HAXLR8R // Blog // Interview with Dave Merrill, Co-Founder of Sifteo

Interview with Dave Merrill, Co-Founder of Sifteo

by Zach.

It's been a year since the last HAXLR8R. What have you been up to since then? Feel free to brag. :)

Dave Merrill Haxlr8rAt Sifteo we have been laser-focused for the past year on building a second-generation Sifteo Cubes platform and a suite of super fun game titles for launch and beyond. The new cubes and games are now out, and appearing in retailers like Amazon, Barnes & Noble, Toys-R-Us and Marbles the Brain Store! We made a new video to inspire developers, and another to show everyone what the new system can do. Indie developers are also working on games; we started a publishing program with some great games in the works, including games by Richard Garfield (designer of the card-game sensation Magic: The Gathering), a team led by the developer of the BIT.TRIP series, and the designers of Johann Sebastian Joust.

Many have said that the hardware landscape is looking more like software with lower costs to entry, better prototyping tools, and faster turnaround times. Do you think this is accurate? What are the critical/important differences in your mind?

Yes, it's definitely getting faster and cheaper to make hardware - however I think the key difference is still the turnaround time on an idea. The first generation Sifteo cubes took us 1.5 years go get to market. The second gen took a year. We're getting faster. But compared to software or web, it's still a much longer road to make a good, solid product, since the hardware really has to be RIGHT the first time when it ships out to the customer.

Can you tell me about a difficult challenge you overcame dealing with hardware (design, production, sales, distribution, etc?)

The architecture and implementation of the new Base's operating system is a modern marvel. The Base is the "brain" of a set of Sifteo Cubes -- it's an embedded system that runs apps on a very low-power processor in an extremely high-performance and secure manner, plays sound and controls the cubes wirelessly. An entire team at a place like Google or VMware would typically be required to build a lightweight virtual machine like the one that runs the Sifteo experience, but our tiny, talented crew pulled it off in record time!

Distribution is also really difficult -- but we managed to get into the new "Toy Tech" section at Barnes and Noble for this year's holiday season!

What has your experience with China been like? Do you have advice for new companies looking to manufacture using Chinese vendors?

Manufacturing in China has a set of unique and difficult challenges related to distance, time-zones, language, and business practices. It's the best option when you're truly ready to scale your production up, but if you go there and get set up before it's necessary, you may be burning cycles and money that could be best spent elsewhere. Make sure you have a local sherpa to guide you in China - hire an expert, or work with a company that helps startups scale production in China, like Dragon Innovation, PCH, or Zao Technology.

What do you think the most promising development in the world of hardware has been in the last year?

The most promising development has been the recent appearance of an honest-to-goodness COMMUNITY of hardware entrepreneurs who have largely emerged from the DIY movement (watch for my article about this in the upcoming January Make magazine). When Jeevan and I started Sifteo we were cold-calling people and working our connections to get introductions to sit down for coffee with the small number of people who had done this kind of thing before. Now there are meetups and accelerators, and people are sharing information online in ways that the software community has been doing for a long time. We are still behind the curve compared to the software community in information-sharing, but we're catching up quickly!

What is your opinion of open hardware and does it have a role in a for-profit company?

I think open hardware is great and I'm glad it's a growing movement, but I see it as mostly a distraction for a for-profit company -- unless opening up its designs raises awareness for the company/product among potential customers. But most consumers don't care. Upshot: if your product is for makers, consider opening the HW. If not, you probably shouldn't bother.

Is Kickstarter still a viable launch platform with the recent rule changes?

Crowd funding has been really interesting to watch lately -- though it's still TBD which of the kickstarted and indie-gogo'd products turn into long-lasting companies. If I were starting Sifteo right now I'd probably combine crowd funding to pay for the manufacturing with some venture investment to support ongoing R&D.

Regarding the recent rule changes: I think the requirement to include a "risks" section and avoid showing CAD models is fine -- it's an attempt to enforce some honesty and requires more upfront work (which is a good filter) in order to show something that looks polished. But with the prohibition on selling multiples, and banning certain categories of products, Kickstarter is risking driving some of the highest-potential campaigns to seek other options. Kickstarter still has the best name recognition among the various crowd funding platforms so it's perhaps most useful as a marketing tool -- but you'll start to see more projects that really just want to get pre-orders move to Indiegogo (e.g. Misfit Wearables' Shine) or roll their own campaign (e.g. Lockitron).

Bonus question - soapbox. Is there something that bugs you about the hardware community? Elephant in the room? Buzzwords that people throw around?

There is a lot of talk about "lean hardware" and "the minimum-viable-product" in hardware - this concept is borrowed from the software startup community. I totally believe in de-risking the product-market-fit and avoiding premature scaling, but interpreting and applying the MVP too literally may be a red herring. Put another way, a MVP for a hardware product is usually a lot less minimum than for a software product. I believe there is a risk that the idea of MVP lulls founders into a laziness about when their product is ready to ship, and they ship too early. For a hardware product to succeed, it has to be correct on all axes: features, price, ease of use, availability of supporting services, etc. Shipping a hardware product before it is truly ready to meet the world - may doom it to failure in a way that a software product could iterate its way out of.