A blog about search, search skills, teaching search, learning how to search, learning how to use Google effectively, learning how to do research. It also covers a good deal of sensemaking and information foraging.
Monday, November 30, 2015
Answer (Part 1): How can you search all the coasts?
So I'm just going to answer the second part of last week's Challenge. We'll get to the "how to search the coastline" later. (As I kind of suspected, it's one of those questions that requires a bit more work than usual. Maybe we should start a new category of Challenges: "Makes you think differently about data," or something similar.
But let's talk about Challenge #2 for today. Remember? It was this:
2. The first picture above shows a stretch of the California coastline not far from where I live. It's a region called "Devil's Slide" which has a tunnel running through a piece of the coastline that perpetually falls into the sea, usually taking the road with it, hence the name. The Challenge is to identify the building that's near the southern exit of the tunnel. This is what it looks like now. Challenge: WHAT was that building originally? Who built it, and when?
Here's what it looks like now.
Here's what it looks like now.
A quick search for [ Devil's Slide California ] takes you to the Wikipedia page (or the State Parks page) and tells us that Devil's Slide is a coastal promontory in California that's on the coast between Pacifica and Montara. You can get the lat/long from the Wikipedia piece and then look at it on Google Maps.
Dropping the lat/long (from Wikipedia article) brings you to this location on Maps. Note the two "bunker" locations in mid-image. If you click on the image on the far left in the image strip (at bottom), you'll see that there's a Photosphere there, so you too can zoom in on the old bunker. See image below...
Rosemary noticed that the pictures were called bunker1.jpg, bunker2.jpg... so that was a big clue. (That'll teach me to give my pictures less useful names!)
The Wikipedia article about the Devil's Slide also has a Military History section. There were 6 such structures built in World-War 2 as triangulation stations to direct fire from the big guns farther to the north. (More on those another day. They're very fun to explore.)
As Ramon pointed out, a query for [ Devil's Slide history ] reveals a great deal including some early photos of the railroad tracks (which predated the road) being buried by landslides... hence the name. It's a naturally unstable bit of geology, even though this particular bit is hanging on tightly.
The bunker on Devil's Peak was originally built during World War II as a "Triangulation and Observing station." It was part of a larger complex of buildings, including a nearby generator hut. When in service, a watcher equipped with a set of binoculars would keep watch out at sea and if they spotted any enemy ships they just would phone in the heading to the big battery a few miles to the north. With the advent of more modern missile defenses the station became obsolete and the entire site was abandoned in 1949, leaving just this empty bunker atop Devil's Slide.
The land was eventually purchased by a private owner but the bunker remained. The earth around the top of the peak was removed for a construction project sometime before 1970. The project was never quite completed, leaving this remarkable fragment of history in place.
When I was originally researching this, I ALSO found a set of blueprints to the station. (Really! They're still marked "secret" and everything!)
For people who love this kind of sleuthing, I'll leave it to you to seek them out. They're really pretty interesting to read.
Search Lessons
There are at least a couple of lessons to draw from this Challenge.
1. Some Challenges are actually fairly tricky! This is definitely the case with the "find a stretch of coastline" Challenge. I'm pretty sure I know how to do it, but I haven't yet found a reasonable way to do it... and I'm still searching for that. I'll give an update later this week.
2. Sometimes simple searches work. In the "bunker" case, the simple searches really do work pretty well. In this case, a quick search plus a couple of links got us there.
3. However, another lesson is that one should take better notes. I can't believe I lost track of the blueprints to the bunker. I'm sure I can find them again, but 15 minutes didn't get me back to them! (I'll update you on this when I find them once more...)
Search on! (I am, I am!)
Wednesday, November 25, 2015
Search Challenge (11/25/15): How can you search all the coasts?
Yes, it really is hard to search...
... all of the coastlines of all the world. In the comments on last week's Challenge, a few readers (looking at you, Remmij!) expressed skepticism that my approach would work in general. And in general, that's true.
However, that particular Challenge had enough constraints baked into it (the coastline at such an angle, the clearly undeveloped countryside, the striking red color of the earth), that I thought it would be possible to do "manual" search. And, it actually worked! We found it.
Obviously, searching for a different aerial image might make it impossible to find. Suppose instead I'd asked about this shoreline:
A random stretch of coastline, tough to find by just scanning. |
That would be a lot more difficult. This is clearly a north-south road near the shore, but it could be anywhere in North or South America, parts of Europe, even parts of Japan and Africa would have to be scanned.
So, can we find a tool or database that would let us figure this out?
Here's Part 1 of this week's Search Challenge. (And I warn you; I haven't figured this one out, so it could be tricky. On the other hand, this is Thanksgiving weekend in the US, so I hope to have some time to experiment along with you.)
1. Can you find a tool, database, or some way to rapidly scan ALL of the coastlines of the world looking for 100 km wide sections of coast that run northwest-to-southeast? (Such as we saw in last week's image.) For this week, all we care about is if the coastline runs mostly "down to the right" at roughly -30 degrees (or course, that's the same as "up to the left" at +30 degrees). For instance, on last week's image, this part of the coastline would match:
Or, in that part of Madagascar there are really only 2 places that match this criterion, the arrow on the left in the following image is the same as the above, just zoomed way out:
And... for people who want to work on something slightly less intense, here's a second Challenge:
2. The first picture above shows a stretch of the California coastline not far from where I live. It's a region called "Devil's Slide" which has a tunnel running through a piece of the coastline that perpetually falls into the sea, usually taking the road with it, hence the name. The Challenge is to identify the building that's near the southern exit of the tunnel. This is what it looks like now. Challenge: WHAT was that building originally? Who built it, and when?
Here's what it looks like now.
Streetview of the strange building (just above the "no parking" sign) |
Let us know how you came up with the answer!
And, as I said, if you're working on #1, I don't yet know if it's easily answerable or not. It really IS a SearchResearch Challenge!
Search on!
Tuesday, November 24, 2015
Answer: A couple of odd questions
From the Cronkite Journalism School in Phoenix...
... I'm writing this is one of the classrooms before giving my talk about "Deep Search," which is pretty much what the topic of SearchResearch is all about. It's a fascinating time to reflect on the connections between what we talk about here, and what I think people need to know to be effective searchers. More on this later...
Right now, let's solve last week's Challenge.
1. This wallpaper showed up recently on my Android device; I was struck by how beautiful it is. Can you figure out where this in on our planet? What's the lat/long? (There's no metadata here, I already checked.)
First of all, excellent SearchResearch skills by the team! You collectively solved it in less than 24 hours. This was a pretty difficult challenge, so I'm very impressed.
As I mentioned, I saw this first as wallpaper on my Android device. As it turns out, I asked a fellow Googler about this, and he took the Challenge and found it pretty quickly! Impressed, I asked him how he did it. He wrote back to say:
I manually searched on Google Earth. I was able to make some assumptions about where to look from details in the original image, but then I basically followed a hierarchical linear search of likely coastlines. I was able to cover most of Africa in a few minutes by zooming in and out along the coastline and skipping obvious jungles, deserts, and mountains.
When it wasn't there, I considered other possibilities and rapidly discarded them (the non-straight, unpaved roads and irregular field shapes really rules out places that have the right potential geography but are even moderately affluent, like parts of South America or Australia -- the roads are all straight along the coast if the ground is flat, and the fields are quite regular).
That left very few possibilities (basically just Madagascar once you recognize that most other places that are likely to have that kind of roads in flat areas are all jungles). In all, it took less than 10 minutes (describing the process in this email was slower).In other words, he recognized that the image has certain properties: it's got a characteristic field layout, the coast runs roughly NW-to-SE (at around -30 degrees), and that it's clearly in the 3rd world. By quickly scanning plausible places (in the tropics) at a 30,000 foot elevation (or 10 km height), he was able to spot the characteristic reddish color and the angle of the coastline.
Interestingly, I didn't really believe him about this, so I started looking around with Google Earth as well, and much to my surprise, I too was able to find it with not-quite-manual search, but informed searches of our planet.
I estimated the size of the photo as being about 3 miles across, so I wanted to start scanning in roughly 3 mile chunks of coastline that was largely red, rural, and trending NW-to-SE.
I know the world is a big place, but if you look at Africa (this is from Google Maps in Earth view mode), you'll see that there actually aren't that many places to look that will match this description:
Only portions of South Africa, Nigeria, and Madagascar look plausible. A quick check of those places shows that only Madagascar makes any kind of sense. Now you could say, "wait a second, there's a LOT more to the world than this!" But keep in mind you only need to check from pretty far up--you can literally skim the maps at a high "altitude" -- like this:
At this height, we can look at very large swathes of continents for plausible places. It's clear only the southern end of Madagascar is plausible. If you zoom in a bit more:
Here I added the dashed rectangle to show a plausible region on the coast, and the red/white bars to indicate 10-mile segments. As this altitude, you can scan around 145 miles of coastline in a couple of seconds. It doesn't take long to look at only the coastlines that are at the correct angles (and land color). This is 145 mile wide section, only the part in the box makes any sense to zoom in. And when you do, this is what you see:
This is at -25.1684274,46.535719 (LINK to Map), and obviously the right place. (Not quite the same color, but with a little shopping, we could get it to be pretty darn close. (If you check the wave breaks side-by-side, you'll see it's really the same image.
2. Some animals have the ability to grow body parts back when they lose them. Some lizards, for instance, can regrow a lost tail; some salamanders can grow back a leg. But SOME animals don't grow the missing body part back, but reorganize their body plan. What's this process called? What kind of animal can do this?I didn't realize this would be quite so easy. The obvious query:
[ reorganize body after losing part ]
gives an article about symmetrization, which is the ability of some animals (such as jellyfish) to regrow body parts that have been lost (either through natural causes, like a turtle bite, OR a biologist cutting off bits). A paper describing this process (and from which the photo below is taken) is Self-repairing symmetry in jellyfish through mechanically driven reorganization.
(M. Abrams, et al., Proc. National Academy of Sciences, vol. 112 no. 26, doi: 10.1073/pnas.1502497112)
From left to right, the process of jellyfish symmetrization, starting with the amputated state. Image Credit:Michael Abrams, Ty Basinger, and Christopher Frick, California Institute of Technology |
3. I ran across this great word Mbaqanga, and I know it refers to a particular kind of music. But WHAT kind of music is this? And... how in the heck do you pronounce that word?
It's pretty easy to figure out that mbaqanga is a kind of African music, specifically, it's a South African dance music that combines traditional elements (as chanting and drumming) with elements of modern music (such as jazz and rock). A quick search on YouTube shows us the Mbaqanga song by Mahlathini & The Mahotella Queens (1991). At 1:54 in this video, you'll hear her say, "... they are the creators of Mbqanga..." As you listen, notice the slight click in her voice mid-word.
Now, how can I pronounce it? I know that both Google and Wikipedia will give pronunciation guides in IPA (International Phonetic Alphabet). Here's what they tell us:
Wikipedia: Mbaqanga, Zulu pronunciation: [mɓaˈ!áːŋa]
Google: Mbaqanga [ (ə)mbəˈkäNGɡə/ ]
But notice that the Wikipedia pronunciation has an exclamation point in front of the á character. That's the click sound, fairly common in Zulu. (See the Wikipedia article on Click consonants.) I don't know why it's missing in the Google pronunciation (I'll have to check on that!)
Oddly, many of the usual sources for word pronunciations (Forvo, etc.) have suspect (or clearly incorrect) pronunciations. Lesson: Never trust the pronunciation of a computer-generated voice! Even the pronunciation you can find on Merriam-Webster (and other dictionary sites) isn't acceptable. (It seems they tried to just pronounce it from the IPA and didn't bother to ask a native speaker!)
But Luís found a fantastic Mbaqanga video which has the musicians talking about this particular kind of music... and pronouncing the word very clearly. You can REALLY hear the mid-word click in this video.
Search Lessons
1. Manual search sometimes works, even when you don't think it will. I have to admit that I was surprised by the idea that you could manually scan all of the relevants parts of coastlines worldwide. The key insight was that by looking for only specific kinds of large patches of coastline, it became a manageable process. This is worth remembering.
2. As usual, it's worth checking multiple sources for consistency, and validating those against ground truth. In this example, the pronunciation of a word like Mbaqanga can be tricky. Here we found two different versions of the pronunciations (and many of the YouTube videos are inconsistent), so it's good to find someone speaking who has ground truth validity. (e.g., the Zulu singers or jazz musicians who say the word in natural context).
As usual, I learned something in the process. Hope you had a good time!
Search on!
Wednesday, November 18, 2015
Search Challenge (11/18/15): A couple of odd questions
I have a file...
... of SearchResearch Challenges that I keep updated with new questions. In that document I've got a few in the "Odds & Ends" category. Here are a couple of quick questions that have come up in the past few months that I can't work into a regular Challenge. So, for today, a special Challenge!
1. This wallpaper showed up recently on my Android device; I was struck by how beautiful it is. Can you figure out where this in on our planet? What's the lat/long? (There's no metadata here, I already checked.)
2. Some animals have the ability to grow body parts back when they lose them. Some lizards, for instance, can regrow a lost tail; some salamanders can grow back a leg. But SOME animals don't grow the missing body part back, but reorganize their body plan. What's this process called? What kind of animal can do this?
3. I ran across this great word Mbaqanga, and I know it refers to a particular kind of music. But WHAT kind of music is this? And... how in the heck do you pronounce that word?
Can you let us know HOW you found the answers to each of these?
We'll give the answers next Monday, and probably have a couple of other updates during the rest of this week. Stay tuned!
Search on!
Tuesday, November 17, 2015
Answer: Fake or real? How do you know?
Generally speaking...
...how do you know if it's faked or real?
These days, with wonderfully competent photo editing software, and the sheer number of people generating content, it's sometimes really hard to tell. Let's tackle each of these Challenges one at a time.
1. Is this a faked photo? If so, how can you tell? (Be specific.)
But let's focus on the plane. If you download the entire picture and zoom in a bit, this is what you see:
I've added a couple of arrows to indicate the direction the light is falling. The sun is clearly up and to the right of the plan (that's why the shadow is falling from the rudder onto the elevator). Notice also that the sunlight makes a pretty sharp shadow; it's full sunlight on the plane.
By contrast, let's take a look at the tower in closeup:
Here you can barely see any shadow. It's a diffuse light, still tinged with the rosy colors of sunrise (or sunset), a kind of light consistent with the time on the clock (although many such clocks are wrong). In any case, as you can see by this small arrow, the shadow is barely there, and it's coming in from the left.
So this picture is clearly a composite of two different photos. Each piece of the picture is illuminated by a different source of light. And in outdoor photography, all sunlight comes from the same direction. (There are exceptions, but this isn't one of them.)
As Luís pointed out (quite correctly), doing a Search-By-Image on just the subimage of the fighter matches up perfectly with this image of a F-16 fighter from the Hellenic Air Force during a solo demo flight at Tanagra Air Base, Greece
Remmij also spotted the roundel on the wing and tried to identify it. It's hard to ID for sure, but it's not a US mark, and it's unlikely that any other country would be flying F-16s in a tight pivot around Philadelphia City Hall.
Remmij ALSO spotted another indicator of being shopped: A cursor left in the image!
Remmij ALSO spotted another indicator of being shopped: A cursor left in the image!
So this image is clearly a FAKE!
2. How could you tell if these hotel reviews are true or faked?
Review 1: Stayed here for two nights with my wife and golden retriever in a pet friendly room. All staff in the hotel are super friendly and helpful. Can't beat the location. Visiting our son at Dalhousie. We are traveling from Maine and have greatly enjoyed Halifax. If you are into jogging, you can leave hotel and make loop in the point.
Review 2: We only stayed for one night but I would have happily stayed here longer. The hotel is beautiful throughout. We upgraded our room for a good price and had a really nice king bed which was really comfy. We ordered room service too for breakfast and were very impressed how quickly the food came.
How do you start on a question like this? How CAN you possibly determine is a hotel review is faked or not?
If you think about it, there are many great reasons to write a fake review (of hotels, restaurants, music recordings, films... anything that people buy AND review will attract faked reviews).
I started this question by searching for some background on how to detect faked reviews:
Somewhat surprising, I found a web site (ReviewSkeptic.com) that claims to do some processing of the language of a hotel review and then give you a score about how fake/deceptive (or not) it is! (You can read the news release about ReviewSkeptic, or the research paper underlying the mechanism here, if you're up for the math involved.)
Running these two reviews through ReviewSkeptic suggests that they're both real reviews. That's encouraging, but could we do more?
One obvious thing to do is to check other reviews by the same person. By doing the obvious query (select the first sentence of the review and search for it), you'll find that this review was written by MarkB, a Level 2 TripAdvisor reviewer. If you click on the TripAdvisor badge for him, you'll see he's written 6 reviews, 5 positive, 1 negative. By following up on all of the reviews he's written, can see see they're all more-or-less in the same tone (and language patterns), and distributed around, not all focused on one place (as would be typical of someone who's doing intentional deception).
The same analysis with the second review finds that Madeline75 , a Level 5 reviewer, has been writing reviews for 10 years on TripAdvisor and has written 59 reviews, scattered across lots of time and places. Again, the reviews are very similar in tone and language, and have a good distribution of opinions--some great, some terrible, just what you'd expect in real-life.
I'd mark both of these hotel reviews as REAL!
3. In this political season, many quotations from famous people (e.g., the Founding Fathers) are being bandied about. Which of these Jeffersonian quotes do you think is real OR fake?
Quote 1: “Do you want to know who you are? Don't ask. Act! Action will delineate and define you."
Quote 2: “That the obliquity of the ecliptic, when these elephants lived, was so great as to include with the tropics all those reasons in which the bones are found."
Quote 3: “Gun control works great for the people who are law-abiding citizens and it does nothing for the criminals, and all it does is put the people at risk.”
Let's start with the obvious approach--just copying them into Google search and seeing what we find. (Actually, what I'm going to do is to remove the quote marks to do my search--the word order preference of Google's search algorithm will find the quote, if it exists, and any near misses.) For example:
[ Do you want to know who you are? Don't ask. Act! Action will delineate and define you ]
The results are interesting. There are lots of results that tell us that this IS a true Jefferson quotation. However, the first result is from Monticello.org, which is (I checked--so should you), the official site of Monticello, the famous home of Thomas Jefferson. The have a full page just on this quote, debunking it as an actual Jeffersonian saying. They helpfully point out that they couldn't find it in all of the sources they checked (and they list the sources--a great practice if you're doing reporting on topics like this), but they DID find it in a Polish authors writings that were first published in 1988! (Specifically, Witold Gombrowicz, 1988: Diary, Vol. II (Chicago: Northwestern University Press, 1988), 130.)
There's so much business in fake Jefferson quotes that Monticello.org has done the world a service and written up a collection of their Rules of thumb on How to Spot a Fake to determine if a Jefferson quote is real or not. This page is basically an introduction to text style analysis, and well worth reading. Among other things, Jefferson basically never used contractions--e.g., "don't" in the above quote, or the use of anachronistic language... such as "gun control."
Quote #1: FAKE!
Same trick with quote #2:
[ That the obliquity of the ecliptic, when these elephants lived, was so great as to include
with the tropics all those reasons in which the bones are found ]
with the tropics all those reasons in which the bones are found ]
This query leads straight to a Google Books result which is a compilation of Jefferson's writings. The Portable Thomas Jefferson has exactly this passage. (Actually, that whole page is really interesting to read. Jefferson had some fascinating ideas about what kinds of animals might exist in North America.)
Quote #2: REAL!
When we do the same thing with the 3rd quotation, we quickly find that it's been widely discounted as a fake Jefferson quote, but is still being used in political speeches by Ben Carson, potential Republican candidate.
Could he have just mangled a real Jefferson quote?
That claimed quote uses the heavily loaded phrase "gun control." If you're going to quote someone, and "gun control" is in the quotation, you're probably not going to get that part wrong.
Let's try to figure out when "gun control" became a common phase. In cases like this, the first thing I turn to is Google NGram. Searching NGram for " gun control " gives us this graph:
Since Google Books includes all of the writings of the Founding Fathers, if it had been a reasonably frequent phrase, it should have shown up in the chart. But it seems that aside from the use of "gun control" during World Wars 1 and 2 (always in the context of how to guide and point real guns in battle), the modern use of the phrase "gun control" seems to have started in the late 1950s or early 1960s.
A quick search for:
[ etymology "gun control" ]
gave me a number of sources (e.g., WordOrigins.com on "gun control"), all of which seem to concur about the introduction of the phrase (in the modern sense) around 1960.
But I thought I'd check a long-running newspaper archive for this as well. I settled on the LA Times (after checking a couple), partly because they have a tradition of writing about "gun control" and also because they have a decent advanced search function that allows one to sort the results by "oldest first."
Sure enough, running the search for "gun control" there on their papers from 1881 until 1981 showed exactly the same pattern--before 1960 it was about military fire control; after 1960, it's a political phrase about gun ownership.
Based on these quick scans of the phrase over time, I'm going to conclude that this possible quotation isn't just mangled...
... it's FAKE!
Search Lessons
1. Pay attention to the angle of the light and shadows in photos. Simple tricks like drawing in the light lines can tell you quickly if the photo has been composed out of pieces.
2. Note the presence of identifying marks on suspect items. There's no way a fighter with a Greek roundel on its wing would be buzzing Philly city hall.
3. As always, searching for tools to help do analysis tasks is a great trick. Before starting this Challenge, I had no idea that an analysis site like ReviewSkeptic existed. Now I have another tool to verify / validate things.
4. Checking quotations is tricky: double and triple check your sources. For these quotes we got lucky. We found one in the writings of Thomas Jefferson, and were able to discount the others fairly easily. But behind each bit of research is a deeper validation of the sites you're reading. Here, Monticello.org is a really authoritative site that gives excellent analyses (including source citations) for all their work. In the case of the 3rd quote, I wanted multiple sources telling me that "gun control" was born (in the modern sense) around 1960. It's a bit more work, but it's pretty compelling.
Another Challenge tomorrow! In the meanwhile... Search on!
---
Update: Regular Reader Passager points out in the comments that I have an error in my 3rd quote. He writes that:
"That the obliquity of the ecliptic, when these elephants lived, was so great as to include with the tropics all those reasons in which the bones are found" is fake. For the real quote you have to replace reasons by regions. Knowing a little astronomy, I noticed the phrase with "reasons" had absolutely no sense, just checked with "regions" on monticello.org and found the right quote.
Ah... Sorry about that. I wish I were that clever. It was just a typo on my part. Moral of this story: If you're going to write about real and fake, be careful to copyedit your prose with care!
Subscribe to:
Posts (Atom)
So... This IS tricky.