Conceptual Puzzles: the Open Source Learning Lab Kit

The Open Source Learning Lab Kit is a conceptual tool for the design of collaborative learning environments. Conceived as a puzzle, spatial concepts are introduced in a playful manner making them accessible to architects, parents, and especially the teachers and the students that will use the space. Through pattern tessellation, specifically the Cairo pattern in this instance, a standard module is introduced to give repeatability to the puzzle. Particular to the Cairo pattern, the pentagon as the standard module or tile develops through 4-fold symmetry to create a hexagon. These hexagons are nested in both horizontal and vertical axis creating a simple field condition. When combined with programmatic intent, this hexagon works particularly well to form a learning cluster with a lab space flanked by two studio spaces accessed through an open collaborative space. The concept of clustering can be understood as a micro-learning community organized around a central common collaborative space which emphasizes that learning is social. Through this basic hexagon unit or cluster, and the introduction of different module types, a manifold of variations and options are possible linking neighborhoods of micro-communities effectively building a network of collaborative spaces.

The idea of a boxed set, or “kit”, takes inspiration from Friedrich Fröebel’s “gifts”. Fröebel introduced these gifts as physical objects to engender a sense of wonder and discovery in the learning process. These are not toys, as such, but introduced in stages as part of a child's learning development. Among many things, these gifts develop spatial thinking most famously an early source of Frank Lloyd Wright’s interest in architecture as well as the source for many other architects and artists as described in Norman Brosterman’s wonderful book Inventing Kindergarten. As finely crafted objects, these gifts introduce a sense of desire and curiosity into the learning process, and furthermore, as physical objects for all to see and manipulate, learning is not individual, but social. Discovery becomes contagious.

Developed with a renewed sense of craft through digital fabrication, the Open Source Learning Lab Kit is intended to provoke curiosity, desire and delight making spatial discoveries through a sense of play. As Johan Huizinga has developed in Homo Ludens, there is meaning, or something “at play” in play. According to Huizinga, play transports individuals from “ordinary life” creating a “magic circle.” This magic circle is an environment of its own - play creates an atmosphere. And yet, there is structure to it, there are rules - in fact Huizinga describes this structure as essential to play. For the Open Source Learning Lab Kit, pattern tessellation provides this underlying structure while the association of programmatic and spatial intent is intended to transport the user from our preconceived institutional ideas of “school”.

Far more than fantasy, this play is intended to expand our sense of the environment and its influence on learning. Inspired by the Reggio Emilia approach to education, after adults and peers, the environment becomes The Third Teacher. This is supported by research in embodied and distributed cognition, and JJ Gibson's concept of affordance: the environment is part of an "ecological niche" that enables and empowers individuals to explore. In this sense, the Open Source Learning Lab Kit has its own set of affordances built into the precision of the puzzle pieces which suggest and provoke an expanded sense of our learning enviornments.

This instructable will first cover how to play in several steps and then get into the making of the puzzle. If you are only interested in the making of the puzzle, you can skip to step 4.

Let's play!

The kit develops from a basic assembly conceived through the concept of a cluster in educational design - combining classroom or studio spaces with a dedicated lab or workshop space, all organized and accessed via an open common flex space. This basic assembly is created with three principle parts: a 3/4” basic unit, a 1/4” top closure piece, and volumetric inserts. There are two basic unit types: a barbell and a hourglass shaped unit. When combined with the closure piece, these basic units create open spaces in different areas. When combined with spatial volumes, different types of spatial use can be explored with colored and transparent volumes. Rather than assigning particular functions or uses to these colors, these are left open for the user to define. These volume inserts are scaled to accommodates about 30 people, such that the basic cluster assembly here would accommodate about 90 students.

While the cluster concepts works in principle, it is only one organizational concept in education design. Furthermore, simply copying the same cluster concept over and over again would only marginally be better then most typical schools! Instead, combine assemblies to create a variety of spaces and configurations including larger open spaces, multiple height spaces as well as different spatial directions. Using different 1/4” closure pieces, multi-height volumes can be created. Spatial direction can be changed by nesting the parts together through rotation. Create open collaborative spaces in the transitions between basic assemblies.

Developing from the combination of assemblies, create further spatial variation and interest through unique units and clear open spaces. Larger enclosed spaces can be created for shops, theaters, etc where acoustical separation is desired. Ramped spatial transitions can be used to create open circulation, open auditorium seating, or open stepped collaborative spaces. Combine multiple 1/4” closure pieces end to end to create clear space open spaces. What will you create?

The Open Source Learning Lab Kit is both a finely crafted puzzle, a physical artifact, as it is a process to develop your own puzzle and by extension, new learning environments. Beginning with pattern tessellation, there is a great variety of mathematical tessellations with even the recent "discovery" of a new tessellation. While mathematical tessellations are a great place to start, tessellations are part of patterns of our daily lived experience, such as this beautiful quilt at Pier 9. You can even create your own tessellations as indicated in the diagram above.

I chose the Cairo tessellation as it is unique, but not alone, in that it develops one shape - the hexagon - from a collection of smaller shapes - the pentagon. Moreover, the proportion of these modules worked fairly well as a room, and most of all, suggested a strong cluster arrangement as indicated previously. However, my primary concern was if enough variation could be developed from the same pentagonal shape. To test this, I created a board game: Cairo Commons. In fact, in researching Cairo tessellation, I came across a board game based on the Cairo tessellation called Cairo Corridor. While much inspired by Cairo Corridor, the goal of Cairo Commons is to organize clusters of enclosed spaces such as classrooms (in black) linked with open flexible collaborative spaces (in white). Adding a rule that each black tile must share an edge a space on the board, provided for an exposed wall to allow natural light into the space (if this were a school). By creating a game, I was able to develop very quickly a variety of shapes triggering new possibilities I had not considered, and even more, by putting the game out for others to play further possibilities were developed (see time lapse). In short, this convinced me that even with a standard tile (module), there was indeed plenty of room for variation.

Developing from the Cairo Commons board game, early 3d prototypes were developed from PLA Makerbot 3d prints as well as my first cuts from 1" thick aluminum on the Omax water jet. The joint between each unit was to be a 1/8" continuous aluminum rod. Variations in thickness was prototyped using the Bridgeport mill. 1" thick HDPE was also cut on the Omax creating precisely fit infill into the voids of the aluminum prototypes. To test this out through a larger kit of parts, multiple copies were 3d printed on the Makerbot. While this helped to visualize the system, it also started looking more like a fixed architectural model. While the 1/8" rods were in fact quite sturdy, they did not allow easy removal. One word changed everything: magnets!

The introduction of magnets as a flexible joint provided several features: by using small 1/8" diameter magnets, connections could be made only at the magnets which kept the pieces on module, the snap connection provided a satisfying feedback to the user when parts came together, and with these ease of connection, rather than milling out openings in 1" thick aluminum, the combination of 3/4" aluminum + 1/4" aluminum top pieces not only eliminated the time consuming milling operation, it simplified the number of unique parts to two basic shapes - the barbell and hourglass - while providing variation through three different top pieces - similar barbell and hourglass shapes + the full module hexagon. This step also developed a subtle but significant shaping of the basic module eliminating sharp corners and streamlining the shapes around the magnets. At this stage the size of the modules was also tested and calibrated with architectural scale - testing units out at 1/16", 1/8" to ultimately settle on a scale of 1:20 as this gave the size and feel to the units in hand. After testing out with user feedback, it was time for production.

While the Makerbot prints are cheap, they are not rapid. Each part takes about 1.5 hours to print. However, on the waterjet, each part could be cut in just less than 15 minutes from 3/4" aluminum. So outside of the cost of machine time (which was not an issue at the Pier), from the perspective of the cost of time, water jet cutting the parts was actually more time efficient than 3d printing, and gave a far more significant feel.

Due to the amount of time needed for this production run, I needed to start final production early on. Cut from 1/4" and 3/4" aluminum, tabs were required such that pieces did not fall into the machine, or the drops did not actually float up and crash the machine. Consequently, each tab had to be cut off and filed down by hand, adding about 10 minutes of post-processing to each part. As precision was a key criteria in the fit and feel of the pieces, the "taper" of the water jet was of particular concern. Through a series of calibration studies, the 3/4" aluminum actually exhibited such little taper that it was insignificant. On the other hand, the 1/4" aluminum had enough taper that the pieces would not lay flat when aligning their sides. Fortunately, the Omax water jet at Pier 9 has an A-Jet. The A-Jet is a 5th axis introduced to counteract for this taper by tilting the cutting head to an angle such that with the taper, the wall of the part came out flat. To my surprise, this was hardly an automated process requiring about a half-day to to dial in the correct angle and tolerances, and in the process I came to know the water jet very well! (I intend to write another instructable on using the A-jet).

With all the tabs filed smooth, the edges of the pieces were not necessarily sharp, but they certainly were not soft to the touch. As I was new to working with aluminum, I spent an entire day polishing the first batch of parts with scotch pads. I also tried sandblasting parts, but as I would be clear anodizing the parts, there was some concern that sandblasting would not be appropriate for this. Fortunately, just about this time the tumbler at the Pier came on line, and Ben wrote a great instructable on using the tumbler. I tumbled batches of parts for between 4-6 hours with coarse ceramic medium.

As aluminum would tarnish over time and use, I chose to clear anodize the basic pieces. There are a few options for anodizing in the bay area, with Metalco in Emeryville as the most local recommendation. However, I was traveling through San Jose once a week so I worked with two anodizer's in the San Jose area: A & E Anodizing and Santa Clara Plating Company. Although I had frequently specified clear anodized as an architect, I had never taken parts that I had personally machined to an anodizer. Consequently, at first there was a lot of mystery in how the parts would come out:

1. Would the process eliminate any surface blemishes?
2. Should I sandblast the parts?
3. Would it brighten up the aluminum pieces?

The answers to each of these questions was in the end: no. Essentially what you see is what you get.

Answer #1: The parts did feel smoother to the touch after anodizing, but surface imperfections, even minor ones, were still apparent which is easy to get when water jet cutting (especially the 3/4" aluminum).

Answer #2: I wanted to sandblast the part as the water jet cut on the thick parts essentially gave a sandblasted finish to the vertical walls of the part. As our sandblaster here uses the same garnet used on the water jet, the sandblasting gave an even finish all around. However, there was some question about anodizing sandblasted parts, and an internet search on sandblasting was not terribly helpful. My recommendation is: give them a sample part! A & E was very helpful and friendly, I dropped off two parts, one sandblasted and one not, went for a long lunch and errands, and about three hours later they had them for me to see. The sandblasted parts had only a minutely more matte finish, and just a little bit darker but most would not even notice, but all the sparkle from the sandblasting was gone. So for these parts, sandblasting made no appreciable difference. Answer #3: While the tumbling gave a nice soft touch to the edges, the water + degreaser that is placed in with the medium gave a slight tarnish to the aluminum, same as if you washed a raw aluminum part in water. To my eye, the anodizing came out looking a little bit milky, if you will, not bright clear anodized. This I believe is due to the tarnishing of the aluminum based on the water in the tumbler. They look great, just not a bright clear anodized. In the future, I would try tumbling without any water to see would happen.

The big question for this piece was about color, covered in the next step.