Update on Patreon releases and ProtonAOSP 13

Alright, time to address the elephant in the room: ProtonAOSP and Android 13.

First of all: I'm sorry for the silence. I get distracted easily and had to shift my focus and priorities urgently last month due to external deadlines and personal plans I’m running very behind on. I was also working a full-time internship, and I'll busy with quite a bit of important non-development work for the next few months as well. So unfortunately, I can't guarantee that it'll be much better in the near future.

Rough timelines for Patreon if all goes according to plan:

• 1-2 days: Repainter v2.0.4 Patreon release (links will be easier to find from now on)
• 1-4 days: PhantomAmp exclusive preview release
• 1-3 weeks: Windows 11 on Pixel 6 + Android 13 KVM blog post and scripts (aimed at developers only; setup is too complex for most users, even with scripts)
• October-November: ProtonAOSP update or discontinuation (more on this below)

Now, to curb the speculation: Yes, sadly it's possible that I'll have to discontinue ProtonAOSP or wait until January. I'll have to decide based on how my schedule looks in November, but I can’t say it’s not on the table. It’s not something I enjoy working on anymore, and based on the data, it may not even be worth the maintenance cost: 26% of patrons downloaded 12.4.1 and 23% downloaded 12.4.0.

Obviously, I completely understand if you canceled or would like to cancel your Patreon pledge. Whether you decide to stay or leave, thank you so much for the support, and don't worry about your decision.

Deeper insights

(TLDR: ProtonAOSP has been stagnant for a variety of reasons, and while it’s not certain, discontinuation is unfortunately more likely than the alternative. Read everything for more details.)

Please DO NOT DISTRIBUTE this memo. It is not an official announcement and is not binding or authoritative. Official announcements will be made once the project’s future is certain.

Let’s face it: ProtonAOSP is more or less a stagnant project. I think the root of the problem is that it tries to cover too many bases and doesn't really nail any of them. Few people will switch to a custom ROM solely for UI/UX or performance, and I've been making less significant changes because they can be very time-consuming for little return. The performance side of things looks more bleak now as AOSP 13 no longer has the data needed for PGO optimizations (now AutoFDO in 13). PGO is a significant optimization that could be behind why people say stock 13 is so fast, so it’s not great to lose.

Some people seek ROMs for privacy and security. ProtonAOSP’s lack of focus on that, combined with the lack of OTAs for bootloader locking (difficult to solve in and of itself due to my connectivity limitations) means it doesn't cover this case very well. I'm not interested in investing more in this area since I wouldn't use/want any of those features myself.

Others seek customization, which is also far from the project’s focus and is not something I’m personally interested in.

Device support also limits the user base significantly. Pixel devices are relatively easy to support, so that's what I've been doing. Most other devices would require CAF support if I actually want any adoption, but I’m not interested in spending time on that. (Security is not the reason for this.)

And to top it off, there's the looming threat of strong Play Integrity checks (hardware-backed attestation) being used more widely, which will bring another major hit to the user base. It doesn't feel great knowingly working a project that will eventually be handicapped with the flip of a switch on Google's side. Yes, not everyone depends on such apps, but too many users do for this issue to be overlooked. Some also depend on the ever-increasing amount of nice, proprietary Pixel features that feel like a waste of time to reverse-engineer and pose a non-zero, if small, risk to add. It feels like I spend more time chasing down breakage and making AOSP usable than making meaningful changes.

There could’ve been some interesting opportunities for expanding the project team earlier, but at this point, I’m not sure if that’s worth the time investment either. If anyone wants to pick the project up: by all means, go for it. Just make sure you clarify that it's separate from the original project and is not an official continuation, and use a clearly distinct name. I won't be able to help much because of time constraints, but an occasional question is fine.

Given that I'll be busier in the coming months and still need to shift priorities (likely leading to updates even more delayed than they already would be), I think it's fair to say that starting ProtonAOSP 13 doesn’t make much sense when it might just get dropped the following month, so it’s too early to commit to that.

I have far more diverse and interesting project ideas, with far larger scopes and target audiences. Repainter is one of them, but that's only the beginning. ProtonAOSP, on the other hand, has been a bit of a dead end and is no longer helping advance my personal goals at all: I don’t enjoy working on it, end up doing far more maintenance than innovation when I do, and relatively few users benefit from the work. The time I’m able to work on it is also limited because it has to be done while tethered to a powerful workstation (AOSP 13 has ~40% more compile targets/steps than 12) as opposed to all my other projects, and even if I don't spend too much time working on it directly, it's still a mental burden to keep in mind.

I've been using stock Android 13 for some time and don't have much to complain about. So yes, I'm aware of why the project hasn’t attracted so many users and I can relate to it myself. But improving it is easier said than done for reasons explained above.

Again, please DO NOT DISTRIBUTE this memo. It is not an official announcement and is not binding or authoritative. Official announcements will be made once the project’s future is certain.