The post Our First Delivery! appeared first on Model B.

It may not look like it from the burn down chart, but we are officially in the home stretch.  The last remaining task of significant size is building the tone rings and we are on track to make significant progress on that front this week.  Plus the sensors are all done and the controllers should be done soon.

Backlog – See SyndCon Task List tab for current sprint plan and Backlog tab for full backlog.

Reports

J.

Did

• Prototyped threaded inserts for tone rings
• Tested and revised controller housings with 3D Hubs printer
• Built 5 new sensor modules

Will Do

• Take delivery of controller housings
• Assemble controller housings
• Finish tone ring design work and order plastic components from 3D Hubs printer
• Draft panel presentation for SyndCon
• Ship components to Bryan

Blocks?

• none

Bryan

Did

• Recruited external resources to do controller assembly
• Recruited external resources for the alpha tester t-shirts

Will Do

• Send updated terms to alpha testers
• Continue working with external resources on controller assembly
• Continue working with external resources on t-shirts
• Recount on hand cables and order any missing ones

Blocks?

• none

The post PY18 Sprint 4 Stand Up 6 appeared first on Model B.

While 2017 was a huge success, I was completely surprised by the specifics of how we accomplished our goals.  As a quick recap, here are our goals for 2017.

1. Finalize the hardware design needed to address the key feedback from demos at SyndCon 2016.
2. Run a private beta test program for Holoseat to vet the design before releasing version 1.0.

So, what happened?  We started the year with some less than great technical and business assumptions.  On the technical front, we were convinced we needed to upgrade out of the Adafruit Feather to a full on system-on-a-chip controller board.  And on the business side, we allowed ourselves to be caught up in the siren’s song of crowd source campaigns too soon.  Sadly, these assumptions took us down rabbit holes most of the year.  Fortunately we caught ourselves in time for a great SyndCon 2017.

Finalizing the Hardware Design

We started 2017 chasing down two seemingly unrelated design issues:  profile management and Holoseat’s sensitivity to pedaling activity.  Profile management covers things like trigger cadence (how fast an individual has to pedal to trigger walking).  And Holoseat’s sensitivity to pedaling actions impacts how long it takes Holoseat to recognize changes in pedaling.  Bryan and I focused the first half of 2017 on addressing profile management.  This effort took the form of redesigning Holoseat around a brand new family of development boards:  the CHIP/CHIP Pro.   Moving to the CHIP boards involved redesigning basically everything about Holoseat and led to an enormous amount of scope creep.  I could write pages retrospecting how exactly we got to that point, but that’s another blog post.

We were still testing the fundamental technologies we needed for a CHIP based Holoseat with only 8 weeks to go until SyndCon.  To say we were off plan is an understatement.   Things turned around when Bryan took charge of designing a custom tone ring to use with a new sensor.  Within four weeks we had a working prototype for these sensor upgrades, versus the four months we spent working on the move to CHIP.  More importantly, the new design’s improved sensitivity yielded unexpected benefits.  Namely, Holoseat was now responsive enough to calculate trigger cadence in real time.

We quickly realized this “adaptive cadence” capability meant we could drop profile management and go back (again) to our original Feather based design.  After conducting additional testing we also concluded we had arrived at our minimum viable product (SyndCon prototype shown above).  And what became of all of the effort we put into researching the CHIP boards? A large chunk of that work went into mastering Python in IoT applications. It didn’t take us long to realize the bulk of that work could be re-purposed into the development of a brand new, and feature complete, desktop configuration app. Further, thanks to Python’s portability, the same app will run on Windows, Mac, and Linux. And should we need to revisit a CHIP based Holoseat, it will work there, too.

Setting Up Our Testing Program

Great Read for OSHW Start Ups

One of the things we have to keep in mind is Holoseat is more than just a technical project. It is also a business and an investment of our time and money. This means we divide our effort between technical and business matters. For example, 2017 included research into preparing for and running crowd source campaigns. One of our conclusions from this research was we were not ready to run a crowd source campaign in 2017. We simply did not have a large enough online following to form the foundation of a successful campaign.

What we did have was a supportive core of fans asking us when they could get their own Holoseats. The value of this core group of supporters became apparent to us after reading Building a Hardware Business.  The authors discuss different growth strategies including building on a small core of passionate users.  This strategy resonated with us and we decided it was time to organize our core fans into a testing community.

So, we put together our Alpha Test Program and offered 5 slots at SyndCon. We came home with 8 deposits for Alpha units. One fan even tried to give us his money during our annual Holoseat presentation! #winning  Having 8 high quality testers is going to be significantly better for us in the long run.  Now the race is on to complete v1 Alpha so our testers can get their hands, er feet, on Holoseat.

Stay tuned for Part 3 where we lay out the 2018 plan and cover how we are doing so far.

The post There and Back Again: An OSHW Project’s Tale – Part 2 appeared first on Model B.

So, what has been going on over the last nine months? I am very happy to announce the project has a second developer, Bryan. Bryan and I play Star Wars: The Old Republic together several times per week, and his interest in the Holoseat is a large part of the reason why I got the project organized on Open Design Engine. Some months ago, Bryan replicated the V0.2 prototype and has been using it routinely to play a couple of games. Between the two of us, we have logged countless hours on Holoseat. The results of all this “testing” is we are very confident in its core features and have identified the principle concern in the current design (namely the sensitivity of the reed switch – which we will be working on in the next version).

Bryan is helping with updating the documentation, and most recently he proposed a very promising design change. This design change is the inclusion of a companion desktop application to control the various settings of the Holoseat, eliminating the need for the slide potentiometer to control the “walking rate” and giving users access to other configuration details including the walk command character. Having a desktop tool also enables new features such as reporting statistics (just how many steps have you taken this week?) and automated firmware upgrades.

Bryan has also been integral in securing our first public demonstration of the Holoseat at The Syndicate’s annual convention this August. We are planning to set up the Holoseat in LAN hall and present the project to the whole guild at one of the plenary sessions. In preparation for this demo we are focusing on wrapping up the features we have and working out a couple of bugs, and getting our demo materials ready.

After the convention we will return to development with the hope of having an improved prototype done by the end of the year. Check out the Roadmap link at the top for all the details and watch this space for more news as we get ready for our demo and push ahead with v0.3.

The post A Victim of Our Own Success appeared first on Model B.