Wednesday, February 14, 2018

SearchResearch Challenge (2/14/18): How much did this menu item cost?

Far be it from me to critique... 

... the food choices of different people at different times and places.  After all, I've eaten roasted grasshoppers (chapulines) in Oaxaca, smoked eel (rauchen aal) and blood sausage (blutwurst) in Germany, along with haggis in Scotland.  

Moo.  "Eat more fish," says Bossie.  

But I hadn't appreciated how much our commonplace and customary dishes have changed over the past 60 years.  Haven't we always been a land of peanut butter and jelly sandwiches?  I recognize that it's very common to eat boiled soybeans (edamame) these days, and it's easy to find sushi in your local grocery store.  That's a big shift over the past couple of decades.  

So I was surprised when I happened to discover that a few popular dishes from the 1950s were broiled liver pudding and boiled calf's head with brain sauce, which seem a bit over the top.  (Although I do admit that my father enjoyed scrambled eggs and brains, so go figure. I didn't inherit that dietary preference.) 

Fascinating.  And this leads to today's Search Challenge:  

1.  If you were in New York City in the mid-1950s, how much would you expect to pay for a good meal of broiled liver pudding or boiled calf's head with brain sauce 
2. (Open ended)  WRT the place you live, can you find something that was commonplace to eat in the 1950s or 1960s, that is very rare now?  Or vice-versa (something common now that was rare back then)?  

In all cases, you have to tell us (A) how you found the prices, and (B) some evidence that a food item was rare (or common) in the time-frames we're interested in.  (For instance, evidence might be an advertisement in a newspaper from 1959 in your locale.)  

Search on! 


Thursday, February 8, 2018

Answer: What's going on in these photos?

The key is asking good questions. 

But you knew that.  
This week, we tried to figure out what's going on with a couple of  decontextualized images.  In this case, these were pictures that I'd taken and recently re-found.  I also found myself puzzling about what they were... (Why, I asked myself, did I take this picture??)  

1.  What's up with these railroad tracks?  They see very odd, yet familiar.  Why are there three rails?  (I'll spare you the metadata extraction task.  This image was taken at:  38.908711, -77.068983)  

Link to original

As many Regular Readers quickly figured out, that lat/long is at the corner of P Street NW and 35th Street NW in Georgetown, Washington, DC.  
If you jump there using Streetview, you'll see more-or-less the same image I show above, confirming that the image is recent and in the correct location.  The map also tells us that this is in the Washington DC neighborhood.  

Since I know that trains that run through small side-streets are typically called streetcars, I started with the search: 
     [ streetcar Georgetown ] 
and quickly found the Wikipedia article about Streetcars in Washington, D.C.  I did a quick Control-F text-search in that article and found that there are several mentions of P Street in the article.  Apparently, there have been streetcars running to Georgetown down P Street since 1876. A few years later, in1895, Congress authorized the Rock Creek Company to purchase the Washington and Georgetown Railroad Company, producing the Capital Traction Company.  In the article there's a comment that "Tracks are still visible on 3200 to 3400 blocks of O St NW and P St NW..."   Checking the Google Map, that's correct.  (Just to be obsessive, I also checked Streetview on O St. and found that it's true: O Street also has old streetcar tracks with cobblestones.)  
Interestingly, a bit later in the the Wikipedia it goes on to say that "...The region's only remaining visible tracks and conduit are in the center of the cobblestone 3200 through 3400 blocks of P Street NW and O Street NW..."  
That's fine, and it's consistent with what we already learned... but I don't know what a conduit is in this sense.  By clicking on the link that's offered, I land on the DC Streetcar Tracks and Structures web page.  There, I learned that the "conduit" is the slot in the center which acts like a third rail, supplying power to the streetcar.  Reading the Streetcars in Washington, D.C.  Wiki article carefully, I see that the 3200 - 3400 blocks of P Street have the only remaining tracks and conduit that are visible in Georgetown and DC.  
Still a little uncertain about what a conduit was, I did another Image search for: 
     [ streetcar conduit diagram ] 
and found this page from the US patent application: 

And another diagram, showing how the conduit connects for power to the streetcar: 

And why would this look familiar to me?  What else does it look like?  
This is a tricky question.  How do I figure out something that's familiar to me??  
The only way I know to do this is to do a bit of visual scanning.  I wasn't sure where to start, so I did a few searches for images using queries like [ streetcar conduit ] or [ streetcar conduit rails ], looking for something that would trigger an ah ha moment.  Unfortunately, these weren't working for me.  
My next query hit paydirt.  I thought I'd try looking for images of the original Georgetown railway, so I did an Image search for: 
     [ Capital Traction Company ] 
and found this for my results page: 

That's when it hit me.  Those streetcars look a great deal like the cable cars I've seen in San Francisco.  What's more, cable cars also have a groove in the center of the street that looks like a conduit! 

Although, unlike the Georgetown streetcars conduit there's a cable down there, rather than an energized electrical cable.  The cable car gripman pulls back on a long lever, closing the jaws of the grip around the cable, which is constantly moving just under the street.  (Yes, it sounds unlikely--the cable is really, really long, and run in a continuous loop under the street.)  
But the cable car slot is NOT a conduit.  It just looks a lot like one, which is why it seemed so familiar!  

2.  Here's another picture I got from a friend, obviously taken late at night at the Googleplex back in December, 2013.  What's the backstory on the dinosaur?  Why pink flamingos? And what's with all the yarn??    (Extra credit:  What's the dinosaur's name?)  

Link to original 

I was there, but obviously don't remember much about this incident with the dinosaur in the night.  Can you fill me in on the yarn, flamingos, and dinosaur?  
This was a fun Challenge.  It's not hard... a simple search with the uncommon terms from the Challenge statement works well:  
     [ Googleplex dinosaur ] 
leads to all kinds of fun pictures. Here's one from the first day flamingos appeared with the T-rex.  I remember walking into the building that day and being impressed that someone went to all the trouble to buy dozens of flamingos for a practical joke.  (This was not long after the T-rex was first installed.)  

It's also fairly easy to find multiple sources (e.g. Business Insider) telling us that the T-rex's name is Stan. (And no, I don't know why "Stan.") 
The yarn thing is a bit harder--the key insight here is to search for the idea: 
     [ yarn statutes ]    or 
     [ yarn covering statues ] 
to discover that knitting yarn-based coverings for statues is called yarn-bombing, which is often a gentle act that is often about reclaiming and personalizing sterile or cold public places. It's also often quite funny, as the artists sometimes do quite elaborate yarn constructions to annotate or comment on public art.  (See:  R2D2, benches, or the Wall Street bull covered in yarn).   
So in some sense, the flamingos are, like yarnbombs, easily removable commentary on the underlying structure.  
You might enjoy knowing that perhaps the cleverest "easily removable commentary" on Stan-the-T-rex happened a few weeks after the flamingos first appeared.  They slowly disappeared over time.  Two went missing the first night, four vanished the next night, etc.  Until after about two weeks, they were all gone.  All that was left was this, below the tail end of Stan... 

Use your imagination! 

Search Lessons 

1. Searching for familiar things that you don't have words for often requires a bit of browsing.  In this case, hunting for "something that looks like a triple streetcar track" took a bit of doing.  We started by trying to figure out what those tracks were in Georgetown, then we had to pull out a little bit and start looking at images of the Capital Traction Company before we found an image that reminded me of the target--San Francisco cable cars!   The skill here is to get to a topic area that's close to what you seek... then browse a bit.  
2. Searching for a concept is often choosing a query that describes the concept, rather than just searching for the thing itself.  Here, it wasn't obvious how to search for information behind the yarn on the statue... until we searched for the concept expressed in the simplest possible form.  THEN we learned that there was a general concept of yarnbombing, which then gave us what we needed to know.  

Search on! 

Tuesday, February 6, 2018

Image searches with chip control!

You might have noticed... 

... that the Image search now has a set of colored rectangles just below the query area.  Here's an example with a simple query.  See those rectangles?  They're called "chips," and they modify the image query.  

If, for instance, you click on the blue "watercolor" chip, you get nothing but images of roses that are done in watercolors.  Notice that the selected chip is moved to the far left of the chip-row, and made grayish.  This indicates that the [rose] query is being modified by the "watercolor" chip selector.  

In a sense, the chips are suggestions for extending or modifying the original query.  

NOTE:  The color of the chips has nothing to do with the color of the image or how the query will be modified.  The chip color is just to separate the queries, putting them into handy categories.  (e.g., "watercolor"  "glitter" and "pastel" are all conceptually related, so they all share the same background color--but they're not searches for blue roses) 

Both Google and Bing offer the same user interface (see the Bing [rose] query below).  As you can see, the suggested chips are a little different between Bing and Google, but it's the same basic idea.   

And, if you click one of the Bing chips, you see that they CHANGE the query to the text in the chip.  

Google's chip also modifies the query.  As you can see, clicking on two chips gives you 99% the same results as adding those terms to the query itself.  See the two SERPs below shown side-by-side.  They're pretty similar: the one on the left is with the terms from the chips added into the query, while the SERP on the right has the two chips selected.  

The chips are there to accelerate your exploration of the space of images.  You can click them on/off rapidly to see what's available near this query.  (This is incredibly handy when you're exploring some ideas for design purposes, or if you're not sure exactly what you're looking for and need to poke around a bit.) 

Notice that the Bing image search also has an additional set of suggestions after the 4th row of images.  These are completely different queries (not just modifications of the original query), but sometimes also useful if you're not certain what you're looking for... 

Search Lesson 

When you're actively exploring / searching for images, notice the chips--they can sometimes help you with fast and furious searching!  

And... a more subtle point:  Notice that the chips user-interface idiom is becoming much more common.  You'll see chips in many places, a probably even more frequently in the future.  Here's an example from my Android, when I was searching for a cast iron skillet.  There are specific chips showing different brands of iron skillets--as with the image chips, these modify the query.  If I tap on the chip with the Le Creuset skillet, the query will become [Le Creuset cast iron skillet] 

Enjoy your new search method!  

Search on, visually! 

Wednesday, January 31, 2018

SearchResearch Challenge (1/31/19): What's going on in these photos?

Every so often... 

... you run across a photo that makes you ask, "What's going on here?"  You take a look, and think "hmmm...."  

Often, these are images that you find in things you're reading; a figure, photo, or illustration that makes you pause and wonder.  And usually, the text on the page will give you a good clue about the backstory.  

But then there are the decontextualized pics.  Like as not, these are photos that you've taken somewhere along the way, and you run across them on your way to looking for something else.  

Here are two such photos I ran across recently on my hard drive--images that made me wonder "What's going on here?"  

Can you figure out what's happening in each of these images? 

1.  What's up with these railroad tracks?  They see very odd, yet familiar.  Why are there three rails?  (I'll spare you the metadata extraction task.  This image was taken at:  38.908711, -77.068983)  

Link to original

Those rails seem awfully familiar.  Any idea why?  Hmmm.... 

2.  Here's another picture I got from a friend, obviously taken late at night at the Googleplex back in December, 2013.  What's the backstory on the dinosaur?  Why pink flamingos? And what's with all the yarn??    (Extra credit:  What's the dinosaur's name?)  

Link to original 

I was there, but obviously don't remember much about this incident with the dinosaur in the night.  Can you fill me in on the yarn, flamingos, and dinosaur?  

As always, you get virtual SRS points for figuring out the Challenge.  But your name will be lovingly inscribed in platinum letters on the SRS eternal heavenly charter if you also tell us how you figured it out.  

What did you have to search for to get the backstory on the triple rails?  What kind of clever searching did you do to discover the dinosaur's name.  (Who knew that dinos could even have a name?)  

Search on!  

Wednesday, January 24, 2018

Answer: How do I find this song that's become an earworm?

It's true: I got an earworm while bicycling... 

I know, I know--an earworm is typically a short, catchy piece of music.  But perhaps my brain (or ear) is wired a little differently and my earworm is a little longer than most.  Whatever the cause, my mind latched onto this tune, playing repeatedly in my head.  The problem was, I could NOT remember the name of the tune.  

Of course, this immediately activated my SearchResearch instinct--how can I figure this one out?  The only thing worse than an earworm is an earworm you can't name!  

Here's a recording of me whistling the tune: 

           playable m4a file of the mysterious tune 

FWIW, I could play it on the piano as well, but I realize that's kind of an odd skill to have, and not everyone has a piano at home.  But (almost) everyone can whistle.  

Now that you've heard it (albeit my tentative whistling), here's our Challenge:  

1.  What's the name of this song?  
2.  If you can't recognize it... how on Earth would you search for this tune? 

As I've said many times before, when you've got a problem to solve, try searching for a tool to help you solve the problem.  

What would I need to search for a song by whistling?  Here's my query: 

     [ whistling search engine ] 

which gave me the following SERP: 

I tried Musipedia, but it requires a Flash plug-in to access my microphone.  Since that's a bit wonky and slightly unsafe, I kept looking.  

Many of the later articles (e.g. GuidingTech or Mashable) all pointed to the Midomi musical search engine, giving it high marks as a good discovery service.  (To be fair, the #7 result for this query is a link to

This is my video recording of my using Midomi and finding the tune.  I'm not doing anything fancy here--I just recording myself whistling the tune (as you heard above), then launched Midomi, and clicked "playback" on the m4a file in a different tab.    

YouTube video link: 

As you can see, it found that my whistling matches the song "Wave," by Antônio Carlos Jobim.  

I was initially a little surprised when I listened to Wave on YouTube.  

For the first few seconds it doesn't sound ANYTHING like what I was whistling.  But I persisted in listening, and 10 seconds into the music, my melody magically appeared.  

Mind you, this doesn't work for absolutely every possible melody, it seems to work mostly with fairly popular and well-known melodies.  I was surprised, for instance, that I could not get it to recognize the popular Japanese folk tune Sakura (at least not with my whistling).  

Search Lessons

1. There are more search engines than you might imagine.  In a future post, I'll talk about some of the variety of search engines out there.  But as a general practice, when you find yourself searching for something, but just can't figure out how to do it, try searching for a speciality search engine.   (These are sometimes called "vertical search engines" because they focus on a particular market, aka a vertical.)  

Search on! 

P.S.  Caution:  Thanks (kind of) to Regular Reader Remmij who gave us a link to the Top 50 Earworms of All Time.  Careful... you might not want to open that page!  If you do, any earworms you get are not my fault!  

Wednesday, January 17, 2018

SearchResearch Challenge (1/17/18): How do I find this song that's become an earworm?

I was bicycling home when it happened... 

wasn't thinking about anything in particular, I was just out for a ride. But somewhere along the way I got a terrible earworm--you know, one of those catchy songs that gets stuck in your head, and won't stop--it just keeps repeating over and over in your head. 

For whatever reason, I could NOT figure out what it was.  I could sing it, I could whistle it, I could even play it on the piano, but I was utterly unable to put a label onto it.

Of course, this immediately activated my SearchResearch instinct--how can I figure this one out?  

Here's a recording of me whistling the tune: 

           playable m4a file of the mysterious tune 

This is this week's SRS Challenge! 

1.  What's the name of this song?  
2.  If you can't recognize it... how on Earth would you search for this tune? 

Naturally, please let us know how you figured out the name.  If you just know it off the top of your head, that's fine, but tell us that.  If, like me, you know the song, tell us WHAT you did to figure this one out.  

This is an Open Internet quiz.  You may use whatever method works.  (Just tell us your secret.)  

Just so you'll know, I figured it out (using my SRS skills).  Took me about 10 minutes.  How long will it take you?  

The good news is that once I'd given the earworm a name, it went away.  I have my sanity back.  Thank heavens.  

Search on! 

Monday, January 15, 2018

Search-by-Image with site restriction

We've talked about Search-by-Image before, but here's a nuance we haven't mentioned...  

When you do a Search-By-Image, you upload an image to Google so it can find similar kinds of images.  To do an SBI search, just click on the camera icon, then upload your image.  

It works remarkably well.  But sometimes, it needs a bit of direction--a little guidance to make your result more precise.  

For instance, you can search for a picture of a caterpillar that you photographed in order to figure out its name.  Here's the one I found: 

When you do a SBI, you see this result: 

That's all true, but not helpful.  This caterpillar is NOT the caterpillar of the European gypsy moth.  They look like this: 

So... how can we modify the query to improve the search result?  

Easy--just give it a bit of a hint--like this: 

Now that you've got some decent "Visually similar images" you can click on that link and quickly find a matching caterpillar. 

It doesn't take long to figure out that this is a Spotted Tussock Moth caterpillar (3rd row, 2cnd from the left).  

But wait... there's more!  

You can ALSO modify the query to include other search operators you know about.  

Here's a thumbnail image of a person that I couldn't identify off the top of my head--although I knew she had some kind of connection with the Rochester Institute of Technology (a college in Rochester, NY).  

When I just a simple SBI search, this is what I get (the top SERP), but if I modify the search to include  (their domain), the result is much more precise.  

The result on the bottom of the page tells me exactly what I need to know.  

Search Lesson 

1.  Remember that you can modify the query in the Search-By-Image operation... including adding a site: restriction.  This is really handy for searching within sites that you know are affiliated with the object of your search.  

Search on! 

Friday, January 12, 2018

Answer: The case of the Missing Island

You think a decent-sized island wouldn't get lost..

And yet, somehow, Google Maps seems to have lost an island off the coast of Taveuni, in northern Fiji.  

To remind you of the Challenge:  I was diving in the Somosomo Strait, between Taveuni and Vanua Levu.  In the map below, I've marked both the islands and the strait.  

Taveuni Island, the Somosomo Strait, and Vanua Levu, the big island to the west.

While I was staying at the Nakia Lodge near the northwest corner of Taveuni, I took a lot of photos, including this one: 

In this photo, I'm standing at the edge of the lodge, at -16.741502,-179.950285 -- the photo angle of the this shot is about like this: 

But when you zoom in on the map, this is what I see: 

And it doesn't matter if I zoom in with the satellite view... there's nothing there, just a kind of smearing of photo edges!  

And yet I know, from personal experience, that this small island is basically due west of the small town of Somosomo.  


1.  What is the name of that mysterious, seemingly missing-in-action island? It's big enough to be on the map, and definitely big enough to show up in the satellite photos... so... what is it? 

The first thing I did was to check other map resources.  Bing Maps looks like this. 

But they have the same issue--there's no island indicated there!  

I poked around for a while looking at different kinds of maps.  For example, I tried National Geographic, but they hold onto their maps pretty tightly--no freebies there! 

I also tried: 

     [ high resolution map of Fiji ] 

but many of these maps are just Google Maps repackaged, and as we know, that doesn't solve the problem.  

Using this approach, I DID find a reasonable map of Vanua Levu that shows us an island there, but despite being labeled as a "high resolution" map, it doesn't have enough resolution to tell us the name!  

That little dot beneath the "o" in Somosomo isn't a typo or drawing error... It is the island we're looking for!  

While I found a lot of maps this way, almost NONE of them were useful for answering our question.  Many of them were on similar topics, but not really useful.  Time for a new strategy.  

Next, I tried the USGS.  Did THEY have maps of Fiji?  Answer:  No.  

How about nautical maps?  I found a bunch of navigation maps providers, but they all want REAL MONEY to let me look at their maps.  If I was to sail in the area, I'd do it... but I wasn't ready to give up just yet.  

Who else might have good maps?  

How about the country of Fiji??  That makes sense... perhaps they have nautical charts for commercial ships and vessels that I could look at.  

So I tried the query: 

     [ site:FJ map Taveuni ] 

which gives me the very useful search results page: 

As you can see, there are a bunch of images, and then a number of real estate offerings.  

Okay.. let's check the images first: 

A quick glance tells me that there's an island there, and all of the maps of the real estate agents seem to have it!  

Riffling through these images, there are many maps with the mystery island marked...and given a name. Here's a nice clear example from multiple sites: 

Looks like our mystery island is named Korolevu! 

But as always, let's double check:  

     [ Koruolevu Fiji ] 

This gives us... a surprise!  It turns out that there's another area on Fiji called Korolevu... "Korolevu, on the banks of the Sigatoka River, is the birthplace of tourism in Fiji, with the Korolevu Beach Hotel first opening in the 1950s..."  Ah... we have to be careful about what we read.   

That's nice, but it's not our island!  What's going on?  I need to be a bit more discriminatory in my query.  So... 

I modified the query to be: 

     [ Korolevu island Fiji ] 

And I got THIS really useful SERP: 

As you can see, it really IS an island (even if the Google Maps insert is showing us a blank ocean).  

But now we can look around a bit through these results and find some great validation of this as the island's name.  For instance, tells us that this island is Korolevu, at -16.766667, -179.983333 (and when you check that location on Maps, it's clearly the same place, right in the Somosomo Strait).  

Interestingly, there's a Wikipedia article on Korolevu Island... in Swedish!  Luckily, Google Translate works reasonably well, but we don't learn much more about the place. So all this does is confirms that someone else believes it's a real place.  

However, when you find the island For Sale (keep scrolling down the SERP to position 40 or so), you know it's for real.  Do a Control-F to find Koro Levu (note that they use a space to separate the terms), and you'll see the ad: 

This is clearly the same place.  The appearance in this image is a lot like the one in my picture--here they are side-by-side: 

So, at last we've got good evidence:  The island that's left off of Google (and Bing) maps is called Korolevu (or Koro Levu), and is an uninhabited island located just west of the town of Somosomo. 

The big question is WHY is it missing from the maps?  The short answer is that it's just an error.  I spoke with a few folks in the Maps image team, and it turns out that it's just an unfortunate meshing of different images from different satellite photos.  And, it just happened, that Korolevu got lost in the overlap from different images.   

A couple of images: 

This first one is a picture I grabbed of an official Fijian map (the original of which I have not (yet) been able to find online).  As you can see, THEY know where Korolevu is.  Note also that it's just about perfectly on the 180 degree meridian (that's aka the International Date Line).  

Dan playing guitar at Nakia Lodge, overlooking the Somosomo Strait. 
Korolevu is in the channel on the left side, just to the right of the mango tree.
P/C to my buddy Audrey Yang.  

In looking at the comments, I was really impressed by some of the findings.  

Remmij used a different strategy:  Don't limit yourself to Google and Bing, but look for more photo sites.  And... Remmij found another mapping where the photos were (mostly) successfully tiled together to show Korolevu.   They clearly have one photo-tile that's not the best choice for that section of Taveuni... 

Image of Korolevu by Wego.Here

Fred and Jon (the Unknown) did a different smart approach with the query by creating a query with the basics of what they know: 

    [ island west of somosomo ]

and found Korolevu that way.  Sometimes the simplest methods work best! 

Luis Miguel found the amazing website by Ulrich Deuschle: his "Generate a Panorama" webtool can basically reproduce a version of the image I took. Here's what was able to create with his tool by setting the camera point to where I was standing, and setting the angle of view to approximately what the camera could see: 

And if you slide the synthesized image a bit to the right, and setting it next to my photo, it looks like this:  

That's a lovely piece of work by Deuschle.  (Obviously, he's using data that's NOT from Google. Digging into the data sources of his app, it seems that the data is coming from the NASA Shuttle Radar Topography Mission.  This is something worth looking into!)  

In his searching, Arthur found the incredibly handy GeoViews website.  If you know the name of what you seek, this has a remarkably deep set of image resources.  Here is the pic of Koro Levu from GeoView.Info

Oddly enough, this fantastic image is originally from Panoramio (taken by  That's odd because Google owns Panoramio, but somehow we haven't managed to connect the Panoramio images with the placenames in Google Maps.  

Search Lessons 

Lots to talk about here, but I'll limit myself to just three points... 

1.  Just because you didn't find it doesn't mean it's not there!  In this example, we weren't able to find the "missing island" because it didn't appear on Google (or Bing) maps.  But since I had proof that there really was an island there, I kept digging.  Sure enough, other resources have the island.  (Moral of this story:  If you know something is true, keep digging. Be persistent.)  

2.  Get multiple--and different!--sources to confirm what you've found.  Since the island wasn't on any of the obvious map resources, we had to go to other sources... including local real estate maps.  Generally speaking, local resources often contain information that's not in the official content.  Of course, you really have to check those sources out, but they often provide great leads and insights.  Don't ignore them, but DO double (triple!) check.  

3.  When you don't know anything, you need a map.  In this case, we really needed to find a literal map.  But this is true for other kinds of research questions as well.  If you're looking for a name of a research area (e.g., "what is the name of a research area that deals with algae?") you might consider finding a "map" of the botanical sciences.  Here's one such query that yields a "map" of botany: 

And you quickly learn that the study of algae is called "algology" and "phycology." This is one of the things that a map is good for--not just identifying the thing you're after, but also giving you a bit of an overview of the area.  In our Fijiian map example, we were counting on the map showing us what's there, and what the parts are called.  

That's a useful skill to have--finding a resource that gives you overview and context.  

Search on!  (Perhaps with a map!)