Hello everybody. I wanted to take some time to update everyone on our progress of late as a lot of you have been asking about us. We’ve been hard at work since the last two updates, and even though we haven’t said much lately, we’ve accomplished one big thing and several smaller things in that time
The big accomplishment lately has been getting the hardware working. Depending on how you look at it, we’ve been working to solve this problem for 2 months (first received prototypes of our custom hardware), 5 months (first contact with our current hardware design company), or 21 months (first approached a hardware developer about our project), so this is big news. It’s also the last major technological hurdle before shipping out all of your orders.
These last two months have been especially tricky as far as product development goes. When we first got the prototype boards in we (naively) thought that we’d be able to plug it in to power, flash our code onto it, and we’d be off to the races. Little did we know at the time, that is almost never the case.
To frame the problem a little bit, one of the reasons that hardware is hard is because there is little to no feedback in the very early stages to let you know what you’re doing right or wrong. To add to that, if you’re not getting the expected output you don’t know if there’s something wrong with your software or if it’s actually a hardware issue. Then, even if you know it’s a problem with the hardware, what’s causing it? Was the board poorly designed and never would have worked in the first place? Was the board designed correctly but something went wrong in the manufacturing process? Or did it make it out of the factory in perfect condition only to have a static discharge cause a short? It’s impossible to tell when your electronics just sit there like a brick.
So the first thing to do it put some power to it and see if an LED comes on. The first day we got the prototypes Jamee and I spent the day running all over Kansas City to find the right power supply unit and connectors so Shane and Zack could work on it when they came in later that evening. When they got there, they started getting everything put together, hooking it up to power, and a solid red light came on indicating it was getting electricity. Great news. Now let’s try plugging it into the computer to see if we can get something useful. No luck. We’re just getting the letter “C” a bunch of times and the light stayed solid. It took a few weeks of research and testing to figure out that we probably weren’t getting enough power to the board so it wouldn’t operate properly. Luckily Zack is still a student and knew where we could find a more suitable power supply unit to see if that would help. It did. So we order ourselves one of those units so we could keep working.
Now we figured out how to get the proper power to it, but we still weren’t able to load the operating system onto the board so we could run it as a computer. We were still getting the Cs but the red light that had been solid was now blinking in recurring pattern, and now instead of one light, there were 3 red lights. That’s all the information we had to go off of. The next 6 weeks consisted of trying to figure out what was going on and at least get more feedback to work with. We sought help online by asking questions in places such as the Texas Instruments website, the Beaglebone forum, Reddit, and Linux forums to name a few. We reached out to our mentors with a hardware experience, met with Zack’s professors, talked it out with the development company that designed it, reached out to 3rd party companies that may be able to help, and took it home over Thanksgiving so our tech oriented family members could help us with. While all this was going on, we kept trying different system images thinking that the problem may be that. Finally, somewhere along the way, we decided to try some different flasher images to try to get the operating system to load. That was a big breakthrough because finally things were starting to happen. Code was being loaded and we were now getting tons of great feedback, all of which was being logged in the terminal so we could read through all the data pinpoint any problems.
At this point in time, the system would try to flash into memory but it would quit at around 60% and restart. To further complicate things, it would restart at a different point each time. Sometimes it would make it to >70% before restarting, but other times it would only reach the mid-thirties. There was no consistency to it whatsoever. That led us to believe that there was a bad connection between to computer board and the fixture we were using to flash the system image.
It turns out that the boards were being clamped to the fixture in a way that was causing it to bow, giving us solid contact in the middle but really spotty contact near the sides, causing the connection to break momentarily and restarting the process. As soon as we figured that out we took the clamps off, held the board in place by hand, started flashing it again and BAM, half an hour later it ran through the whole process, flashed the memory chip, and welcomed us to Linux.
So yeah, that’s done now. And like I mentioned above, that should be the last major technological hurdle before we’re able to ship out orders, and it’s finished. We still need to bring the other board up before we move on to the next step, but that one should be an order of magnitude simpler, and within the last week we’ve already gotten about 90% of the way there.
In other news, we received a grant from the KC Digital Sandbox in September. This program was put in place to help startup companies by paying for one vendor contract of up to $20,000. After receiving this grant, we elected to have them help us out by paying for our patent. Without getting into too many of the details of it, we’ve been working with our intellectual property attorney over the last few months to draft a non-provisional patent application to help protect our invention and make our company more enticing to potential investors who will help us grow and get the Read ‘n Style pen into more hands. We received the final draft on Friday and should be filing all the paperwork this upcoming week.
One sad but awesome bit of news is that Becky, the industrial designer that we’ve been working with from the start, received a really great job offer in San Francisco and has decided to accept. Luckily for us, before leaving Becky put us in touch with a friend of hers named Jack who she met in industrial design school. Jack is very talented and quite accomplished already having designed a wide variety of new products that range from high end custom flatpack furniture to soft goods like hiking boots and backpacks, all the way to ergonomic virtual reality controllers just to name a few. He’s also skilled in graphic design as well which will come in handy down the road.
Jack has some mighty big shoes to fill but we're confident that he’s up to the challenge. His job now is to take over where Becky left off. He’s working to come up with the best configuration of the components to minimize the bulk and evenly distribute the weights so it’s comfortable to hold. As he’s doing that he’s modifying the enclosure design that hold all the parts together. The current thinking is to include a few ribs to give the body added strength and durability, place a few bosses in there to hold the parts in place, squeeze in a few pieces of cut foam to make sure everything fits snugly and doesn’t wiggle (and to give it some extra padding in case it’s dropped), and then hold the two molded halves together with a tongue and groove joint around the seams.
Currently the design looks a little something like this. It’s already pretty small but we think we can further slim it down by about 30% in width. In the final version the exterior will be more stylized with different lines and textures, but this will likely be the rough shape of it.
Despite making some substantial progress, there is obviously still more work to be done before we can get it in your hands. If you’re anything like us, you’re probably more than a little frustrated by this. We get it. We are too. If we had it all to do over again, we probably would have waited an extra 6-9 months before launching the Indiegogo campaign because we really weren’t as far along as we thought we were at the time. Hindsight is 20/20 though and all we can do right now is continue to work hard so we can get the product to you as quickly as we can. Right now, we’re probably looking in the ballpark of early summer for a ship out time. I honestly think it could be a bit sooner than that. As you can tell though, I don’t exactly have the best track record for accurately predicting these sort of things, but early summer does seem to be pretty reasonable.
Thank you all for reading and I hope you had a wonderful holiday season. Happy New Year. :)