Eat steak with chopsticks,
Grab a spoon for those noodles,
Something's not right here.

The Lego Movie (which was awesome, btw), my recent (and forever ongoing?) work with OpenHand, and my current side-work on 3D printer modifications has had me thinking more about modularity and its place in physical design. Like I (poorly) mentioned before, modularity is integral to software design. Everywhere I look nowadays, framework-this and framework-that. Whether you think it's B2B or B2D, the new thing nowadays (imo) seems to be building platforms for others to develop and expand. Mechanical hardware development is often more time-intensive and less forgiving than software development, which is why I've been more and more fascinated by the growing DIY and maker communities, and the market being developed around them.

Back in my undergrad days (when we walked uphill both ways to do anything, yada yada), things Arduino, Raspberry Pi, Makerbeam (and assorted extrusions), or Electric Imp weren't as readily available as they are today. Other writers can and probably have already put this more eloquently, but the maker community has really been growing thanks to an explosion of accessibility, thanks to the internet and companies willing to make these starter hardware kits. The barrier to entry has steadily been lowering, and DIY enthusiasts working out of their garages (or equivalents, for those of us living along the coast) no longer need to specialize in a bunch of fields to really synthesize a multi-disciplinary project.

That said, I would argue that the vast majority of projects built on these entry-level hardware platforms are little more than extensions on the simple examples provided by the developer, and as long as that's the case, there's only so much these platforms can support (pun intended). For a framework or platform-type product to succeed commercially, I think either it must perform its functionality better and more efficiently than any other alternative, or someone else needs to find a new purpose for it that no other alternative can fulfill. The former is why I think it's so difficult to create a successful framework (on purpose). Most products (again, imo) are built for a particular purpose, and they're judged on how well they achieve their goals. How do you properly design a product whose primary purpose is unclear from the start and by design? The problem with being a jack of all trades is that you end up sucking (just a little bit) at everything. Despite this, everything I just wrote seems to fly out the window when you're dealing with the DIY and maker market.

Assigning a consumer value to the build experience is something I think is unique to the maker community. I had a post-doc recently mention that (in the robotics space) if a company is touting the educational and academic applications of their product, then it's likely they've already failed commercially. In those cases, the company more than likely did not get the reception they expected for their product's intended use, and falling back onto the academic space, which can always extract value from just the act trying new things, was the next logical (and possibly only remaining) option. Products made with the maker community in mind seems to embrace this (seemingly backwards) trajectory from the start. Depending on what project you implement with an Arduino or Raspberry Pi, odds are good that the vast majority of its output/input pins and ports remain unused. For me, I might run XMBC or a Octoprint on my Raspberry Pi, but that's it. If I had known this prior, and if the option was available, I'd prefer to purchase a dedicated board for just those particular objectives. The maker community seems to (rightfully) value flexibility, even if they only end up using the product in just one particular way. So, hackers want something designed for hacking, even if they don't end up hacking. Time will tell just where the ceiling is for this sort of market, but my (pessimistic) guess is that it's a lot lower than most would think.

(If you're still reading my rambling stream of consciousness up until this point, God bless you.)

At the juxtaposition of the traditional and DIY markets are kits, designed for assembly (hopefully) but not assembled. Kits are designed for a very particular purpose and yet still provide some leeway for the more hands-on consumers to modify. Usually, I would bet kits came about as a cost-cutting maneuver, especially when it comes to some of the earlier 3D printers. It's nice to not have to worry about assembly and quality assurance costs (as much) when you're just starting. Kit parts are more often than not cheaply made (ie. lasercut plywood or acrylic) with a standard set of fasteners or joints. This actually offers a neat opportunity for makers to fine-tune and upgrade kits, and that's certainly evident with the RepRap movement. Some user modifications even make it to the product's next stock revision. However, as most desktop 3D printer companies have shown recently, it's more profitable to produce a fully assembled (and more expensive) printer to tap into the more traditional commercial market than to keep producing kits. The vast majority of consumers don't want to keep tinkering, they want plug and play. Your average Joe is probably not going to dedicate time to building a 3D printer, much less anything more complicated.

So why write so much about kits if they're still mostly ignored by the masses? One word (name): Ikea. They've managed to disguise 1% of the world's wood and some fasteners as furniture. Perhaps their greatest success (and challenge) is making their products as easy to assemble as they are, but it doesn't change the fact that their business is built on selling kits. Entire communities have even formed around modifying and extending their core products. That last bit may have been accidental, but I think that may be the direction that the consumer market will (or should) head towards next. Like I said before, the average consumer doesn't get much benefit from a product that can do A and B when they primarily want a product that does either A or B well. Even when you think about Legos, the epitome of modularity, they're used to build something very particular in mind. Sure, they can be taken apart and used to make something new, but if it's meant to last longer than playtime, you want a very particular set of parts. Hence all the custom Lego kits and movie tie-ins.

Instead of focusing on whether to make the product that does A or B, or the product that does both, why not figure out how to eliminate the production costs of switching between building the products themselves? How do the economies of scale change when a huge selection of products use the same set of core materials and building blocks? More importantly, can Ikea's model be expanded to more than just augmented, static pieces of wood?

tldr; This ended up being a re-hash of what I rambled here, except now I'm talking about Ikea instead of Google. Oops. Sorry.