a rubber replacement never worked, but it turned out that it could be used as chewing gum!  (An excellent book to read is Jennifer Mathews Chicle: The Chewing Gum of the Americas, From the Ancient Maya to William Wrigley (2009).  Check out chapter 2 for even more amazing stories about the sapodillo tree.  

Santa Anna was also, by the way, that Santa Anna who led the battle at the Alamo.  

So, yes, chicle is the basis of chewing gum.  Or it was until the 1960s, when most chewing gum companies switched from using chicle to butadiene-based synthetic rubber which is cheaper to manufacture.  (In a strange twist of fate, Santa Anna tried to develop chicle as a rubber replacement; but now rubber has replace chicle in chewing gum...)  

Search Lessons

There are a lot here, and I'll continue this commentary tomorrow.  But let's start with a few obvious ones... 

1.  Take good notes!  I made a big mistake by not keeping decent notes as I was working on this, which led me to have to back up and repeat a bunch of searches... all because I didn't write down the queries and the results that later proved to be useful.  Notetaking is a powerful skill--especially for people doing research. 

2.  Know when to use INURL:  Recall that sometimes links to pages go out of date.  They usually aren't deleted, but they ARE frequently moved around (e.g., the web page example above).  To find the new location of this page, I had to do this search:  [ inurl:chicxulub.htm] -- that site: limits the search to the site and the inurl:  looks for webpages with the name of the page in the URL, chicxulub.htm - the actual page was still on the site, just elsewhere.  

3.  Finding missing data is hard... so think of another way to frame your queries.  In the case of the apparently "missing" cenotes, I had to search for the names of the other states (Yucatán and Campeche) in order to find much of anything.  It was also handy to get a single cenote (near the town of Miguel Colorado), and then go check Google Maps (or Earth) to see what the ground truth is.  

4.  Remember YouTube as a source for "physical" information.  In looking for the pronunciation of Quintana Roo, you can look for the International Phonetic Alphabet version (kin.ta.naˈro.(o) - but remember that this is IPA for Spanish).  In such cases, it's easier to find someone to just DO it in a video.  

More tomorrow... in our next exciting chapter of SearchResearch! 

Search on. 

Monday, August 14, 2017

Detecting photo manipulation--the classic way

Before I write up the answer about the Yucatán.... 

I want to tell you about something fascinating I heard over the weekend. 

RadioLab is one of the best podcasts going.  It covers a broad range of topics in science, education, environment, health, and life-in-general.  Their podcasts are incredibly well written and produced.  To my mind, they set the standard for what a great podcast can be.  

Usually I listen to their casts as I'm running errands on the weekend. On Sunday, I listened to a RadioLab podcast about a search process that's of real interest to SRS readers.  

Truth and Cannonballs (22 mins) is about the quest of documentary filmmaker Errol Morris to understand how a famous pair of images from the Crimean War (by 19th century photographer Roger Fenton) were manipulated.

Here's the pair of photos side-by-side:  

It's pretty clear that at least one of these pictures was manipulated by moving the cannonballs around.  The question that consumed Morris was "which of these was first?"  

That is, did Fenton first photograph the road WITH the cannonballs, and then move them away--OR--was the road empty, and he placed the cannonballs there for photographic effect?  

The podcast has a great discussion about what motivates someone to pursue a SearchResearch question like this, and then what he did in order to figure it out.  

tl;dr -- a colleague used a version of the blink comparator (that we discussed a few weeks ago) to find some stones that moved between the two different versions.  That was enough to say with high confidence that clear-road version was first.  (The insight was to realize that the stones all moved downhill, suggesting that they were accidentally kicked as the photographer moved the cannonballs into place.  After all, it's very unlikely that all of the stones would move uphill!)  

Listen, and enjoy. 

Wednesday, August 9, 2017

SearchResearch Challenge (8/9/17): Questions about the Yucatán? (#1000)


Yes, this is SearchResearch post #1000.  Remarkably enough, we've been talking about search, sensemaking, and how to be better at this for just over 7 years, since January 31, 2010.  In that first post, I wrote about a method for monitoring a web page on a topic.  (Of course, the method I describe there no longer works as written... which is why this blog has been ongoing for so long!)  

In my next post, I'll write a bit about the process of writing SRS, and where it's going.  There may be a thousand posts under our collective belts, but there's still so much more to do!  


Onward... to this week's Challenge... 

Last week I was in Cancun, Mexico, attending a friend's wedding, enjoying the beaches, and (naturally) slipping in a quick dive trip on the side.  

This is a picture from our dive at one of the Dos Ojos cenotes

As you can see, a cenote is a sinkhole exposing groundwater below. Cenotes were sometimes used by the ancient Maya for sacrificial offerings, but we were there to explore the caves.  There were four of us, following the yellow trail marker line into the heart of the cenote, swimming through impossibly clear water.  At times, it seemed we were flying through a fantastic vast cavern, rather than swimming along well underground.  

Naturally, all of this made me very curious about cenotes, the Yucatán, and the whole place in general.  Here are a few of the questions I had (and answered!) this week.  

Can you answer them as well? 

1.  Cenotes seem to appear all over the Yucatán peninsula.  If you look at a map of the area, it seems they all line up just inland from the Riviera Maya coastline.  But I know there are cenotes in the north of the peninsula as well. Is there a larger pattern of cenotes at work here?  If so, what caused that particular pattern of cenotes to form? 
2.  As you know, Cancun is on the eastern side of the Yucatán, in the state of Quintana Roo.  That name--Quintana Roo--has always struck me as slightly odd. Where's this name from, and why does it sound so non-Spanish?  (Extra credit:  How do you pronounce "Quintana Roo"?  While there, I learned I've been saying it wrong all these years!)  
3.  Speaking of Quintana Roo, when did it become a full-fledged state of the United Mexican States (Estados Unidos Mexicanos)?  
4.  While walking around, I found a tree (apparently native) that is said to have been the basis for chewing gum.  Really?  What kind of tree is this? What's the story here?   

As you search, take note of HOW you found the answers, and let us know in the comment thread!  

¡Sigue buscando! 

(Search on!)  

Tuesday, August 8, 2017

The difficulty of searching for something dimly remembered...

I saw it, but only briefly.... 

Now I want to refind it.  But how? 

A true story:  A few days ago I saw an interesting article in my Facebook feed.  It was one that I distinctly remember as something I wanted to find and read in detail later. But I was busy, so I skipped over it, and figured that I’d be able to find it easily later. Maybe you've done this as well.  

But then....  it took me 90 minutes to RE-find it.  What happened? 

What I remembered about the article:  I recalled that it was a post by The Atlantic magazine (which I follow on FB), and that it was about how millennials (or Gen Z, I forget which) now believe that “there is a great deal of information that cannot be found on the internet.”   That was the phrase I read and remembered. I thought my memory was pretty good, so searching on Facebook should just work, right? 

But the obvious search doesn't turn up anything.   (This is true even if I limit the search to a particular source, in this case, "The Atlantic" stream by entering the name of the publication in the "Posted by" option.)  

However, I did a LOT of searches, limiting the time scope to the past month, site searching the magazine, searching FB… couldn’t find it. 

What was going wrong?  

Short answer:  I was trying too hard with long queries.  I should have known better... 

To actually find the article, I went back through my feed MANUALLY, one post at a time working my way back in time.  Of course,  I finally found it, and here’s what the post actually looked like:

Now, WHY couldn't I find it except by brute force? 

Because my searches were close, but not precise enough.  They were all variations on this Facebook search:

     On Facebook:  [ students “important information” not on the internet ] 

 ... and all obvious variations don’t work.  Note that all of these words appear in the post. 

 Hmm.  This search should have worked.  

Insight #1: Search on Facebook is picky.  You have to give it exactly what you know is there, and nothing more. 
Insight #2: You need to have a fairly precise search to find a specific article that you’re looking for on Facebook.  

Ultimately, when I found the story (by repetitively clicking), it linked to "Millenials are Out-Reading Older Generations," which is well worth reading.  

After I found it, I spent some time backing up and trying to use regular Google searches that would work to find this article by searching just The Atlantic site.  Here are the Google searches that worked for me:

    [ students “important information” “not available on the internet” ] – 1 result

     [ students “important information” “not * on the internet” ] – 1 result

     [ “important information” “not * on the internet” ]  - 1 result

     [  important information “not * on the internet” ] – result #2

     [“important information” not on the internet ] result #3

     [  information “not * on the internet” ] – result #5

     [ “not * on the internet” ] – result #8

Notice that a lot of variations on this theme worked just fine.  With quotes, without quotes, using the * operator.  

The important thing to notice here is that what works on Google might not work on Facebook search.  

Another thing worth noticing, the shorter queries work just fine.  Keep this in mind.  

In the interest of fairness, I went back to Facebook search with the article in hand and re-did my search with an exact quote.  Naturally, it was the first (and only) hit: 

How is it possible that I didn't try THIS particular variation of the query?  Methinks it's because I was trying with too many terms... I was working too hard.  

The Facebook query I SHOULD have tried at first is just with the single term that I was really sure was in the article--the relatively rare term "millennials."  

This is a much better search strategy.  It's doesn't give me the result I want in position 1... but it's at position 5 in the Facebook search results, which is pretty good.  This strategy would have given me the result in less than 1 minute, rather than the 90 minutes I actually spent!  

... scroll down to the next page to see result #5 


There are a few Search Lessons here: 

1.  Different sites have different search behaviors.  Facebook's search works rather differently than Google.  You have to understand the limits (and capabilities) of each.  (For instance, Google can't search the posts on Facebook.)  

2. Being persistent means trying variations on a theme... and trying different resources.  I should have followed my own advice--advice I've given many times before... 

3.  Start as simple as possible, then add keywords as you discover what's not working.  

4.  Sometimes, the best way to search on a site is by using Google. As you can see, the Google queries are pretty robust--many of the query variations return exactly what you're looking for.  (This is a general comment: it's not just limited to Facebook.)  

Keep searching!