Answer: What is that thing in the sky?

First off... it's not a UFO, as a couple of people suggested!


As I mentioned yesterday, finding things like this--objects or concepts that you don't already know the name for--is pretty hard.  And this is no exception.


If you do the obvious search:  [ bright spot sky  ] none of the first page of results is useful.  While starting with a few words is often a good strategy, the query [ bright spot sky ] is a bit too generic.  In essence, you haven't said much that the search engine can get its hands on.  This query is just a few words, and not any usefully descriptive phrase.


If you add just one word to make it a slightly more descriptive phrase, a la  [ bright spot in sky ] you'll find something useful way down in the 10th result (way at the bottom of the page).  There you see the first reference to a term you might not recognize: sun dog. 

The last result mentions "sun dog"--could that be it? 

All of the other results are about astronomy and the night sky.  Unfortunately, the link is to a video of a rocket liftoff showing shock waves in an atmospheric ice crystal layer.  More about that later, but it doesn't seem directly relevant.


Except for that term "sun dog"--why is it there?


A quick check with define, as in  [ define:sun dog ] teaches us that a "sun dog" is "an atmospheric phenomenon that creates bright spots of light in the sky, often on a luminous ring or halo on either side of the sun"  (Google found this definition on Wikipedia).


Now that we've learned this likely sounding term, we can go to Google Images and look for pictures of [ sun dog ]

Image from Google Images (3/18/2010) 

This second image looks so much like the picture that I took with my phone that I'm pretty confident that this really IS a sun dog.  A check on Wikipedia (query for [ sun dog ]) shows a pretty nice article with multiple images of sun dogs.


And that's what I saw on Tuesday--light reflecting off of a high, thin layer of ice crystals--a sun dog that's caused by sunlight reflecting off the ice crystals way up high.  Often they're invisible.  (But that 10th result from the query [ bright spot in sky ] is a beautiful short video of a rocket launch.  The shock wave as the booster goes supersonic is made visible by the rippling of the ice crystals as the pressure waves ripple the crystals, just like rippling water from a stone tossed into the still pond.)


Interestingly, while that video wasn't useful in and of itself, the text of the snippet suggested that there maybe was a term we could use in OUR search.


That turned out to be the trick.


As you search, you learn.  As you learn, you can improve your search targeting by using terms that you pick up along the way.   


Keep learning.  Keep searching. 

Wednesday Search Challenge (March 17, 2010): What is it?

One of the most difficult things to discover is what something is called when you don't know its name....

This is actually a double-edged sword.  Because you don't know what something is, you might very well miss noticing it.  Yeah, it's true--you tend to miss things if they're not familiar or have a name.  This is a really interesting distinction--because you can SEE something, but then when you look away, you'll have a very hard time remembering all the things you saw that don't have convenient names!

Here's a real-life thing that happened to me yesterday.  I was out riding my bike in the hills near home when I turned the corner and saw this:

This isn't the best picture (it was taken with my Droid phone), but it shows that bright patch of sky to the right of the sun with a little bit of a bright arc.

As it happens, I now what this is.  Perhaps you do too... but pretend for a minute that you don't actually know what this is.

How would you find out more about it?  What's this thing called?  And... how do you learn more about this phenomenon??

Everything you need to know in order to figure it out is right here in this photo.

Search on!

Making sense of conflicting information: the case of the wild radish

One of my interests in life is understanding the world around me.  When I moved to the San Francisco Bay Area, I was initially astonished at how different the botanical world was all around me.  I'd look at the hillsides and have NO idea what I was looking at!  Low flowering plants, trees, bushes... all of this amazingly beautiful stuff, and I didn't know a thing. 


So I did what I like to do... I started learning.  Went to the bookstore and picked up a few wildflower nature guides and identification keys, a sketch pad, a few colored pencils and went to the local hills and started figuring out what flowers were what.  I'd work through the keys (eventually learning what an "inferior ovary" was, and learning how to count stamens).  And slowly, laboriously, I started figuring it out.  


It took me a couple of years (literally!) to figure out 90% of the flowers, and I'm still working on tree identification, but the super-abundance of wildflowers is still my favorite thing to look for in the springtime.  


But this past Saturday I was struck by something really odd.  


Wild radish (Raphanus raphanistrum) is a very, very common flowering plant in the area.  It blooms very early in the spring, and grows nearly everywhere. 

It's in the Brassicaceae family, which means it's related to mustard, broccoli, cabbage (and all "cruciferous" vegetables--so-called because their flowers are cross-shaped).  


But I was seeing it with many different colored flowers.  Some were white, some were yellow, others blue and some even a kind of tawny-gold color. 


Huh?  Were these all the same flower?  Or were there four different species, all of which looked like wild radish?  


I spent a few minutes looking around.  The first place I checked was the Wikipedia article on wild radish, which isn't a bad summary, although a bit short.  


The next thing I did was a regular Google search on [ Raphanus raphanistrum color ] and discovered a pile of articles in Google Scholar (which contains many, many journal articles written by professional botanists). 


{ Why did I use the Latin name of the flower?  Largely because I wanted authoritative articles on the subject.  And I know that the Latin name is used by botanists and horticulturalists, that is, people who would know what's going on.} 


Now, to back up a second to one of the points of this blog... 


Sensemaking, as I think about it, is what people do when they're trying to understand something fairly complicated.  It's the combination of searching information out and then re-organizing what they've found in order to come to a deeper understanding.  


And that's what I started to do with respect to this question of "do wild radishes really come in more than one color?"  


I started with Wikipedia, which told me that they vary "...colour, usually from white to purple but sometimes light orange to yellow, often with colour shading within a single petal"  


Wikipedia also told me that wild radish is sometimes called "Jointed charlock" (a name I'd never heard before), but seeing that also told me that using the Latin name would be the right way to go on my next search.  (That's the problem with wildflowers--they ALL have multiple names!)  


Now, back to the Google Scholar results.  I started looking at the various papers that Scholar found.  The first one I clicked on was "Reproductive biology of petal color variants in wild populations of Raphanus sativus" by a botanist at UC Davis.  I know UC Davis is a good agriculture school with a strong botany department, so I'd tend to believe most of what I find from them.  


In that paper, the author writes that the flowers vary in color from "white, yellow, pink and bronze."  Bronze?  Ah, that's the technical term for what I'd called "tawny-gold."  I also learned that honeybees are the most common pollinators, along with syrphid flies in second place.  ("And what are they?" I think.  Oh ... save it for later.)  


But you see what's happening?  I'm learning more and more from every article I read about wild radishes.  I'd already confirmed my hypothesis about color variations, and now I'm starting to learn why and how such variations occur.  


The article I'm reading now ("Reproductive biology of petal color variants...") is suggesting that the color variations are what drives individual honeybees to prefer one color over another.  (Pink flowers only want to be pollinated by other pink flowers?) 


The next article I look at implies that petal color is determined by one gene site, and that no pure, pre-European contact versions of the wild radish exist... they've all been hybridized by now!  


I could go on.  This turned into one of those diversions that leads to an immense amount of background knowledge about local plants (which is one of the characteristics of people who spend hours identifying plants with a plant key in hand).  


But you see what I'm getting at:  Sensemaking often proceeds from a clearly defined goal ("do wild radishes have color variants in their flowers?"), towards something less defined  ("what's the value of color variants on the fitness of the plant?").  If all you wanted to do was to answer the question, I could have done that on Wikipedia and been done with it.  


Often, though, the beginning question leads to a richer, more nuanced, more informed version of the question.  Sensemaking is often also about how to reformulate your question into something that's richer and deeper, usually by learning about the topic area as you proceed.  


That's what happened in this case.  I now know a lot more about why flowers might want to have variations on their flower colors.  And that's probably knowledge I'll use again in the future.  


I'll see if I can't get some decent pictures of the color variations.  If I succeed, I'll post them here. 

Answer: Alaskan sled dog questions

Yesterday I posed a couple of related questions.  The first was "What's the average weight of an Alaskan husky running in the Iditarod in 2010?"  


This is pretty simple.  The query [ Iditarod sled dog ] brings you quickly to the Wikipedia page, and the weight range of 45 - 66 pounds (20 - 25kg).



But the second question is a bit trickier.  "Who actually handed the vaccine to the doctors in Nome in 1925?"  


It's pretty well known that Leonard Seppala and his lead dog Togo were involved in delivering diptheria serum, but it turns out that Gunnar Kaasen was actually the last musher into Nome with the diptheria serum


How did I figure that out?  


It's not hard to figure out from the query  [ Iditarod history ] that Seppala was somehow involved in the race.  (Just read any article about the Iditarod, and that will give you his name.)  


But if you then read any of the many bios of Seppala on the web CAREFULLY (such as the Wikipedia entry), you'll find out that he actually wasn't the last guy in the chain of dog sled teams to Nome, but was the musher who went the longest distance through the most brutal weather.  (Not that any of the trip was a milk run.) 


So... The man who handed off the canister with the serum to the doctor in Nome was Gunnar Kaasen, who made the last leg of the journey with lead dog Balto.  (Oddly enough, Balto was actually owned by Seppala!)

And... for extra credit, how much do you think HIS dogs weighed, on average?


It turns out that Kaasen's team was all Siberian Huskies.  They weigh between 45-60 pounds--about the same as the Alaskan huskies.  




The teaching point for this challenge is that it's often very easy to assume you know more than you really do.  The Iditarod race celebrates Seppala's heroic mushing effort in 1925--but you shouldn't assume he was the hero who actually rode into Nome.  Be careful, and although your mother always tells you this, it's really true: check your facts!  

Wednesday Search Challenge (March 10, 2010) -

As I mentioned, I spent last weekend in Anchorage, giving a talk at the Alaskan Librarians Association meeting.   During that same weekend, the Iditarod dog sled race started from just down the street.  And that gives us our challenge for today.

Quick:  What's the average weight of an Alaskan husky running in the Iditarod in 2010?

As you probably know, the Iditarod dog sled race commemorates an amazing feat of outdoorsmanship--the transport of a 21 pound container of diptheria vaccine to Nome in order to quell a potentially lethal outbreak.

But if that first question was too easy for you, let's try a harder one.

Can you figure out who actually handed the vaccine to the doctors in Nome in 1925?

And... for extra credit, how much do you think HIS dogs weighed, on average?

Mush on!

A few days in Anchorage at the Alaska Librarians conference: perceptions about search skills

As I mentioned yesterday, I spent a few days at the Alaskan Librarians conference (AKLA 2010).  It's the annual gathering of the librarian tribe, and this time it coincided with the start of the Iditarod trans-Alaska sled dog race.   (Tomorrow:  A search challenge from the Iditarod!)

During the conference, I had the chance to talk with lots of librarians about what they see their patrons searching for (or not searching for), and what their particular challenges are.

One thing that came across clearly was that most people think of themselves as being pretty good searchers, but in fact, this seems to be a bit of an illusion.  People's perception of the skill of searching (particularly their individual skill) is fairly positive.  Many times you'll hear people express complete confidence:  "I can find anything that's on the web!"

While that's great, it's also not quite true.  For the most part, the search tasks that people tackle are reasonably straight-forward: What time is the movie at the megaplex tonight?  What is the population of Japan?  Why are Alaskan husky dogs in the Iditarod so small?  (FWIW, they're much smaller than I'd expected!)

For these kinds of search tasks, the web has a large information base that's straightforward to locate.

But the really hard questions--the kinds we're interested in for our purposes--are a bit harder to answer.

I sat for a few hours at the University of Alaska library reference desk and observed as the reference librarian answered questions ranging from the simple ("Do you have a copy of 'The Great Gatsby'?")  to the more difficult.

U. Alaska Anchorage:  Great Room at the library

I find these sitting-at-the-reference-desk observations to be invaluable for my studies.  While the range of reference questions you see at the desk is still broad, some of the harder questions are genuinely tough to answer, requiring the best skills of the professional researcher... and the educational knack of a skilled teacher.

One patron asked for a "ranked list of the top sociology journals in the world."  That sounds as though it should be simple--but the hard bit here is defining what "top journal" means.  The reference librarian did a masterful job of trying to elicit what the patron really wanted--does "top journal" mean by subscription volume?  By citation counts?  By number of appearances in the popular press?  By number of issues published each year?

Any of those are possible to figure out, but to answer the patron's question (or by extension, their Google search) they need to understand what tradeoffs and choices are possible.

That's the function of the reference interview, to drill down into what the patron really REALLY wants, as opposed to what they say they want.  As my clinical psychologist friends often say, "the presenting problem is rarely the real problem."

But knowing how to drill down and identify the shape and nature of information that actually will satisfy the patron's request requires broad knowledge about the world, about what resources are available and more than a bit of human psychology.

The patron who comes in to ask for "everything about the civil war" definitely requires a guiding hand to hone in on what would be a good answer.  Now, if that patron has a British accent, the librarian might want to ask "which civil war do you mean??" since they could be asking about any of several different civil wars.

An expert searcher incorporates much of this skill into their own search strategy, asking themselves:  What is it that I'm looking for anyway?  What would a great answer be?

This kind of meta-cognition is, to my mind, the next great frontier for teaching people how to search effectively.  It's hard to do and slippery to teach.  That makes it a great challenge.  More on this in times ahead.

Now... on towards Nome!!

A quick note about Google Books

For the past few days I've been in Anchorage, Alaska attending the 50th annual Alaskan Librarians Association meeting.  (More on that tomorrow...)

In the process, I taught a tutorial on Intermediate Google Web Search skills (click on that link to see what I talked about).  And in the process I showed a bit about Google Books.

In particular, I highlighted a couple of neat, somewhat unknown features of Google Books.

I was even MORE surprised when I found out that few of them used Google Books.  It's not because they didn't want to, but because they couldn't see why they would ever want to use Books.  A quick demonstration fixed that problem. 

Surprisingly, only one or two librarians out of the 400 there knew about the "Find in a library" link to WorldCat.  (Just out of curiosity, did you?) 

My Books demo was to:
 (1) show them the WorldCat link ("Find this book in a library"),
 (2) illustrate how to search for the text of a passage in a book, 
 (3) then how to locate archival books about topics of local interest (primarily Alaska state history;
       example: find old books that mention the Chilkoot trail),
 (4) how to use the shelf feature in Books,
 (5) how to search the magazines collection in Books. 

While all of these features are wonderful, if you're a teacher on Kodiak Island trying to get a back issue of Life magazine, it can be quite a wait for the interlibrary loan to get the back issue to you.  But on Books, that's a 10 second operation--Example:  search for magazines in the collection that mention Alaska statehood.

And the "Find in a library" link is incredibly useful if you're looking for a book that's (physically) nearby.

I did a search for books by Erin Mckean and found "Verbatim" (a wonderful book).  Now, to find it in a local library, I click on "Find in a library."  

Ah.  This is what you see if you're in the town of Kodiak on Kodiak Island, a smallish fishing village without an extensive collection.  You can see I entered the zip code for Kodiak in the "Enter your location" box.  But now you know where to look. 

As it happens, I actually live in Palo Alto, California, not so far from Stanford University.  

When I put in THAT zip code, I can see that there are several copies within easy biking distance of home.  

Now you know the secret of finding a book via Google Books.  It's handy on Kodiak, and it's handy where you live.  (And it's REALLY handy when you're looking for a super-rare book too.)

Answer: Finding an insect

When she asked this question, I felt very much like a reference librarian who's asked to identify an insect that a young patron has brought in.

With a question this open-ended, and since the insect in question wasn't around, the only approach that would make sense is to look for images of the bug.

Note that in a case like this, doing a regular Google Web search for [ california insect brown stripes on abdomen slow-moving ]  just isn't a high probability search.  There are LOTS of insects like this.

But more importantly, would she be able to identify the right insect given a textual description?  Probably not.  She wasn't an entomologist (if she were, I would use a totally different approach).

Instead, I was hoping that the insect was big, slow-moving and obvious--and so I bet that someone had taken a picture of such a bug and had posted it on the web.

So I turned to Google Image search and did the simplest possible query that I thought might work:

[ California insect ]

I know that's too open-ended, but I was hoping.

But no luck.

So THEN I tried filtering on the color brown.  Did you know you can filter images by dominant color?

Here's how to filter image search by color:

See the "Show options..." button on the last line of the screenshot?

If you click on that, you'll open up the "Options panel" and it'll look like this:

This gives you a whole slate of new options / new ways to filter the hits you're getting back from Image search. 

Since she said it was a brown insect, I thought I'd try filtering on the color brown. 

I didn't KNOW that this would work, but it only takes a couple of seconds, and I thought MAYBE it would be a common enough insect to show up in the first page or two.  

Success!!  The big, ugly, brown insect with stripes on the abdomen showed up in the lower right corner of the first page.  

The bug is a Jersualem cricket, a large, flightless cricket common in California.   (see: Stenoplematus fuscus,  http://www.insectidentification.org/insect-description.asp?identification=Dark-Jerusalem-Cricket for more details).  

Bottom line:  A good tactic when looking for common things is to just try doing the obvious search (in this case, on images, since any textual description is likely to be confusing, and confirmation is much easier by sight), then applying filters to get to the details of what you want to see. 

Wednesday Search Challenge (March 3, 2010) - Identifying an insect

This is a true story.

During one of my search-education classes for K12 teachers, I was asked if I could "search for this big brown bug I found in my back yard last week."

I was dubious that this would be easy to search for--insect identification is well-known to be hard under the best of circumstances, and I wasn't at all sure I would be able to find the particular bug she was talking about.

But what the heck... let's give it a try.

"Tell me a little more about the bug," I said.  "How big was it?  What color was it?  Anything else to go on?"

It turned out that the insect was pretty big, very ugly, slow-moving and brown with dark stripes on the abdomen.

With just that amount of information, I was able to do 2 searches and find the answer.  She confirmed that this was in fact the bug she'd seen.

Now... Can YOU figure out what this bug might have been?

A big hint:  The thing that confirmed it for us was the presence of stripes on the abdomen.

Give it a try.  Come back tomorrow for the answer!

What a graphic can tell you

A while ago I used to co-post in Kathy Sierra's blog "Creating Passionate Users." I wrote a post there that I've copied over to my blog because I wanted to be able to refer back to it (and I don't know if the archive edition of CPU will hang around forever). (Link to original post on "What a graphic can tell you.")


So I'm copied my original post here... and also wanted to make the point that I'll ALSO be talking about various sensemaking and information visualization topics in this blog as well.

________________________________

I love looking at things in a new way. More to the point, I love it when I see things about the world that I thought I knew... but then realize there's more to the story.

And for the past few years I've been doing various pieces of research into information visualization--the science and art of making complex information visually perceivable. It's cool, it's fun, and every so often you learn something really, really interesting.

I was just sort of doodling the other day in Powerpoint. (If you must know, I was trying to prepare a talk to give at some university, and I started wondering how many dots you could usefully squeeze onto a Powerpoint slide.)

Now that sounds crazy: doodling in Powerpoint? Yeah, I know. But come to think of it, why not? It's got a bunch of okay tools for drawing and while you probably won't make great art, it's fine for quick sketching. You won't confuse the output with a masterwork in oil or water color, but it's really great for a quick sketch. And it IS the tool I spend a lot of time using, so I've grown pretty accustomed to its idiosyncracies.

So I drew a bunch of dots. Then it hit me--if a dot stood for a day, what would my life look like as a set of dots on the screen?

(You can click on each image to see it full-size.)

If that's 20 years worth of days, then what would the rest of my life look like? Could I fit a lifetime's worth of days onto a Powerpoint slide?

And how would you think of your life in terms of segments? I fooled around some more and came up with this as a way of thinking about the various parts of my career.

I certainly hope that my life and career last more than 60 years! Nevertheless, I found this an intriguing way of looking at the progress of your life.

Or, to put it more pragmatically, where's your dot?

Question for you: What information graphic was most informative / influential / insightful for you? Was there a great piece of infoviz that really just did it for you?