Friday, October 25, 2019

A new "How Google Search Works" video


I don't often blog about new YouTube videos, but when I do, there's a reason.
After showing the long-lived and beloved Matt Cutts video "How Search Works," Google has finally released a few version of that brilliant explanatory video.
So, this here's my Public Service Announcement for teachers who teach search research skills or information literacy more generally.
Worth a view if you're looking for a 5 minute summary of how Google search works (at least as of 2019).
(Link to the video for use in your classes: https://youtu.be/0eKVizvYSUQ )

Search on!


Wednesday, October 23, 2019

SearchResearch Challenge (10/23/19): Where is this lobster image from? What's the story behind it?


I sometimes find things... 

... in my files that are funny and odd... and I have no clue about what they are, how they got there, or what the backstory is.  

After last week's epic SearchResearch Challenge, I was cleaning up my files. (Yes, I admit it.  I tend to drop files all over my desktop in a huge disorganized pile, and then clean things up afterwards.)  

In the cleanup process, I found this cute little lobster image: 


Obviously, this isn't a lobster I know about and it has nothing to do  with the contour lines from last week, so it must have appeared at some point in the past.  This looks to be some kind of older document.  

I've got lots of lobster photos from dive trips I've taken (here's an example).  


But where did that first lobster  image come from?  

1.  Where is this lobster image from? 

2.  What's the backstory of the source?  (IIRC, it has a kind of crazy story.  But I don't remember what it is!) 

Can you enlighten me about this lobster image and the tale of the source that goes with it? 

As always, let us know how you found the answer!  

Search on!  

(After last week's Challenge I pre-tested this.  I was able to find the answer in less than 5 minutes, although the backstory really is crazy!) 


Tuesday, October 22, 2019

Answer: Why are there contour lines there?

Questions come from everywhere... 

... in this case, I was just staring out the window and wondering "how did all of this landscape come to be?"   

As I said, when I looked into the Sierra foothills  just northeast of San Francisco, it seemed as though some giant hand had drawn contour lines on the hills below. 


That struck me as very surprising.  How could this happen?  It made me curious.  



How could I be seeing what looks like contour lines in reality...?  To remind you, here's what I saw.  The green wavy lines are clearly trees or shrubs that seem to be following some kind of constant edge through the hills.  


Alas, I didn't have my camera handy.  This is an image from Google Maps of the place I
saw out the window.  It really does look like this!

The Terrain view of this location is really interesting.  





If you put these side-by-side, they're amazingly similar. 




This observation led to our SearchResearch Challenge for the week: 


1.  What is going on with those naturally green "contour lines" that seem to show up naturally?  Why are they growing along the lines of constant elevation?  What is going on?

The first thing we need to do is to figure out what this place is called.  When you look at it from the air, there aren't any handy labels.  Even on the map, the only nearby named things are  "Wells Place" and a few creek names, most notably "Mill Creek" (as seen above).  

If you zoom out a bit, you'll quickly figure out that this patch of land is halfway between the city of Red Bluff and the Ishi Wilderness (if you keep zooming out, you'll find that this is the southwestern-most part of Lassen Volcanic National Park).  And if you zoom IN a bit, you'll find that the Lassen Trail passes through the bottom part of the photo.  This is a great overview of the entire area:  


Terrain view of the area between Red Bluff and the Ishi Wilderness

So when I started searching, I used the local terms I know and can see on the map of the area ("Ishi Wilderness" "Lassen Trail" "Mill Creek").  

But what am I searching for?  

I started with the obvious terms and did searches like: 

     [ arc-shaped vegetation "Mill Creek" ] 
     [ crescent plants hillside "Ishi Wilderness" ] 

and so on.  Trouble was, none of this was working for me.  I got lots of hits, but nothing that was useful in any real way.  I can SEE these things, but I have no idea what they're called.  (Or even, really, what this place is called--I'm looking for something that's between several named entities. An unnamed region is pretty hard to search for!)  

But as I was doing this, I realized that the last west of the Ishi Wilderness (that is, in the middle of our search area) is fairly covered in springs.  Since springs mean water, and since water leads to green growing things, I wondered if there might not be a connection.  Are all of those springs somehow making green things grow in arcs that follow contour lines?  

My query was perhaps not obvious, but works quite well: 

     [ springs Tehama county ] 

Why Tehama County?  I know that in the US, the United State Geological Survey (USGS) often organize their content (e.g. lists of geological features) by county.  A quick check of Google maps showed me that the boundary of  Tehama County included our interesting location  at 40.081801, -121.903088 (link to Google Maps at this location).   



See Wells Place just to the right of the city of Red Bluff?  That's in the middle of what we're looking for.  So, after I did that query, I found a list of Tehama County springs.

With all of those springs, I made a quick map of some Tehama County springs using Google's My Maps service.  (link to my map)



What I found so interesting about this map is that there's a long line of springs uphill from the green contour-following arcs of green.  Remember how they seem to follow along the edge of the creeks?  

Here's the view of Mill Creek we saw earlier.  


This is a closeup view of Mill Creek (40.090698, -121.907125) with clear arc-shapes
that seem to follow a constant elevation. 


Here's another view of the same thing as seen in 3D Google Maps.  This is a 3D view of the satellite image from above.  (You can make this first clicking on the 3D button in the right-hand side control panel, and then title by holding down the Control-key while moving the mouse up and down to tilt the surface of the Earth.)  



From this perspective, it's clear that those arcs of bushes are really more-or-less horizontal lines.  They look a lot like bedding planes that you might see in a canyon cross-cut.  Here's an example from the Grand Canyon--bedding planes are the layers of rock you see in the canyon wall.  


Grand Canyon cross-section showing bedding planes.  Note how plants seem to grow
only in certain horizontal sections, and not on others.   P/C Dan. 
It's pretty clear that in canyons like the Grand one, the river cuts a cross section across the bedding planes exposing places where you can see the structure of the rock layers.  It's like you've cut a layer cake and can see the different layers, as well as the frosting.  



Here's a diagram I drew to make this really clear.  Instead of white frosting, here's a valley with the bedding planes / contours shown with red dots.  


Contour lines along the Mill Creek drainage.  (Hand-drawn by Dan.)  


Except  in this case, it looks like the contour lines are following the bedding planes... as do the green growing plants.  What if the bedding planes highlighted in red (on the sides) transport water?  

Putting two and two together, we know there are lots of springs in the area, but do they supply water to the bedding planes that then flows to the place where they're exposed in the creek valley?  

What do we search for now?  

I'll spare you the next few hours of searching.  I tried lots of variations of search terms having to do with bedding planes, hydrology, water flows, etc.  

But I didn't have much luck until I started looking for versions of this query:  

     [  geology Tuscan formation ] 
    [  hydrology Tuscan formation ] 

Where did "Tuscan Formation" come from?  

I  all of my reading on the topic, I  finally learned that this whole area (from east of Red Bluff to the Ishi Wilderness) is mostly covered in the "Tuscan Formation" (on the bottom) or the "Riverbank Formation" above.   

I first saw that term after I did a search for:  

     [ geology map Tehama County ] 

The thing is, once you go down that rabbit-hole, there's a LOT of things to learn about.  When you look at a map of the area, they often are extremely complicated--to wit--here's a section of the geological map near Mill Creek (this is of Cottonwood Creek, just a bit to the north of Mill Creek at 40.209742, -121.966270): 




Everything labeled with a Q-- (e.g., Qrb) are parts of the Riverbank Formation from different times--they're basically different layers that were laid down many years apart.  These are all parallel layers in the complex layer cake of the geography in this area.  

If you look carefully, you'll see that this map translates into this image: 



The green line is the watered foliage along the creekside--no surprise there.  But the arcs of green follow the Qrl / Qrb lines. That is, the different lines of layer boundaries.  The same story is repeated everywhere in this region.  When one formation contacts another, there's a dividing line, and water flows along that plane, coming out at the edge of the creek erosion, where plants can absorb it and grow.  

So now we have a pretty decent idea about what's going on.  

The green arcs really ARE on contour lines.  The contour lines are actually the places where two different layers of the formations meet.  Groundwater in present in these formations (as we saw in all of the springs uphill from this greenery), and slips between the layers until there's a river cut in the soil, exposing the water along the edge, following the contour lines exactly.  


Lessons Learned 

1.  Sometimes simple questions get very hard.  This was hard because I had to learn a lot on the way to be able to ASK a decent question.  I learned about the different kinds of springs (toespring, linear seep spring, etc.)  because I needed to understand what caused each. This really was a challenge because I had so much to learn!  

2.  Names matter!  Identifying a place name is important, especially for those places that don't really have a name.  In this case, the name "Tehama County" included the whole area, but it was incredibly useful as using that name let me find all kinds of resources, even if the area I was searching for was only a piece of the whole.  

3.  Great information resources can be overwhelming, especially when they're produced for professionals.  I'm not a professional geographer (or hydrologist), so when I see these immensely complex and detailed maps (e.g., the USGS Geologic Investigations Series Maps--here's the one  for the Tuscan Formation), the trick as a searcher is to NOT be intimidated.  Someone understands these things; you can too.  You might have to read a bit of background material (I certainly did), but it's not impossible.  



I have to admit that I spent MANY hours on this... probably 10 hours or so.  There was a lot of stuff to learn in order to even begin to understand what was going on here.  
In fact, I had SO many maps and guidebooks and documents open at once that I finally had to bring in a 3rd monitor!  This was my setup in my home office.  Luckily, second (and even third) monitors don't cost that much anymore.  And it vastly simplified what I was trying to do.   

As many other, wiser people have said before me... now that I've started this, I realize how much I really don't know.  
But I don't mind the hours spent.  I was learning fascinating stuff all the time.  It's a kind of graduate education that's difficult to get anywhere else.  
Hope you  enjoyed this quest as much as I did.  

Search on!  


Wednesday, October 9, 2019

SearchResearch Challenge (10/9/19): Why are there contour lines here?


As I was flying to Seattle... 

... I had the window seat that faced east as we flew north.  

Not long after we left SFO, I was staring out the window (because looking out an airplane window when you fly over wilderness is just the best thing you can do), and I saw the most remarkable thing.  


It looked like some giant hand had drawn all of the contour lines on the hills below. 


What???


As you know, contour lines on a topographic map mark a constant elevation.  For instance, here's a sample of a topo map with contour lines at 5400 feet, 5600 feet, etc.  As you can see, there's a steep cliff on the right side of the map, with a more relaxed flat spot on the left.  



Sample topographic map with contour lines.

But the idea of "contour lines" is an artificial way of looking at the land.  They're imaginary lines that are drawn to give the reader an idea about what's going on with landform.  

So you can imagine my surprise when I looked out my window and saw this, with what looks like contour lines in reality...  



Alas, I didn't have my camera handy.  This is an image from Google Maps of the place I
saw out the window.  It really does look like this!

When I got home to my laptop, I searched along the flight path and found the location.  (See above.)  

Interestingly, if you look at the Terrain view of this location, you'll see something remarkable: 





If you put these side-by-side, they're amazingly similar. 




The "contour lines" in the satellite photo are clearly lines of trees or bushes or something growing.  This amazes me.  I thought contour lines were an artificial construct, a visualization device mapmakers would draw on a map to give a sense of shape on an otherwise 2D image.  


But how is this possible?  Did some maniac cartographer go out there and plant bushes along the contour lines?  


This leads to our SearchResearch Challenge for the week: 


1.  What is going on with those naturally green "contour lines" that seem to show up naturally?  Why are they growing along the lines of constant elevation?  What is going on?


Any ideas?  


In truth, I have no idea what's going on here, and what's more, I don't have a great idea about how to find out!  We'll figure this one out together.  


One useful piece of information:  The center of this map is at 40.081801, -121.903088 (link to Google Maps at this location)  


When you figure it out, let us know the answer... AND tell us how you did it.  We learn from each other, so also be sure to tell us the paths you pursued that didn't work out.  



Search on!  


Thursday, October 3, 2019

Answer: Why are palm trees so tall?

Palm trees are wonderful, but very strange... 



As you might suspect, they're not ordinary trees, but something very, very different.  


They're generally super-tall, able to remain standing in hurricane force winds.  If you cut one down, you'll see a very strange and wonderful composite structure that looks nothing like an ordinary tree.  There are no tree rings, but a bundle of fibers that are key to its extraordinary resilience. 

P/C Wikimedia / Kadeve.
  
Our Challenge was:   

1.  Why are palm trees SO tall? 


As you know, asking Why Questions can be really difficult and tricky.  What's a good answer to a why question?  

To get a bit of background, I looked at the Wikipedia entry about palm trees, and quickly ended up on the entry for Arecacae, the latin family name.  There I learned that: 
The Arecaceae .. can be climbers, shrubs, tree-like or stemless plants.. There are 181 genera with around 2600 species are known, most of them restricted to tropical and subtropical climates. Palms are distinguished by their large, compound, evergreen leaves, known as fronds, arranged at the top of an unbranched stem. However, palms exhibit an enormous diversity in physical characteristics and inhabit nearly every type of habitat within their range, from rainforests to deserts...

I started this Challenge by doing an obvious search in the form of a question.  Note that asking questions like this triggers a special kind of Google search processing--it's not just searching for those terms, but the query is handled much more as a knowledge-based request... 

     [ why are palm trees so tall ] 

The search results are pretty good.  

(Click to see at full size.) 
The first 4 results are right on topic and give us a diversity of content.  Here's what I see and think when I look at these hits.  

#1 is a link to a Quora (a well-known question-answering site) discussion with a question about the evolutionary benefit for palms to grow so tall.  That's a great approach to answering the why question--understanding the costs and benefits from an evolutionary perspective would be good. 

#2 (skipping over the "People also ask" section) links to a reddit "Explain like I'm five" question/answer about "why are palm trees so tall?"  The explanation might be simple, but there's almost certainly an interesting discussion there. 

#3 is from Mother Nature News, a kind of gee-whiz site with breathless articles like "4 ways tardigrades are nearly indestructible," but might give us some interesting tidbits about tall palms.  

#4 links to a StackExchange forum, pointing to the more generic question "Why is it beneficial for trees to grow that tall?"  I expect this to be a more general discussion of tree height--perhaps we'll learn something about why trees grow so tall in the first place.  

I read the targets of these links and found out that: 

* Not all palm trees are tall!  (In retrospect, this is obvious--different species of palms have different heights. For instance, the Allagoptera arenaria  (Beach palm) is less than 2 meters high.  But clearly, we're curious about tall palm in this Challenge.)  

* Palm trees in their wild and natural setting often compete for resources.  In the wild, palm forests are often densely packed, requiring the palm trees to do something to grab their own light, water, and nutrients.  Growing extremely tall is one solution.  

Here are a few images of wild palm tree forests. You can see there's a lot of competition for sun and water.  

Eastern San Diego county, packed into the bottom of a dry ravine. 

A palm forest in Indonesia.

Even beach locations can be competitive! (Image by Pexels from Pixabay.) 
(I note that it's a little tricky to find images of palm trees in their wild and unstructured settings.  Many palm trees, even dense forests, are often former coconut plantains, which isn't the same.)  

Result #2 tells us that palms are often the fastest growing trees (although as with palm tree height, growth rates vary from species to species).  So they compete in height, and rate of growth in order to get the resources they need.  

Meanwhile, #3 tells us that the tallest palms are the Quindio wax palm (Ceroxylon quindiuense), growing up to 60 meters (180 feet), which Wikipedia tells us grows in dense forests in the wild--so height is important there as well.  

And #4 asks the more general question, "Why is it beneficial for trees to grow that tall?" Keep in mind that this is a discussion on a StackExchange site that encourages experts to answer and discuss questions.  It's heavily moderated, and the quality of the discussions you find there is pretty high.  

This particular thread discusses why palms in forests grow so tall, and work through various alternative explanations... but it all comes back a fitness advantage for taller trees to have more sunlight.  

An interesting twist...  Just for grins (and because I know that shifting media types sometimes gives an insight), I did a search for: 

     [ palm tree height ] 

and looked at Images.  It was pretty much what you'd expect.  Looked like this in the center of the SERP: 


That scatter plot chart in the middle made me think--perhaps there's something interesting here! 

Turns out that this chart comes from a scientific paper about the age and height of oil palm trees, Tree height and crown shape as results of competitive games (J. of Theoretical Biology, January 1985) and that made me think about doing a search in Google Scholar.  

In Google Scholar: 

     [ palm tree height ] 

led to a bunch of fascinating papers (which time and space prevent me from summarizing, but there's a fun intellectual rathole to explore one day).  

But the paper Competition from below for light and nutrients shifts productivity among tropical species
seemed to potentially hold the answer to our question.  Turns out that it didn't... exactly... but it DOES make the fascinating observation that 

"...In 2 cases the novel competitive mechanism responsible for the shift was reduction in crown volume, and therefore light-capturing capability, of overtopping deciduous trees by intrusive growth from below a palm." 

Which kind of captures what we found elsewhere.  

Why are palm trees so tall?  Answer: Palms compete for light by growing tall and fast.  In this case, they overreach the (ordinary) deciduous trees by growing up and through the canopy to reach the pure sunlight above the shade cover of the deciduous trees.  But in palm forests they're competing with their peers...  where they compete just as hard.  

Remarkable. 

Search Lessons 


This Challenge points out a couple of lessons to learn and take to heart. 

1. Looking across a number of different sources is valuable.  I know I keep saying this, but as a skilled SRS-er, do NOT lock in on any single result, especially if it confirms your beliefs.  A better strategy is to look broadly across a number of results and look for insights that are reported consistently across a number of different authors, different sources, and different perspectives.  That's one way to find your way to truth.  

2. Try different sources to get a different perspective.  Here I did another (but related) search on Google Scholar (after having been prompted by seeing a scientific chart in an image), and found lots of high-quality (but sometimes dense) articles on palm trees and their growth behaviors.  

Hope you enjoyed this romp through palm tree botany.  As always, there's a LOT more to say about this topic.  (If you're interested, a great query is [ varieties of palm trees ] -- they're an amazing group of plants with wildly varying shapes, sizes, and niches.  As they say, worth a trip...)  


Search on!  

Wednesday, October 2, 2019

Slight delay... I'm giving a talk at a librarian's conference today!


Strangely, 

... palm trees line the Embarcadero, the grand boulevard that wraps around the top of San Francisco.  


I'm not sure how they manage it here--it's cool and often foggy, nothing like the deserts or tropical isles where palms are native.  

But they do survive.  Barely.  

I'm here in San Francisco today to give a talk entitled The Nature of Literacy and Learning in the Future.  You can see the program for the Pacific Library Partnership conference here.  

You won't be surprised when I say that learning how to do effective research is a critical skill.  

So I'll be at the San Francisco Public Library today, enjoying chatting with librarians and researchers.  

And as a side-effect, I'll write up my ruminations on palm trees tomorrow.  

Search on!