Learning Design Patterns as an Alternative Model of Course Design

I’ve discovered Chris Alexander’s work on architecture, and I cannot read it without hearing every line as a statement on problems in course design.  Alexander approaches architecture not through top-down design or bottom-up chaos, but through generative constraints, that is, he begins with the environment and then runs through a “grammar” of building. The design emerges through these constraints the way a sentence arises from grammatical templates/rules (depending on your view of grammar).

I’m new to this, but the way it seems to work is you have a pattern such as “Light on Two Sides“. This pattern notes that people naturally gravitate to rooms with windows on at least two walls. This pattern is made possible by attending to some higher order patterns dealing with the “edge” of the house: Positive Outdoor Space, Wings of Light, and Long Thin House. If Light on Two Sides is impossible in spaces where you need it, you may need to revisit the higher order patterns.

This Light on Two Sides pattern when executed then becomes the higher order pattern to a number of even lower order patterns: Windows Overlooking Life, Deep Reveal, and Roof Layout. In other words the work proceeds in the way you might write a song or form the plot of a book. You write a series of notes that determines the key, which generates more melodies and suggests a bridge in the relative minor key. You add a drum beat which suggests for you the rhythmic structure of the bass. The bass suggests a keyboard hook. Layer by layer the song emerges as a living thing. Everything ends up original, but it is the process of putting one piece in place guides the next piece of work.

In pedagogy how would this work? I have a 150 year old book on teaching I recently read, and scattered through it are comments such as “Place Hard Work First” (e.g. students have limited metal effort to expend, put the activities requiring the most effort towards the front of the lesson). That might be placed alongside a pattern about Peer Learning — and if peer learning is effortful it would suggest the position of peer learning in your course. The use of peer learning might require a higher order pattern about the length of the class or the format of the furniture.

So if the patterns are the same, how do designs end up different? Because you start with different constraints. The same way the shape of the land or relationship to other houses will trigger certain solutions in Alexander’s pattern language:

“Above all, the shapes of the building must spring from the land, and buildings around, like a tree springing from a coppice — it fits perfectly, the moment of inception.”

Your students have certain backgrounds, various strengths. Your institution has certain facilities and your technology has certain affordances. Just as a limited number of grammar rules produce an infinite number of sentences based on the needs of the moment, so learning design patterns combined with the circumstances and aims of instruction can produce infinitely expressive learning designs.

What excites me about this is it is a way to combine research and practice without succumbing to an industrial paradigm. This isn’t “wing it on the whiteboard” or the class as the artistic expression of the instructor. This is a framework which is as rigorous in its own way as any ADDIE-fueled design monster. It’s premises can be challenged. It identifies right and wrong ways to go about things. It would evolve in reaction to new data. It is a distinct methodology which can be shown to produce either good or bad results. But, unlike many methodologies. it seems to me to work in the natural directon of our thought. And as Alexander points out it’s this pattern of a work reacting to itself and its environment that gives it the spark of life.

I am placing some quotes of Alexander’s here to give you the flavor of his thought. They come from random places with no original sourcing, so I paste them here without links. If you find them as powerful as I do the context will eventually present itself.

“In an organic environment, every place is unique, and the different places also cooperate, with no parts left over, to create a global whole – a whole which can be identified by everyone who is part of it.”

“In the past century, architecture has always been a minor science — if it has been a science at all. Present day architects who want to be scientific, try to incorporate the ideas of physics, psychology, anthropology in their work . . . in the hope of keeping in tune with the “scientific” times. I believe we are on the threshold of a new era, when this relation between architecture and the physical sciences may be reversed — when the proper understanding of the deep questions of space, as they are embodied in architecture will play a revolutionary role in the way we see the world and will do for the world view of the 21st and 22nd centuries, what physics did for the 19th and 20th.”

“Every building, every room, every garden is better when all the patterns which it needs are compressed as far as it is possible for them to be. The building will be cheaper; the meaning in it will be denser.”

“I’ll tell you a story. I was in India in 1961. I was living in a village most of the time. I studied that village, tried to understand what village life was all about. And I got back to Harvard, a few months later, and I got a letter from the government of [the town in India], saying ‘We’ve got to re-locate our village because of the dam construction. Would you like to build it?’. I think about 2000 people were being moved. And I thought about it. And then I was very sad. And I wrote back, and I said, ‘You know, I don’t know enough about how to do it. Because I don’t want to come in and simply build a village, because I don’t think that will make life. I know that the life has got to come from the people, as well as what’s going on physically, geometrically. My experience of living in the village is that I do not know enough about how to actually make that happen. And therefore I very very regretfully decline your kind offer.’ And I was actually chagrined beyond measure, that I had to give that reply. But it was honest, and in fact, it was because of that letter that I wrote A Pattern Language. Because, I thought and thought, and I said, ‘You know, this is crazy. What would I have to do, to put in people’s hands the thing with which they could do this, so that it would be like a real village and not like an architect’s fantasy?”

“We are searching for some kind of harmony between two intangibles: a form which we have not yet designed and a context which we cannot properly describe.”

“If you have a feeling-vision of the thing – a painting, a building, a garden, a piece of a neighborhood – as long as you’re very firmly anchored in your knowledge of that thing, and you can see it with your eyes closed, you can keep correcting your actions… It’s not a question of holding onto every little detail, but of holding onto the feeling.”

“From a sequence of these individual patterns, whole buildings with the character of nature will form themselves within your thoughts, as easily as sentences.”

“Nowadays, the process of growth and development almost never seems to manage to create this subtle balance between the importance of the individual parts, and the coherence of the environment as a whole. One or the other always dominates.”

Peak StreamMode

Bonnie Stewart has a great post over at her TheoryBlog on the state of Twitter. The post attempts to pull together the problems of the New Orality of social media, which seems to have somehow combined the worst aspects of conversation and print:

Because lately secondary orality via digital media seems like a pretty nasty, reactive state of being, a collective hiss of “you’re doing it wrong.” Tweets are taken up as magnum opi to be leapt upon and eviscerated, not only by ideological opponents or threatened employers but by in-network peers…because the Attention Economy rewards those behaviours.

This is a pretty important point. Conversation works both because interlocutors share context and because listeners and speakers work hard to see the world through each other’s eyes. Sperber and Wilson, in Relevance Theory, go so far as to say that our default assumption in language is that speakers will always be maximally relevant, and when they seem not to be our default assumption will be that our understanding is incomplete, not that the the speaker was being intentionally obscure or sporadically inconsistent.

In fact, so ingrained is the language instinct to try to see the world through our conversational partner’s eyes that we use this as a sort of trick in pedagogy. Have a student explain something to another student, and suddenly the speaker opens up a whole new level of self-analysis. The words that made so much sense in the internal monologue fall apart as we try to speak them to another, surfacing unanswered questions in things we thought we had down pat. Somewhere the model is wrong.

Alas, whatever the evolutionary hacks are that cause that sudden emergence of dual consciousness. they haven’t caught up to Twitter. And as our contexts become more fragmented, we don’t know enough about individuals to know what seems out of place and what doesn’t. In conversation, I know that Bonnie is very far from racist — if she perhaps disagrees that the cop who shot Michael Brown should not be immediately named and states that my mind is going to try to reconcile that with the model of Bonnie’s worldview I have. And that Sperber and Wilson principle of cognitive efficiency all but demands that I resolve the contradiction without tearing down my entire mental construct of Bonnie. This is why in normal conversation such moments can be so powerful — the cognitive dissonance between a statement and our mental model of the speaker pushes us to build a more complex model of both.

That used to be true with Old Twitter too, but as Twitter has expanded it’s fallen apart a bit. It’s not only that we know so little about some people that the dissonance never arises; it’s also, as Bonnie notes, that the attention economy rewards flame wars, scheduled outrage, and intentional misunderstandings. Find someone above you, wait for an inartful tweet, kick up publicly to the cheers of many. Watch your follower numbers grow. Repeat.

I think there’s an additional issue as well, which I blab on about a bit in Bonnie’s comment section — we’re pushing too much of our output through what I have decided to call StreamMode (that running serialization that sees all things as sequenced speech events) and too little of our output though StateMode (that iterative mode which sees our work as existing as a sort of snapshot of us and our ideas). We used to work in hybrid forms — self-contained blog posts that were serialized to RSS, Flickr collections with new photos feed. Increasingly, however, we are abandoning StateMode altogether. Instead of Flickr we have Instagram, instead of blogging we have Twitter and Tumblr. Everything is placed on a timeline, and very little is integrated in any greater way than “X came after Y”. You can take my location history, interleave it with my tweets, my Netflix viewing patterns, my Facebook likes, my GoodReads additions. It’s all just one big soup of timestamps.

StreamMode has some advantages, but it’s curious how quick it’s swallowed everything. I remember when I first saw the Facebook lifestream idea (the running log of what you had done on Facebook) back in maybe 2006 this seemed very new — this idea that it could all just be stream. Now I don’t even think people realize it was new at one point, or that there are alternate ways of ordering online experience.

These thoughts are too nascent to spend much time on yet, but throwing them out there in case someone has any suggested reading for me.



Using Federated Wiki in the Classroom: Getting Started

This post assumes that you’ve read some other posts on federated wiki. There’s a few dozen on this site if you have not. Click the federated wiki tag and then scroll down to see them all.

If you know what federated wiki is, the following description should get you started with federated wiki use in your classroom.

Make Page of Site Creation Links

Set up a page in your class federated wiki (owned and managed by you) that links to not-yet-existing sites for each one of your students. You’re running federated wiki in farm mode, so going to these sites will create them for the students who go there. The page will look like this:




Have Students Set up Their Sites and Bio Pages

When a student clicks on their link, it will give them a new site with the name you specified. I chose a convention of “first two letters of first name + first two letters of last name” which allows me to quickly identify a student while still giving them internet anonymity if they want it. Here’s what it looks like when the student clicks it:


Under “Pages about Us” have the student put in a name as a link. It could be their full name, their first name, a nickname. Just as long as it is recognizable to you. After adding the name as a link, they click on the link. This new page will be their bio page. At this point I show them an example bio of myself — something relatively lighthearted but substantial.

Students will draft their bio pages. A lot of students will make boring bio pages at first, but here’s part of the genius of wiki — have the students look at other student bios after they are done making theirs, and generally this will help some of the students conceptualize theirs. At the end of this you’ll end up with a lot of very cool bios.



IMPORTANT: After students set up their bios, it’s a good time to have them “claim” their sites with the big “Claim” button at the bottom. This uses a Mozilla-based Persona login that sets the student up as the sole editor of the site. If you forget to do this early, students will end up unintentionally editing other students sites, which isn’t the end of the world, but is a bit of a headache. Have them claim the site early.


Create a “Class Circle”

Now it’s your turn — you have to create what we’ll call the “Class Circle”. This will be a page that students can load to see the work of all the other students in the class — not just in the recent changes feed, but in search results, “twin” notifications, and the like. To make a circle create a bunch of factory drop areas on a page named “Our Class Sites” or something similar:


Now go to that page of links of all the student sites, and for each link:

  • Click the link to go to the student page.
  • Click the link to the student bio.
  • Drag the student bio onto an empty “factory” drop area

This will pull a “reference” to the student site and the first paragraph of their bio into your page. (Note: I did this with the “link launch page” described above to streamline the process and standardize site names, but you could also have student self-select site names and email you the link).

I had 20 students — the process took about 10 minutes. It’s the most time-consuming part of the setup. But when you are done you should have a page that looks like this.



Tracking Student Work Using the Neighborhood

The circle page is pretty cool, because anyone can load it and see all the class activity (to be technical: it pulls class sites into their “neighborhood”). Students can (and will) fork it back into their own sites. Unlike FeedWordPress and other “hub” designs, however, the power to make circles is given to the students as well — the students can easily create their own circle page entitled “English majors” if they want, and pull in all the references to sites by English majors in the class. They can set up circles for their group, or for the three people who always do exemplary work.

Once you have your class circle in place, you be able to track the work of the class through your recent changes page. Here’s a snapshot of it the day after class:


Here I’ve loaded my class circle, clicked recent changes, and am looking at a recent submission by a student on the “redefinition” aspect of SAMR. One thing to note here is how well the form supports a “notes” aesthetic — the student here writes very well, but is allowed to put half-formed thoughts up and questions up to which they can later return.. If the metaphor for the student blog is the personal journal, the metaphor for federated wiki is the researcher’s notebook.

We also see the usefulness of the colored icons here. Scanning this changes feed, we can see that:

  • The student we are looking at right now, with the teal gradient, has been very busy, and has in fact gotten all their work for next week already done.
  • Four other students have done a page on the SAMR model of educational technology impact,
  • Another student (purple gradient) has done the SAMR assignment, although maybe not the “note-taking strategies” assignment.

Since I used a naming scheme (first two letters of first name and first two letters of last), I can hover over these icons and know immediately which student they represent. The teal icon here has a hover text of “krde.mits.wsuv.wiki”, which tells me this is Kristin D’s work.  If we click on the teal icon at the top of Redefinition, we can get her Welcome Page. Another shift-click opens up her bio page as well (click replaces the page to the right of the page clicked, shift-click adds a page in the first empty spot, giving you the page in an added column — it sounds odd, but feels awesome when you get the hang of it).




We can also look at just Kristin’s feed now that we’ve collapsed our “neighborhood” to just her.




Using “Twins” as a Student to See Other Approaches to an Assignment

Reloading our class circle and going to the page on SAMR model, we can start to see how the federated aspect works in the classroom. Any student or teacher can easily use the “twins” notification up top (that part that shows links to older and newer versions) to pull up different student work on the same subject.


The assignment was to find some articles on SAMR and to summarize them. In this case, a day after class, a couple students have found the article they want to use, but not done anything yet. One of the neat things here is I can check on work in progress — see what articles they’ve selected and the like. For the students, one of the neat things is that by seeing other student work in progress, they have some idea of what the target they are trying to hit might be.

That’s enough to get you started. We did more in class than this, but I’ll write up the next part later.

Noteworthy Problems

I found the process to be pretty smooth by edtech standards. Certainly orchestrating mass registration in a class always has a bit of a herding cats element to it, but this process actually compared favorably with something like signing up for Google Sites or setting up a blog. That said, there were a few issues I’d make more effort to plan around.

Claiming Sites

As I mentioned, you should be very insistent that students claim their sites early on. We did have one issue where a student looking at other student bios ended up claiming someone else’s site inadvertently, which was a bit of a mess to sort out. Before the students start to wander off their newly created site, have them claim it.

Creating the Class Circle

I found it a tad difficult to create the Class Circle while simultaneously assisting students in setting up their bio pages. I think what I would do in retrospect is have them set up bio pages, claim them, surf other bio pages, edit their own pages again — then I’d call a break. I could probably get the circle page made in about five minutes while the students go get a soda. When they came back, we’d continue.

Logouts and Yellow Borders

I’m not sure how this happened, but a couple students logged themselves out and started getting “yellow-border” pages, indicating their changes were not being saved to the server. Additionally, in the flurry of 18 people hitting the AWS micro instance at once it may be that one or two of the edits did not post because of that (note: this is only speculation). In any case, I think I would have started off explaining blue and yellow borders to students, and showing them what to do if they got a yellow (check to make sure you’re logged in, then fork the page to the server to save your offline edits).


The biggest surprise is that no one really had trouble wrapping their head around the tool. It was no harder for students to understand than blogging or social bookmarking. We even did an activity where students forked a page with a  George Siemens video on it, took notes on the video, checked the notes other students had written through using the “twins” links, collaborated with students in their group on a page, then did a cross-tab drag and drop to fork the resulting video summary to their site.  One or two students out of the class didn’t quite make it, but the vast majority of the class did this easily.


(If Warhol did George, it’d have looked like this).

This might all fall apart as we get deeper into the tool — here they are just executing actions without really understanding the underlying interaction model. So I don’t want to celebrate too much yet. But it may be that federated wiki is easier for people who have no extant understanding of feed-based blogging communities or standard wikis since we don’t have to unseat any exisitng ideas of how the web is supposed to work.

Then again, it could just be I got lucky — this was a heavily guided activity, and the question is whether they can do it without the guidance. We’ll find out next week.