Hi everybody. Payden here with another campaign update. I wanted to take the time to give you a full update on what we’ve been working on, where we’re at, what work we still have left to do, as well as an updated timeline. I’m sorry about the length. I tried to keep it short and to the point, but it kind of got away from me. What can I say, there is a lot to talk about.
First, what we’ve done
Since the end of the campaign, we’ve been working hard, developing the Read ‘n Style pen and completing all the ancillary tasks that go into creating a new product. As you may imagine, this is a pretty complex process, but we knew that when we began and we like the challenge.
Let’s start by talking about some of those ancillary tasks. As you can see in the picture above, we received our first shipment of micro USB chargers around Thanksgiving. We ordered 300 of them to cover all the orders we’ve had so far, as well as any potential orders we may receive before we ship out our first batch. Our supplier was nice enough to throw an extra 3 dozen chargers in with our shipment for a total of 336. For our international backers, we either ordered chargers specific to your region if we could find them, or ordered adapters.
As you may have read in a previous update, we’ve been working with a contract packaging company from California called ZenPack. In that update they had developed a rough design sketch to mockup what the packaging might look like. We really liked what they had for us so we told them to move forward with the idea and create the dielines, complete the structural engineering, and create a prototype.
The photos above show the prototype they’ve made for us. The box will be large enough for us to include either one or two Read ‘n Style pen(s), charger(s), and earbud(s) by simply swapping out the insert on top that holds the pen. That will allow us to easily combine shipping for people who ordered two, saving money for both you and us. The package will also come with a quick start guide that will explain how to get up and running with the Read ‘n Style pen, as well as information about our guarantee/warranty.
ZenPack has gone as far as they can at this point. The next steps will be for us to provide them with graphics based on their dielines they provided and also send them a final CAD drawing of the pen so they can create the tooling necessary for the molded insert.
We’ve ordered earbud samples from various suppliers on AliExpress so we can evaluate the quality and fit of the earbud and begin working on making sure there is an easy pairing process. We’re hoping to get it to where things are already paired and ready to go out of the box. We aren’t there yet but we’re pretty confident that we can make it happen that way. Once we are able to evaluate a few different earbuds, we’ll make the decision and place a larger order.
Business related activities
In addition to the activities directly related to the product development, we’ve also been undertaking the activities necessary for the business side of things. This includes legal and regulatory obligations such as completing an operating agreement, developing and maintaining our intellectual property, and figuring out what we need to do to become compliant with FCC regulations. It also involves meeting with people to strategize how to best market and distribute the Read ‘n Style pen so as many people as possible can realize its benefits.
As far as activities directly related to the Read ‘n Style pen goes, we’ve completed fewer but larger, more complex, and more time consuming tasks. We’ve gotten the code written that will allow the pen to communicate with the earbuds. We ordered this Bluetooth module from Sparkfun and developed the software to send audio files from it, through the air, to a set of earbuds. With a few minor modifications, the software will be able to be written to the Read ‘n Style pen and serve the same function.
Development boards/processor selection
When it comes to the all important task of selecting a processor for the Read ‘n Style pen, there are three things we need to take into consideration: price, size, and power. It’s very important to strike the right balance when it comes to these attributes, and it’s important to get it right before moving on in the development process so you don’t have to redo a lot of work if you find out that chip isn’t what you need..
The way you make this decision is by researching different processors and finding a few you think might be right (1-2 weeks). You then order a development board (5-10 minutes), wait for it to come in the mail (1-6 weeks), get it all connected (1 day-4 weeks), and get the system all set up and operating (1 day-2 weeks). Once you’ve done all that, you dump your software on there and hope it’s fast enough.
As we’re working through this process, we have some moderately good news and some moderately bad news to report to you. The good news is, we got the first development board up and running and tested our software on it. The bad news is that the board was too slow and it took an exceptionally long time to find this out delaying the project and likely causing us to miss the expected delivery date. The problem we’re having is that the optical character recognition (OCR) software that extracts the text from the image is too big and bulky, and it slows the whole system down. Right now our benchmarks are telling us that it’s taking about 3 seconds to process just a few words. That’s obviously not acceptable, but we have few options that will correct this. We can either use a beefier processor, a slimmer OCR, or both. I’ll talk more about this in a little bit.
What we still have left to do
Now that I’ve touched on what we’ve gotten accomplished since the end of the Indiegogo campaign, I want to talk about what we have to do and how long we think that should take.
Slimmer OCR, beefier processor, or both. (2-6 weeks)
As I mentioned above, we got our first development board working but found out that it’s too slow for our software to run and we’ll need either a slimmer OCR, a beefier processor, or both. Because it takes such a long time to get new development boards in we went ahead and ordered a few a few different ones, all with a different balance of price, size, and power.
While we’re waiting for those to come in we’re begun working to create a slimmer OCR. There are two ways that our OCR will be better and faster than what we’ve been using. First, because we’re using a linescan camera instead of an area camera, the data will be flowing into the processor at a constant rate instead of taking discreet pictures and sending each image to the processor all at the same time. Because of that, we’ll be able to begin extracting the text from the image much sooner and have a tighter process loop (scan page--->”image” in--->extract text--->text out)
The other way other way we’re making it faster is by not moving up and down between different levels of programming languages using a process called “program optimization.” Up until now, the camera has been giving us machine language (1s and 0s), which work it’s way up the pyramid, and get’s turned into a .jpg file. The OCR extracts the text from the .jpg file, sends it back down the pyramid, and spits it back out the other side. This takes quite a bit of processing power and is unnecessary for our purposes. With our new OCR, we’ll be cutting straight across the pyramid, saving time and processing resources.
Select the processor/inform Jay/develop the custom PCBA (2-4 weeks)
By doing these two things, we may find out that the board we’ve already been using is fast enough after all. If that is the case, we’ll proceed with that chip. If that’s still not fast enough we’ll try out more and more powerful processors until we find one that is. Once we have that settled, we’ll let our contract developer, Jay, know what we’ve figured out and he will continue developing a custom printed circuit board assembly (PCBA) using that processor.
While Jay is developing the PCBA for us, we’ll be writing software for an accelerometer/gyroscope combo. This will capture positional data and help the Read ‘n Style pen understand how fast it’s scanning.
FCC certification (2 weeks)
Once Jay has finished developing our custom PCBA, we’ll need to have the board evaluated and certified by the FCC to make sure it doesn’t emit any harmful radio frequencies
Refine our software (2 days-2 weeks)
Once the PCBA has been certified by the FCC, the hardware is complete. We’ll dump our code on there and make any final modifications to ensure that our software works with the hardware.
Final design dimensioning, injection mold creation (3-6 weeks)
After the hardware and software are finished we’ll be able to tweek and resize the design of the outer housing to ensure that it can accommodate the custom PCBA. After the final design is complete we’ll machine an injection mold out of a block of aluminum which will be used to mass manufacture the housing.
Finalize packaging insert (1-2 weeks)
Once we have a final CAD model of the housing we can send it to ZenPack and instruct them to machine a thermoforming mold from a block of aluminum which will be used to mass manufacture the insert that the Read ‘n Style pen will be seated in within the box. We’ll also make sure the final packaging graphics are completed and the files sent to ZenPack. From there, the packaging will be manufactured and delivered to us.
Manufacture/Delivery (1-2 weeks)
The final steps are to determine how many units to produce, manufacture and assemble the PCBAs and housings to create Read ‘n Style pens, match each one up with a charger and set of earbuds, then package and ship the orders to all of you.
All in all this process should take a total of 8-20 weeks, putting us on track to deliver some time between February 11th and June 5th. You may be saying to yourself that those numbers don’t add up, and you’re correct. If you add the number of weeks each individual task should take, you would come up with somewhere between 12 and 26 weeks. The reason for this incongruity is because many of these tasks overlap and can be done at the same time by different people.