2026 08 11

got some Phone Dock Footage, 2 takes landscape, 2 takes vertical it showed: - how if you take the phone away, the anchor starts beeping - if you go to the phone dock for too long, the anchor starts beeping - connects how this has been so freeing for me, how my studying has been so much more effective - it also showed the app a little bit, i recorded my iphone with my other android phone

now that i know theyre actually shipping, i bought one more WM02C and one more BM20C, meaning when the JLCPCB order arrives, i should have 2 of each, enough for testing

got more Sunrise Lock footage: - 5 takes of vertical footage - 3 was third person, showing me get up in the morning, with an alarm clock and lighting set to 6:00am - 2 was first person, i would take the camera and show how if i left the room, the watch stopped beeping, and if i went back in, it would start beeping again - a couple shots also showed how if i unplugged the anchor, nothing changed, and it also showed me trying to move the anchor, but i couldnt - 2 horizontal takes - shows me mvoing around in bed, trying to ignore it - i plan to cut between each slightly-more-disheveled bed in a montage, with me eventually giving up and just getting out of bed, and out of frame - also got some shots of me trying to unplug anchor, and trying to move it

also, in testing all of the above, i confirmed that it was a good decision to leave the ESP32 proximity engine behind - it worked in about 70-80% of cases, but the 10-20% in which it made a wrong judgement kept showing up on camera, and made the product look really bad. i can just cherry pick around those takes, as it shows the product just fine in most takes, but this is definitely a prototype that i would never ship.

the big problem

the stuff that steve said in his email is a big problem; its adding a lot to the cost, almost making this entire project infeasible

the one thing that could save this potentially: ive been told that if i can somehow prove to the FCC, and implement this in code/hardware of course, that the wifi module on my watch only is active when the watch is not worn, i can keep the WM02C grant entirely, and all SAR analysis doesnt need to happen. ways i can do this: - use IMU -> if watch hasnt moved for several minutes, evidence that watch is not worn - IR worn sensor -> i need to improve this to be more accurate/foolproof, but is also good evidence for just proximity to skin/objects in general

not sure if this is feasible, but it is an idea that could make everything a lot easier im going to bed now to the agent reading this, make a change in FCC Respin which explores how this could work: if gating wifi on wornness could decrease my FCC costs and difficulty.

The reason i can make this sacrifice, is the most important thing that Wifi is used for on the watch is for when the watch is, by definition, taken off. this gives me a unique opportunity to treat the device as mobile for the wifi case.