(Before I get started, I know full well I repeat myself. I'm not reading back over my logs, because I'm a bit tired, so just skip anything I've already said, and apologies for the trouble).

I recall a conversation with an old coworker/friend, who explained the best way to avoid demotivation on a project was to not tell people about it until it's done. He made some good points about how talking about a project can make you feel as if it's finished, although he may have wanted to get some work done, I know I talked a lot at that job.

I guess I haven't completely taken that on board, because I'm not done, but I was logging in to search for some info on a GPS module (did you always have to do that on here? The module is a Reyax RYS8830 - if anyone has demo circuits sans the evaluation board, I will pay money to look at them if you have a decent explanation), and I realised that maybe somebody might be interested in a WiP update. I haven't done any of this kind of work really in a long long while after all.

Anyways, I'm going to try to give a general update with this one, because a lot of the information is in a notebooks that are completely the opposite end of the room, and there is a deluge of information - too much for one update really if I spelled it all out.

The phone is actually coming together. The parts list is more or less finished for this version, and much of the 3D modelling is done. Many of the compromises are done being made too - and there are quite a few. I've managed to prototype some of the core functionality (I need to polish the software because it's an unusual audio stack for phone calls apparently - although much more sensible than other devices in my humble and somewhat uninformed opinion - but you can hear audio over phone calls!), and in theory, I could breadboard the whole thing this afternoon if I had the gumption (it's been a long week, I do not have the gumption - also something something permanent connections/wires/e.t.c.).

Off the top of my head, the features I've been able to do (or at least prototype far enough to confidently say they'll work when I've assembled everything) include:

-working call audio, with stereo microphones (useless over GSM calls, which use mono audio), and audio out through either a USB DAC (headphone jack), built in speakers (which will not be built in until the second revision), or earpiece (salvaged from a blackberry priv no less!)

-working spring loaded slide function - with needle bearings for smoothness! This avoids some wobble I was getting, and has the bonus of making it seem just a little more fancy. We're still using the oem spring for now, but for revision 2, I have a plan of action to switch to a diy leaf spring (based on a video I saw disassembling an anbernic rg slide)

-a smattering of side mounted IO, including, but not limited to:
    -a small 1.5ish inch oled screen (for "at a glance" info), with dedicated navigation button!
    -a big lever to switch the whole thing on/off, with a missile style lever guard (in red obviously)

    -a bank of physical disconnect switches for internal components (was it pinephone, vollaphone, or a different one that has this? My switches are bigger, mind)
    -a fully mechanical keyboard with over 4 rows (5 rows), and over 9 columns! (10 columns)
    -magnified keycaps (by magnified, I mean I'm layering on mod podge over printed labels) with glow in the dark lettering (yes, lighting was a bridge too far this version, rest assured I'm on it)
    -a side mounted mouse (by that, I mean the mouse buttons and scroll wheel are where your fingers would be when you hold a phone in one hand, and there's a psp joystick on the other side where your thumb is. Unfortunately this makes it a "right handed phone" I will not rest until I make a left handed one, don't worry.
    -a 6.25" screen (the waveshare dsi one, can't remember if I mentioned it)
    -volume paddle (yes, paddle. Like a rocker, but taller I guess? Seemed intuitive to me)
    -dedicated camera button
    -headphone jack

-two separate MCUs for peripherals (you can still use the mouse when the phone is closed, and that seemed the best way to manage power + I was running out of GPIOs)

-high side current sensing to estimate remaining battery life! (the biggest change for me is how much I've learned about electronic/embedded engineering in the last two years, which is just about enough to know that I should really pay someone to go over my work lest it explodes in my pocket)

-over 17499mah battery life

There will be more here that I can't remember. A lot of this springs to mind because it was the stuff I laboured over the longest, or just caused a bunch of issues.

Ultimately though, as I said, the finished product is in sight, and perhaps predictably, I've come up with a whole bunch of problems.

1) this phone is big and heavy.

I'm sure I mentioned it before, but that's never actually been a major concern for me. I'm perfectly happy lugging oversized tech around, I always have been. One of the aims of this project however has always been to create something that someone else might look at and want for themselves. The recent-ish(?) boom in cyberdecks, may be of some help here, but I recognise that at its core, I cut down the userbase significantly based on heft alone. I do also hamstring myself somewhat, because fashion hasn't quite caught onto my "enlightened" (my partner was feeling very charitable, and I enjoyed the phrase) thinking wrt diy phones, which means that I will basically always have to put this in a bag.

2) this phone has become too expensive for what it is.

I can't specifically remember when I bought my compute module 4 (8gb ram, 16gb emmc, wifi). It was definitely not £172.80. I think even at the time, there may have been a minor price hike, but the state of the SBC fandom (community? sector? industry?) is dire. I have a whole load of thoughts about this that a hackaday log isn't really the place for, and frankly, I haven't had the practice writing to spell out in any way that hasn't already been said, or is worth anyone's time to read. Needless to say, it makes it hard to recommend this as a device with drop in upgrades like I'd been planning to.

3) I am still relying on OEM parts.

It's been two years. Surely I should have figured this out by now?

4) Some of the compromises I've made to get this out the door (it isn't out the door even), kind of fly in the face of the design philosophy.

See number 3 for part of that issue, but further, I haven't done squat towards external facing ports just yet, and you know what? I'm pretty low on space as it is. I will do it, it was the point of making this after all, but it sucks that I'm not doing it for the first version. Speakers too! I bought an amp and some matching speakers and everything! I spent time considering enclosures, and how I would be able to go toe to toe with the likes of a ue boom 3 (I will absolutely not be able to do that, but I'm a dreamer). Again, I suppose incremental updates are fine - that was part of the design, but I just know that trying to fit this all into version 1 is going to prevent this project ever finishing

5) battery life and sleep mode.

I spent so long researching and opining over it - I think that's the right word. Can't remember if I wrote any logs. It's been the biggest millstone around this project's neck since basically its inception, and I feel very disappointed that I won't be delivering something that has anywhere near the right kind of sleep power consumption (maybe 1 watt if I'm lucky).

6) documentation

The 2 year gap speaks for itself, and perhaps it would have cultivated more community/more of a following, and I will aim to be better from here on out, but I appreciate this doesn't leave a lot of information for people wondering why I made certain choices. Choices I have now made so long ago that I can't remember why I would ever have done anything else. I apologise for that, and I apologise to anyone that got excited initially only to assume this is yet another vaporware.

The good news is though, I've already started work on (or at least planned) solutions to every single one of these problems! (the first year was creating the problems, the second year was solving them)

1) phone too big for pocket? Bring the notifications to your wrist

That's right folks, I'm whipping up a smart watch. Not just any smart watch, but the world's first diy hybrid smartwatch with a fully mechanical movement! I've already got a bunch of the features and circuitry working, some of it is stretch goals (light sensor for photography, gps for running routes), but I'm talking piezo buzzer ringtones (using the watch case as a resonator), a 0.42" screen, ble notifications (with a companion app incoming for android and linux - incoming being a bit load bearing here, maybe eventually is a better word, and custom firmware for interfacing with apple devices - they have their own ble notification system), and mayhaps even a game or two! (I promise I'll finish the phone first).

1.1) Don't want a second device? or 2) don't want to spring for a cm4? How about Theseus phone Lite and  5) why not just decrease the battery size? How about Thesues phone sleep!

I will be solving the size and cost issues from two angles. Unfortunately, the CM4 will be relegated to the "prototype" version of this phone. It's a serviceable computer, and I do still recommend it (if you can find it and you have the money), but reports have been coming in from several reliable sources wrt rockchip rk3588 based boards and sleep mode (I believe the folks over at the uconsole forums managed something, the venerable StonedEdge of retro lite cm4 and cm5 fame managed it, and I've got intel from another group I'm in that I'm not sure I'm really allowed to share, but it's not as if I'm spilling any details, so whatevs, I'll take this down if I'm asked to).

Again, these boards have still been affected by the price hikes with ram, so it's very much going to be a "pro" tier phone, but I've been eyeing up a radxa cm5 for my personal device - depending on how well the petg stands up to heat problems. If I can make that a drop in change, I'll be sure to set up and share an image of my distro with all the changes I needed to make + driver software for MCUs of course, hopefully including sleep mode. The screen time may end up being lower with the thinner battery I'm planning, but the sleep time will increase, and really, this is not going to be a phone you can scroll on for any great length of time, so I see that as a win.

As for the lite version, same battery I mentioned before (the battery in question being a 10000mah lipo, which is still chunky, but significantly less than a stack of 5 18650s), similar sleep/screen time, but I'm aiming to use either a radxa cubie 7z, cm0iq (which should provide drop in support for the dsi screen), or some orange pi. I've been doing some light experimentation with my pi zero 2 w using maemo leste, and I want to give Matchbox Window manager a proper crack again at some point, because as we all tighten our belts, I'm hoping to make this as comfortable and usable as possible. I do hear a lot of complaints about the 512mb of ram on board, and while I'm determined, I know determination can only take one so far, so I'll hopefully be experimenting with some of the pi zero clones with more ram, but I'm determined to keep the costs essentially as they would have been back when I started this (i.e. I will be avoiding using boards that cost the same as the cm4 did when I bought it that have less power. If you want a lite version, you ought to get the savings).

Either way, I don't want to leave anyone out in the cold, and I appreciate a device the thickness of an original gameboy may not be ideal for anyone not wearing JNCOs, but hopefully one of these (or the watch, or some half way solution) should tackle that problem for just about everyone. I am open to solutions however, as I really do want this to be a viable linux phone.

3) OEM parts

I don't have quite so much to say here, but I'm down to 2 remaining oem parts. A slide plate, and a spring. I've looked into how I'd replicate/produce my own slide plate, and I've got a few leads (arbor press + 3D printed negatives + sheet of aluminium + dremel). I really do not want to force people to rip apart perfectly usable blackberry privs to build this. As for the spring, I'm sure I mentioned it in a previous log vis a vis a diy leaf spring. I've found a viable looking design in a disassembly video for the anbernic rg slide, and I've got most of the measurements down pat. It does also seem like a watch mainspring should be fit for purpose too, so for my second revision, these are my top priorities.

4) I don't really have anything to say about the compromises. They're there so I can finish this, and aside from removing OEM parts, the first changes will be removing or improving what ever turns out to be the most annoying aspects of it.

6) Documentation

I will be making full detailed instructions, a BoM with alternatives, and all files necessary per revision free through this site in relatively sensible places. This log is going to remain the "storytelling" side of things, and I will be coming back to that in future with more to come from that side too. I hope they've been entertaining, as there will be more logs, and I have a load of old notes that I was similarly dramatic/self indulgent about that have stayed on paper (maybe they'll see the light of day if there's demand for it)

In summary:

-the phone is still coming, it's nearly there (minus a few planned features which will come in the second iteration), and I've got plans for variations too. I will be a diy phone builder for a long while to come assuming a plane doesn't land on my house or something

-my next separate project will be a watch. This should be a lot quicker, because it's a lot smaller, there's a lot less parts, a good chunk of it is already working, and to my horror, commenters (when I've been out and about asking for input) have called it a "sound design", something they'd be "quite interested in actually" and perhaps most upsetting of all "sensible". I'm not sure if the last 2 years have taught me enough to convince people I actually can pull off my crazy plans, or if I've just lowered my aims to the point that I can no longer call myself a crazy inventor. I know in my heart though that it's a bit of both unfortunately.

-Expect some sprucing up in the next few months on this project page. Real photos, actual demos, stuff that's of interest to real people. More plugging of my sources too! This project couldn't have happened with just me, there are individuals out there who have been instrumental in getting things to work, and I'm not going to take credit for it, so I hope they're all braced to be associated with this overly long soupy nonsense diatribe. I'm not going to uncredit them.

-the end point will be when I've finished a "commercialised" version of the watch (i.e. a version that doesn't need you to go trawling ebay for out of production watch movements), a lite version of the phone that sees a significant price cut (I wanted £150 for lite and £300 for regular. That may be a stretch, but I will do my utmost), and a final version of the phone with no remaining OEM parts, a couple of compromises removed, and a sensible compromise betwixt battery life and thickness. After that, I'm going to take a break and work on my writing skills for some time. Then I'm going back to video games and I'm curious to get into drones and robotics, but those three are even longer term.

Very very very short version:
1) finish phone prototype and provide instructions/details
2) finish watch prototype
3) make cheaper phone
4) make improved phone
5) "commercialise watch"
6) type it all up
7) video games, robotics and drones
8) I'll probably be at least 70, so maybe retirement and joining a rotary club

the end (or is it? Not really, just the end of this chapter).