Friday, February 12, 2016

A syntopical style of online reading

Where did the morning go?  

I ask myself that question at least twice a week, wondering where a couple hours of my day has gone.

I finally have the answer: I think I’m doing something that’s different, and a little unexpected… I’m “syntopically reading.”   

And I mean that in a very particular way.   “Syntopical reading” isn’t just reading (I’m not just reading the latest John Le Carre novel, although I’d like to), and I’m not just hunting around searching for interesting / cool stuff on the net.  Instead, what seems to be happening is a kind of combined semi-directed investigation on topics of interest.

This is probably one of the biggest surprises from my omphaloskepsis into the depths of my time management. 

As you probably know, I carefully write up what I do each day.  The record has around a 5 minute accuracy and is pretty complete (and before you ask, no, I don’t write all THAT down—there are some categories that show up as *personal* with no further breakdown required). 

Yet, on a typical morning, I’d see:

     4:30 – 5:30 – writing (work, conference paper)
     5:30 – 7:00 -- ?
     7 – 8:15 – prep kids for school; breakfast

…and I’d wonder what happened in that gap.  Why couldn’t I remember?  Don’t worry: I’m not suffering from early onset dementia.  But these breaks are very dream-like gaps in my memory—these episodes are times of nearly continuous activity, but difficult to reconstruct what the point of that time was after the fact. What WAS I doing?  

So I finally broke down and started interrupting myself at 5 minute intervals and writing down exactly what I was doing during those otherwise unaccountable times. 

Found out that it’s a whacked combination of web surfing, reading (online and offline), moving data around, web searches, writing tiny programs to transform text from one format to another, etc etc etc.  

It was all just a ton of apparently arbitrary activities.  Hmmm. What's up with that?  

So then I started writing down the goal I was working on at each 5-minute interruption, not just what I was doing.  The goal timeline looked a bit like this: 

      7:00 - 7:05 - looking up what "Baidarka" is 
     7:05 - 7:10 - looking up "chines" are on kayaks 
     7:10 - 7:15 - checking a map for a nearby ravine 
     7:15 - 7:25 - reading an online book describing
                   traditional Inuit hunting practices in Alaska
     7:25 - 7:35 - looking for and scanning scholarly articles about Inuktitut

That’s when it hit me: I was doing something that I didn’t have a name for.  This wasn't just me skipping around without a point, instead it was a kind of intense, focused behavior that I couldn’t recall because I didn’t recognize it: it was un-nameable.  

My friend Tom Erickson mentioned to me that Mortimer Adler’s notion of “syntopicalreading” is exactly this (although he meant it about ordinary print on crushed-trees kind of books). 

It's a way of reading on a topic that is both broad and deep, covering many different kinds of resources and content types, that leads to an understanding of a topic that is synthesized from all of the materials just read.  

Well, now I have a good description of what I was doing, and a name for it.  Syntopical reading is what I find myself doing in those time gaps.  I read one web page, then do a search to understand more about it--I switch media types, the kinds of things I'm reading, and topics all swirling around.  That often branches to another topic, and another, and another.  

What I find interesting is that it’s NOT all just time-wasting link-following indulgence; instead syntopical-time is when I find myself going deeply into a topic (yesterday’s topic, I discovered from my notes) was trying to understand the connections between French chansons, Gypsy melodies and Balkan scales.   Sure, I sometimes get side-tracked onto YouTube videos, but even they turned out to be important resources for concepts like “Gypsy melody” or “Balkan wedding music.” 

This mini-syntopical-time started as a search to get background information about a band I was about to hear (“Rupa and the April Fishes,” if you’re curious), which led to chansons, and I was off on thematically linked sequence of readings and searches. 

Oddly enough, even for this inveterate notetaker, I find syntopic behavior to be so engaging that I completely forget to take notes.  If I find an extraordinary thing, I might write that down… but the process is wholly absorbing as I switch from reading something, to looking up something I didn’t understand, which often leads to yet another thing. 

If the topic is work-related, the syntopical reading also often involves picking up data from one place and changing it into something else (an activity that takes up a big fraction of all my analysis time)… but when I’m in syntopic-mode, this is all in service of working towards a larger understanding.  I’m not just plugging away on a tiny nit of a problem, but really working the edges of perception to see if I can grok an entire picture at once. 

These syntopic times are, like regular reading, both absorbing and illuminating.  I wonder if this isn’t what 21st century reading is really all about.  I think of monks who spend lifetimes working slowly through manuscripts to gain a deeper knowledge.  That’s one kind of knowing, one way of looking at knowledge over time.

My sense is that syntopic time is qualitatively different: when I’m engaged like this, it's a flow experience and I feel as though I’m moving fast, kayaking over the knowledge stream, looking to pull all of the things I read and data I have into a single, unified cohering understanding.  It doesn’t always work out the way I expect, but when it does, it’s creates the sense of being in the flow, swept away into the river of ideas, and not of this earth.  

This syntopic blend of online reading and search gives a kind of reading-in-depth that hasn’t been possible before.  

No wonder I lose track of time. And what glorious time it is! 

Wednesday, February 10, 2016

Search Challenge (2/10/16): Why an arc of lakes?

You know that odd sensation you get... 

... when you first notice something that you should have seen before now?  

I mean, sometimes you'll look at something you've seen for years, and then a new feature just sort of pops out at you.  

Ever have that happen to you? 

It happened to me recently when I was looking at a map of North America.  We've all seen this about a zillion times, and yet, this time, I saw something different.  Here's the map as we usually see it: 

From Google Maps. The blue dot in California shows you where I am. 

And THIS time, it struck me that there's a big line of big lakes stretching across Canada and into the US.  Here's another way to look at it that makes the arc of lakes very clear: 

Here's another version of the same map.  (I've labeled some of the lakes to save you a couple of minutes of searching.)  

Image from Google Earth, with my annotations of the major lakes on this arc.

See that?  Although I've looked at maps of North America just about my entire life, I'd never noticed that the lakes line up so neatly. This striking observation leads to today's Search Challenge. 

1.  Is this arc of lakes across the top of North America just an accident of geography?  Or is this actually the consequence of some gigantic geological process?  

I suspect I know the answer to this, but I'm going to enjoy the search process along with you. 

The Search Challenge here is "how do you look for something like this?"  If you're not already a geologist, how can you find out enough about the topic to ask a reasonable question of Google?  

Let us know your answer... and your process (including the most useless time sink diversions that take you off the path to the answer)!  

I'll be reading the comment stream this week, and will chime in from time to time.  

This will be fun.  

Search on! 

Tuesday, February 9, 2016

Answer: Mysteries at Universities

These three images need different skills... 

Remember that our Challenge this week was to figure out where these pictures came from and what's going on in each of them.  All you know is that they come different universities AND that I took them.  

1.  What IS this strange animal perched on the edge of the building?  And what's the story that led to it being installed on a university campus?  (Yes, you could call it a gargoyle, but can you be more specific than that?)  

If you zoom in a bit, you can see the post in the lower left of the image says "Arizona State University."  A query search for [ Arizona State mascot ] tells as that it's... a Sun Devil!  Interesting, but not quite right.  On the other hand, in reading around a bit in this list of hits, I DID find that the

I then checked the EXIF metadata, and found the lat/long is:  33.450469, -112.066600.  I went to Google Streetview and found a somewhat better image, clarifying that it really IS a cat (and not a dragon, or something similar).  

Since these are ASU (Arizona State University) buildings, I tried the query: 

     [ Arizona state cat statue ] 

 and then checked images.  That's when I saw a beautiful image that was MUCH better than mine.  If you click through and visit the page, you'll find the whole image by Cobalt123.  

What's striking about this closeup is that there are obvious diamonds on the back of the cat.  What's up with this? 

So I did the query: 

     [ Arizona cat diamonds on back ] 

BUT as I was typing the query, this suggestion popped up: 

I can't pass up a suggestion like that (especially when it's so close to our goal).  Lo and behold, I discovered that there is a baseball team in Arizona called the Diamondbacks... and, there are a lot of pictures of cats in Arizona Diamondback uniforms.  That's slightly funny, but why so many?  

When you do the query [ Arizona diamondbacks cat ] you quickly learn that their team mascot is D Baxter, the Bobcat of the Arizona Diamondbacks.  

And that, my friends, is as far as I got.  It's clearly a cat, and it might be a reference to the local baseball team, but I wasn't able to run it down.  (But I suspect a connection here...)  

2.  What's this odd pattern of bricks over the window?  Why is it there?  

Interestingly, Remmij found the answer to this Challenge by using the "crop the image to just the brick pattern" and then doing a Search-by-Image for that.  

I have to admit that when I tried that method, it didn't work for me.  

BUT there's still hope.  You know that I took the photo, and if you use the metadata viewer, you can figure out that this picture was taken on February 28, 2012  at 9:34:09AM.  (Note that there's no lat/long metadata in this image.)  

How can you figure out where I was on that date?  

Since I'm a bit of an academic kind of guy, I have a home page that lists most of the talks I've given over the past several years.  A search for:

     [ Daniel Russell home page ] 

takes you to my page, and a Control-F for 2012 reveals that I was visiting Princeton on February 28th!  

Now the search is much simpler:  

     [ Princeton brick pattern ] 

search in Google Images leads you to another picture of the bricks on the Princeton Computer Science department web site.   If you then modify the query to be much more specific...  

     [ Computer Science brick pattern Princeton ] 

you'll quickly find the page that explains it all.  

This brick pattern is on the west wall of the CS building and was built into the wall in 1989.  The pattern is actually a bit pattern of five 7-bit ASCII values.  The vertical lines to the left and right are "framing bits" (that is, they just outline the pattern).  So the pattern reads (top-down) like this: 

1 0 1 0 0 0 0   P
0 1 1 1 1 0 1   =
1 0 0 1 1 1 0   N
1 0 1 0 0 0 0   P
0 1 1 1 1 1 1   ?

If you look up [ASCII] you'll find a table of values that map from the bit patterns to the five text characters: 


And if you look THAT up, you'll find that it's one of the great (and as yet unsolved) questions in Computer Science!  

3.  What's this? Where is it? And how long has it been there??   

Here, there's even less metadata--no location, no time stamp.  So I did the simplest possible query based on what I know: 

     [pooh house tree stump university] 

And that leads quickly to a bunch of pictures of this small "House of the Pooh Bear" which is on the Harvard campus near the Science Center.  As this Quora article points out, it's been around for a while, the exact origins are lost in history.  But it IS a little bit of whimsy on the Harvard campus. 

Search Lessons 

1. Not all questions are easy to answer... though you can get close.  When I wrote the "cat" question, I had assumed that it would be straightforward to look it up.  I was counting on the ASU student newspaper to have an article on it... but no.  I spent about 2 hours trying various things, none of which worked.  (You have to stop working on the Challenges sometime!)   

2. Remember that other information sources can be useful to triangulate on the answer.  In this case, knowing that I was in Princeton on the metadata-derived date of the second photo made the search fairly straightforward.  (Of course, actually figuring out that the brick pattern was an ASCII code required some digging.)  But the bigger point is that information from another source can vastly simplify your primary search challenge. 

3.  Sometimes a simple query can get you what you need.  That's the lesson of the tree-stump house for Pooh Bear.  I was surprised at how simple that was.  I guess I somehow thought that ALL universities have locales out of the Christopher Robin / Pooh Bear books!  

For Teachers

One of the things I always tell my teachers (and try to enact in practice myself) is that even simple questions are sometimes very hard to answer.  Even though I thought the Arizona cat would be simple, it really wasn't.  Surprise!  This will happen to you.  Moral of the story:  Pre-test the search questions you give to students, lest you too be very surprised by what you (or they) find when doing the research! 

Search on! 

Friday, February 5, 2016

Teaching beginners how to search--lessons for teachers

Next time I won’t react with such a blank stare. 

I was surprised when a student in my Internet Search Skills class at the public library asked a question 30 seconds into my class:

 “Umm… when you say that the presentation is linked off your home page… I need to know… What’s a home page?”

Now my home page isn’t exactly hard to find. It’s the first Google result for  [ Dan Russell ].  A click takes you right there.  

But in the context of a “how to use Google better” class, it was surprising to find a student that didn’t have any idea what a personal home page was. How was it possible to live in the heart of Silicon Valley and NEVER hear about a “home page”? 

Things were about to get more complicated. 

I found 4 students who didn’t know you needed to have spaces between the query terms (e.g., [ DanRussellGoogle ] vs. [ Dan Russell Google] ). I found students who couldn’t accurately click on a query box and enter the search.. and I simultaneously found students who have been using advanced operators in their searches.  

Hmmm.  This was going to be a challenge.  

I was teaching a class of twenty students how to do basic internet search.  My class is offered regularly at the Santa Clara public library, and attracts students with a wide range of search skills.  I’ve taught this class about 25 times, each time with a broad cross-section of students.  I’ve had classes of kids (ages 7 to 17), classes of older adults (ages 55 to infinity), and mixed classes with both kids, parents and older learners.  In other settings, I’ve also taught advanced search skills classes to school teachers, librarians and even rocket scientists at NASA Ames. 

But this class of real beginners crystallized my thoughts about teaching search skills.  Here are some things to keep in mind as you think about teaching search and basic information skills for the average student / user.

Screen layout and behavior is really complicated to someone just learning.  To searchers without a lot of practice looking at web screen layouts, even something as simple as the default Google home page STILL has too many places to look.  When you’re teaching beginners, you really have to show them where everything is and what those things are called.  They won’t remember the names, but for the duration of the class, you can direct attention to specific places on the screen. 

As one example of complexity, a student was wondering why there are three places to enter a query (toolbar, chrome search and address bar).  I pointed out that they’re really different.  “But they’re not,” he protested, “when I type my search into any of them, they all do a search!”  And he’s right.  Except when it doesn’t work.  An entry like [ sofa ] in the address bar does a perfectly fine search.  But an entry like [ whitehouse ] might not—it can take you to landing page if it autocompletes. 

My favorite example is when a student entered “blackberry” into the Address bar (in IE) and was taken to, the makers of a particular brand of smartphone.  “Why did it do that?” she cried, “I just wanted more information about pies!”

Her previous query had been [ blueberry pie ] (also in the address bar, which works just fine in IE).  When she entered “blackberry” into the Address field, the behavior was utterly different, and she didn’t know why. 

Yes, she didn’t understand the difference between the Address bar and the chrome search box.  Question is, should she have to? 

As you teach, take the time to give orientation.  

Even managing the typing is hard for some.  We don’t think about this much, but older folks and people who haven’t used computers much find entering text to be incredibly tricky.  They don’t know about the subtleties of where to click, when to press-and-hold the mouse, or even how to select a whole word.  And once an input zone is selected, they’ll accidentally click the mouse one more time, or move the input focus without even knowing they did something wrong. 

For my true beginner classes, just teaching the mechanics of how to enter search terms is a major challenge.  As a teacher, it takes a huge amount of patience to slow down enough to guide the hand, coordinating the arm, wrist and finger motions—but it’s absolutely necessary.  If the student can’t select and enter their search, they can’t do anything.  

For students having these problems, you need to slow down and manually show them how to move, how to click. 

Basic find (control-F) isn’t widely known.  I ask every class I teach if they know how to locate (or find) a specific word on the page.  I illustrate with an example.  I’ll go to a random page (say, and then spot a word that appears on the page (but not in caps).  Suppose it’s the word “energy.”  Then I’ll ask  “can anyone find if the word ‘energy’ appears on this page?”  I go through this elaborate process to avoid cuing them with words like “find.”  It usually take a minute or so before someone finds the word.  I then ask that person how they did it—usually they say “I used control-F.”  That’s when I ask how many people know this trick?  (Or using the edit menu to do a find.) 

Consistently, I’ve been finding that right around 25% of the class knows this basic skill.   More generally, when I ask this question in larger groups, it drops to 10%.  Somewhat to my surprise, I also find that even when I teach technically advanced audiences, not everyone knows this.  Almost all librarians (skilled searchers) know this—but surprisingly, it seems to be around 85% know this, which continually baffles me. 

For the love of Pete, teach EVERY class about Control-F / CMD-F!  

Scoping a search to focus on the topic of interest isn’t a common skill.  I always ask my class to do some searches of their own—things that they really want to look up.  As the class works away on this, I go from person-to-person and talk with them about what they’re trying to do.  I want to understand their intent (in their own words) and see what they’re doing. 

Some times a student will say “I want to learn about the Civil War.”  And then do a query like [ Gettysburg ].  That’s a great search to find out about that particular battle or the famous oration, but not a good search for general Civil War information.  For beginners, this isn’t a crazy approach—it’s a way to get a start on a topic when they don't know what else to ask. The problem is that they’ll get stuck in the corner of the topic they already know, and not learn about the rest of the topic.  

Contrariwise, a student might start that same search with [ Civil War ], which is a good beginning, but then they need to figure out how to reduce the scope of the search.  That’s a problem with such a huge topic—how to focus in on the part (or parts) you really want.

When you think about it, a good part of expertise in learning about a topic is coming to understand what the scale and scope really is.  And doing a search on a topic, especially for a beginner, a wildly divergent thing to do. 

Talk about scoping a search explicitly.  Not everyone knows... 

Choosing query terms is not an obvious skill.   It’s striking to talk with a student about what they’re trying to discover and hear the mismatch between what they’re thinking about, and what they actually type in for search.  For those of us who have used search engines for years, it’s so obvious as to be transparent.  But for someone who’s new to using a search engine the question is often “how do I start my search?” 

“I just want to find a cheap hotel for my vacation.”  He’d entered [ cheap hotel ] as the search query, and was puzzled to not see any mentions of hotels in Chicago, where he was planning on travelling.  I suggested he add the name of his destination, and all was well. Although he was still surprised why Google didn’t understand that  he was interested in Chicago.  “I’ve been looking for all kinds of things to do in Chicago—didn’t it see me doing all of that?” 

This kind of dropping of terms implies that some beginning users see the search process as a kind of conversation rather than as a stateless transaction. 

Have a unit just on "query term selection."  

 “Google” as a concept is often confused with “the internet” or with “the browser.”  We see this kind of misconception all the time in the complaints Google gets in the mail.   (“Why can’t you fix my phone bill?” a dissatisfied user will ask, not realizing that we’re not AT&T, we’re just the search engine that got you there.) 

But people often confuse the tool with the embedding ideas.  When I worked at IBM we’d constantly get complaints from people about why they kept getting viruses on their laptops.  Of course, that’s not IBM’s issue—but to the end-user it doesn’t matter—IBM is the logo they see on the laptop and hence who they know to contact (of course, it’s not Lenovo, but that was then).  When you think about it, who SHOULD own the virus problem? 

In a similar vein, Google is the home page for many people—it’s the first logo they see and recognize on a web page, so the conflation of “the internet” or “the browser” with Google is natural. 

Of course this raises Cain with a user’s idea of how things work when they learn that you don’t need a Google email account to use Google services, or when they get an instruction to close their browser, but the browser isn’t on a Google page.  (“How can I close the browser?  It’s not showing me a Google logo because I’m looking at another website.”  It sounds incoherent, I know—but this is the way I hear many people reasoning.) 

It’s difficult to learn a coherent model of how a search engine works.  When you use something, it’s good if you can have a coherent model of what it is and what it does when given certain inputs.  Unfortunately, Google behaves wildly differently when you give it slightly different, sometimes imperceptibly different queries. [ hunt ] vs. [ hunting ] are pretty different; but something like [ Ron ] vs. [ Ronnie ] are sometimes bafflingly different to beginners.  For users with some experience, the difference is clear.  But when a beginning user does a search and then doesn’t  find one of the terms on the page, it’s a mystifying moment. 

I don’t mean to paint a dire picture here.  For the most part, students are incredibly happy with Google—they find things they could only imagine in their dreams. 

At the same time, for a portion of the population (probably the vast non-technical majority), using a search engine isn’t like falling off a log; it’s a highly technical skill that’s full of mysteries and not a small amount of worry. 

What’s perhaps most worrisome is that there’s no good way for beginners to climb up the learning curve.  Of course, they could look for online materials and tutorials, but these people are beginners at search, it’s not the first (or last) thing they’d consider. 

Of course, they could take my class.  And while I always have a full class (and always with a waiting list), just my teaching doesn’t scale.  (Which is why I offer the scalable version of my classes at -- that's always open and available for free.) 

We really need to figure a way to get people over the hump of first time use to become reasonably good, reasonably proactive searchers.  Once they can search for their own help materials, we’ll have a self-sustaining population of new users.  We just need to figure out a good way to get them over the initial hurdle of not-knowing.