Archive for January, 2007

Ambient information as a replacement for classrooms


The small one is down for a nap right now, so I can write. Little kids look at the world in funny ways. The other day she came up and proudly proclaimed that “my throat is burping!” Turns out she had the hiccups.

In a few years, this kid will go off to school, sit down in a classroom, and Learn Things. Continuing on from the last post, I’d like to examine the idea that the classroom she’ll grow up to be in might not be a classroom at all. The reason: a growing surge in ambient information.

When people are asked how the internet will change our educational lives in the next decade, this is probably what most of them think of. With increasing amounts of smaller, cheaper, faster technology and connectivity, information becomes pervasive, accessible wherever you ask for it and most places you don’t (witness the rise of context-sensitive advertising).

Since it’s easier to use data, more data comes into being; with increased amounts of data to understand, we begin letting our computers do the understanding for us via the explosively maturing semantic web. Tapping into many minds other than your own will become commonplace (how many of us already look online for consumer reviews before we buy a computer or refrigerator?) and “queries to the metabrain” will tap into both human and machine knowledge.

So far, this is the acceleration of business as usual; no computer-literate teenager would be surprised by what I’ve just written.

The tipping point comes (and is coming now) as we go post-digital, taking the technologies for granted and using them as natural ways to live our human lives more fluidly. The heralded “classroom without walls” will give way to lives without classrooms. We’re learning how to learn without placing our bodies in a specific geographic location. Next up is learning how to learn without placing our minds in a specific mental frame - “in school” will be an archaic phrase because we’ll be simultaneously in and out of school all the time.

In kindergarten, a good teacher notices something you’re doing and takes the opportunity to turn it into a mini-lesson without pulling you out of the flow of your activity. We haven’t been able to scale this into higher grades too much. While one teacher might be able to cover the knowledge of 25 five-year-olds well enough to spot and teach lessons to them all, it’s the rare human being that can cover the depth of knowledge that 25 ten-year-olds will pursue, let alone 25 university students delving into the depths of 25 different subjects. (We’ll talk about the ability to mentor someone both in your specific field and in a depth not your own in a future post on apprenticeship.)

But what if the teachers didn’t have to do this by themselves? What if we had “smart recognition” of places where context - little mini-lessons - could be inserted? And what if, instead of waiting for a teacher to notice and instruct her, the student had the ability to tap into the needed data stream in the field (or better yet, it was done automatically)? If lessons were taught in context, we’d let kids out of the classroom so they could experience more contexts to learn from. Teachers and parents would have to make sure that the lessons were high-quality and level-appropriate but not limited or biased in potentially horizion-narrowing ways, but these are skills that our children (and our computers) should be learning anyhow.

Along with bringing children out of the (increasingly nonexistent) classroom, this would bring adults back into it. I didn’t have lessons or textbooks at my job this summer; there is no course called “How To Be An EE Intern 101.” Schools give people enough tools for them to start learning on their own by asking questions. “What do I do when this part doesn’t work?” “What does this chunk of code do?” “Did you run into problems setting up that micro last year?” You used to have to either sift through a book to find the 3 relevant pages or wait around for an older engineer to give you the right information. Taking the brains-as-factories analogy, this is the equivalent of industrial production before the advent of Just In Time manufacturing. Idle hands. Wasted time. Interruptions in the workflow. General underutilization of potential.

Just-In-Time ambient software will bring adults “back to school” by acting as a / supporting a real mentor peeking over your shoulder or a helpful buddy at the keyboard. “That’s the bypass cap, and we use them because…” You learn on the job, within the job, without being pulled out of your task to go sit in a classroom. Computers will suggest where to look for help if they can’t assist you themselves. “It looks like you’re writing a resume. Remember to ask Leslie to check it - here, I’ll pop up a blank email for you.” (The trick, of course, is to do this without becoming annoying or intrusive.) Soon it’ll answer questions I didn’t even think of asking. “Mel, why are you drawing JPEGs of your equations? Ever heard of LaTeX? Here, gimme 15 minutes and I’ll link you to a crash course…”

There are many things best done in classrooms; there are many things best learned through formal instruction. But there are equally many subjects and people that don’t take naturally to classroom instruction, and I expect to see these modes of learning freed from this constraint. Classrooms will exist because they need to exist, not because they’re the “only way” to do something.


Tipping points for the online autodidact


I believe that everyone in the world deserves a fair shot at becoming a good autodidact in whatever subject they wish. (I also realize this is a broad, idealistic blue-sky dream, but we can talk about that later.)

The internet’s been helping mightily in this during the past five years. When I started high school, MIT’s Open Course Ware was just beginning, Wikipedia was still a gleam in Jimmy Wales’ eye, and in terms of quality online educational resources past the K-12 level, that was mostly it. Now we have lovely things like the resources described in this Lifehacker post. There are widely available free educational materials, some of increasingly impressive quality. We have syllabi and textbooks reprinted online, occasionally lecture notes, and even homework assignments and lecture tapes themselves.

So far we’ve predominantly been discovering better ways of representing standard course materials on webpages. This corresponds to the first phase of a new form of media: that of a new way to do old things. “Cell phones are like landlines without cords.” At some point a paradigm shift occurs, and the new form of media isn’t “Old Media With Feature X” but a separate thing in its own right, and gets used in apps that weren’t even on the radar before; smartphones, SMS, and location-based messaging, for example. The two phenomena snowball into each other, and soon enough the world is chang’d, at least in some small way.

We’re seeing the beginnings of a transformation for education as seen through the internet corresponding to just this kind of shift, where we move beyond the “put the old class material on html” and into… what? I don’t know, but here are three trends I’ve got my eyes on; ambient information, communities of apprenticeship, and public reflection. I’ll cover each of these in a separate post later on, with the disclaimer (courtesy of my friend David) that it’s tough to predict a horizon that’s shrinking towards you; a few years from now we’ll probably look at these posts and laugh.

In the meantime, take a peek at Carmun. It’s a web 2.0 startup designed to help students track and share reference lists - for instance, if I’m taking a course in educational theory, I can put articles and books from class in Carmun and they’ll be there for seamless referencing and bibliography creation later on when I’m writing my final paper. Better yet, if you take the same course next semester and ask me about good books to read on the subject, I can send you my reference list via Carmun (with links to the original papers and everything) instead of copy-pasting my pdf’s endnotes into an email (where you’d have to laboriously re-search for each article in JSTOR anyway). You can write notes on papers and books, rate and tag them, and generally use it as an all-purpose reading list for whatever you’re interested in learning.

Still a few rough spots to smooth out and some features I’d like to see (the first of which would be an API - I want a LaTeX plugin!) but overall pretty nice. Cool people, too. I’d like to learn more about how they’re doing this.

The small cousin has woken with a wet diaper. Time to go.


The flesh is willing but the spirit is a total pushover


What the title means is that last night, despite the protests of my body, I pulled my first true allnighter in two years. Thus ends an era of triumph - I was trying so hard to sleep regularly, but when I get really sucked into a task, I don’t do anything else until it’s done. I haven’t allowed myself to do that in a long time, because it’s so dangerous to do so. Hopefully this isn’t an indication of my lapse back into seriously wacky sleep patterns (two allnighters a week was just not a good idea).

Why the allnighter? Our design of echo is a finalist in the Japan Design Competition, and we needed to submit new slides to Osaka so the judges could preview a more detailed version of our original entry. I am one of “the artists” of the team (the other “art dude” being the talented Joe Kendall) and also apparently subject to Murphy’s Law while travelling, which led to me spending two days intended for drawing en route to my aunt’s house instead. The long story involves a busted transmission, a late bus, and spending the night on a couch approximately 60% as long as I was. (Surprisingly non-uncomfortable.)

Those five images represent 16 hours of nonstop work, by the way. Add a 17th hour driving around town in a panic trying to find internet before the deadline because wireless has died 10 minutes before I was ready to send the files. The library was closed, Starbucks cost money I didn’t have, and… the Olin library was open. Yay Olin library! I did get a few “Mel? What are you doing back here?” comments, though. (Dear staff and faculty - I swear I’m on vacation, kind of. Sometimes.)

Those five slides also represent the reason why I’m not a professional graphic designer (although it’s reinforced my resolve to learn that as soon as I get a better grasp on engineering). In a followup to my “I’m learning web design!” post, echo is also my first “real webpage.” Very much a work in progress. Grainy graphics, non-elastic images, not quite cross-browser perfect, almost w3c-compliant… very obviously the work of an n00b (you don’t even need to look at the code - just look at the color scheme) but at least I’m making something. And I’ll be taking a second pass to fix it up before we leave for Osaka at the end of January.

If I’m going to get up in 5 hours, I need to sleep now.


Long-distance triage: sometimes it’s good stuff.


People with office hours or students from Ozgur’s Distributed Engineering Design course will especially appreciate this.

Collocation Blindness in Partially Distributed Groups: Is There A Downside To Being Collocated?

Basically, remote teamwork functions in favor of specialists with rare skills. Because of the distance, communications with them are perceived to be of higher value and allows the specialists to be more pragmatic about distributing their time instead of simply spending it mostly on the folks nearby. This is contrary to conventional wisdom which says that in-person contact is almost always the best way to go.

What does this mean?

Distance allows us to step back and see the larger picture of things. We’re more removed from a situation, but the tradeoff is that we’re better able to gauge its priority and act accordingly. This is why IT helpdesks tell people to email in tickets instead of always stopping by; it’s why people say “email me if you have any questions,” why customer service representatives are now available via chat (you can handle multiple chats more easily than multiple phone conversations), why some professors hold office hours over AIM, and why I get a lot done when I shut my room door or go home for vacation.

It’s one reason working from home is so popular. I get to decide what’s important and what I’ll do next - not a supervisor breathing down my neck or customers clamoring that their problem needs to be fixed now, now, now. This has a particularly big effect on suckers like me who can’t resist a request for help; those desperate puppy-eyes make you feel so darn guilty that you push back important things to solve someone else’s small but immediate problem. (The warm fuzzy feeling is worth it, though.)

This also means we have to trust specialists to prioritize things the right way. If I don’t think the medic will prioritize my grandpa’s stroke over Johnny’s bleeding arm, I’m going to make a scene until he gets the attention I think he deserves. It also means we have to agree that the way they’re triaging is appropriate, since we certainly can’t all go first. If Johnny thinks his gash is horrendously vital (maybe he’s a hemophiliac) he’ll start shouting as well.

Finally, the specialist needs to prioritize reponsibly and in a rational manner instead of abusing their magical long-distance triage powers. Even if our hypothetical medic is afraid of stitches, she needs to patch Johnny’s arm within a reasonable amount of time even if it’d be “easier” to ignore his yells until he faints from loss of blood. Similarly, just because someone can’t remind us at dinner to complete a task doesn’t mean we have license to blow it off. (I acknowledge I slip up plenty on this last criteria, but I’m really trying to work on it…) If, on the other hand, Johnny’s got a trivial paper cut and is screaming for a lollipop he shouldn’t have before dinner, then the ability to ignore Johnny is very good.

Speaking of which… I’m slowly growing less and less behind in my pile of work. Two major tasks to complete before I’m 100% on-schedule for winter break again! (Ah, unscheduled family trips. I wish I could tell my parents to send me an iCal, but that would just make them confused or mad, depending on on whether they know what an iCal is or not.)


Another short eclectic reading list


I swear I must have posted this before: How to live on 24 hours a day. The only trouble is that it’s written with people with the exact reverse problem of what I’ve got. According to this book, most folks fritter their time away and don’t understand how much more they could and should want to cram into their 24 hours.

Another read that made my mind spin: The underground history of American education by John Taylor Gatto. He used to be the New York Teacher Of The Year; now he denounces the same educational system he once worked for. From reading his book and some of his essays and speeches, he’s a radical (or at least presents himself as such). I definitely don’t agree with everything he says, and I don’t think the extremist viewpoint he proposes and the fiery voice he speaks in is the best way of getting things done. I do greatly admire his eloquence and courage in saying such things, though; this is one man I would love to meet.

Gui recommended Zen and the Art of Motorcycle Maintenance by Pirsig. It’s good; I read it on the bus to the Media Lab when I was working there two summers ago. This is one of those books that slowly crackles away in the back of your mind as you read it; you don’t notice until you look up, something goes snap! and you shake your head and wonder what’s rattled loose to make you see things so differently.

Writing Down The Bones is a book I’ve bought and given away multiple times. Short, poetic reflections on writing itself that will transform the way you write; it’s one of the things on my inspirational writing rotation (Shakespeare is another common appearance) that I read through whenever I start feeling that my prose is growing dead and mechanical.

Flatland: A romance of many dimensions is not sappy in the least; it’s mathematical (and therefore I love it). It’s a story told by the denizen of a two-dimensional world (A. Square) who is trying to grasp three-dimensional space. I discovered this book in high school simultaneously with linear algebra, non-Newtonian physics, and non-Euclidian geometry, and it was pure delight. Speaking of non-Newtonian physics, anything by Alan Lightman (especially Einstein’s Dreams) is just beautiful. Reading Lightman’s essays gives me the same quiet feeling I have when I stare out a window and see thick white snowflakes swirling softly in the lamplight. He expresses the loveliness of mathematics and science in a way I would like to… not imitate, but admire and honor by finding my own way of writing about such things.*

Sequels to Flatland (Flatterland, Sphereland, the Planiverse…) have been written. All the ones I’ve read are good; I read them over a spread-out enough time-scale that I’m not 100% confident in my rating of which are best. Someday I’ll collect them all and do a storybook reading in order - I wish someone would publish a single volume collecting them all, as they’ve done for Lord of the Rings, the Chronicles of Narnia, or even the Feynman Lectures on Physics.

*Lightman teaches at MIT’s Graduate Program in Science Writing. I wish I could just stay a student forever - there are so many teachers I want to learn from, classes I want to go to, schools I would attend…


Clothes shopping: the aftermath


This is the result of the shopping trip.


And this is what I will do to anyone who laughs.

Actually, that was the least painful clothes shopping experience I’ve ever had. Mom dragged me to Macy’s, and by a stroke of fortune we passed by the office for the personal shoppers, where Mom explained my predilection towards t-shirts and jeans to a wonderful lady named Ilana, who proceeded to help me find a suit that was… comfortable. For something that wasn’t a t-shirt and jeans, anyway. I could probably play football in it, if I wouldn’t be strangled by my mother for doing so. And it looked good enough to satisfy my mom.

I’m going to use a personal shopper whenever I need to get “nice clothes” from now on; my sense of fashion is nonexistent enough that I desperately need the help. Hopefully “from now on” means “maybe in 2 or 3 decades I’ll start thinking about clothes shopping again.” Mom still wants to sign me up for that fashion consultation thing, though. I’m working on persuading her that a nice set of O’Reilly books would add more long-term functionality to my repertoire, but this is the woman - though I love her dearly - who signed me up for modeling classes once upon a time. And yes, I had to take them. And no, I’m not kidding.

So there was the suit, and then the second pair of pants, and the dress shirts, and the shoes (can’t we just make black sneakers acceptable business attire?) and the final price tag which makes me hyperventilate and is probably by far the most that has ever been spent on my clothing. This is not difficult, as I’m used to hand-me-downs, discount stores, and free t-shirts as my main source of garb. Shirts over $5 are expensive, and walking around a “normal mall” leaves me in a continuous daze of sticker shock (will someone explain this “$50 for a logo on your shirt” phenomenon to me?)

We didn’t have an awful lot of money when I was small; it carries over into your adult life, and I’m glad for it. It still feels really weird to live in a nice house with a dishwasher and a new oven that works; it feels like we’re too rich to be right. I’m not complaining, mind you. I freakin’ love that dishwasher. And I know that in the grand scheme of things we’re really upper-middle middle class at best. But it makes me feel guilty to see money being spent on stuff I don’t need - going to a nice dinner with friends, getting a space heater for my bedroom (the draftiest part of the house) instead of wearing double-pajamas and socks, having a graphics tablet (okay, that really does make me more productive.) And it makes me uncomfortable buying a nice suit.

I need to stop having this vague guilt over spending money when it’s spent wisely.