A bit of a tangent, but I'm really surprised one of the android vendors hasn't focused on the niche of easy to repair phones. Maybe even making them four or five times the thickness of the latest i-sung devices. extra hot swappable batteries and the like.
I know the ostensible reasons waterproofing, planned obsolescence, looking cool, and being light weight. Though I suspect the real reason is they're just chasing the big players and are afraid to be different. It just doesn't make sense to me from a business perspective, if you know you can't compete in the general market, why not carve out a smaller market and serve that one well. An example of this is the Jitterbug phone.
Fairphone[0] is a European manufacturer which does this (modular replaceable components, 10/10 on iFixit and usable by non-techies).
In my experience it's the annual release cycle of Android coupled with the lack of OS updates by manufacturers which contributes to e-waste, closely followed by the lack of battery replaceability. My first smartphone (OnePlus One - 2014) still has the specs to be a perfectly usable phone, at least until 5G is widespread and 4G networks are decommissioned.
Can anyone enlighten me about 5G? To me it seems to be absolutely nothing more than a marketing tool meant to sell a new generation of phones and phone plans. The fact that it is pushed so fervently just sets of red flags galore for me.
4G is plenty fast and works well for me. I see nothing gained for my phone by going from 100Mbps to 1Gbps(?). Nothing. But its pushed like the second coming of Christ.
I'm excited. Not for a specific use-case, but because it removes constraints and that lets developers push the boundaries of what's possible.
As a European, I predict it'll be a catalyst for American companies to rethink data caps and data pricing, as you can blow through a data cap 10x as fast. That'll be massive and measurable progress.
Paraphrasing Liebig's law of the minimum[0]: progress is hindered by the most scarce resource. I'm sure that bandwidth will have been that resource for some ideas. These ideas will have been 'before their time' a few years ago, and are now viable.
Take spellcheckers[1], and electron-based apps: they've moved very quickly from "impossible" to "an everyday occurrence", I hope 5G enable this for another class of problem, and I hope it's unpredictable.
I don't have a 5G phone, I don't plan to get one any time soon, I'll have one in 10 years time, and I'm excited to see what comes of it.
Generally there is said to be 3 parts to 5G.
The first is eMBB: Enhanced Mobile Broadband. In other words faster mobile internet. This is where most operators start.
The second is URLLC: Ultra-Reliable Low Latency Communications. This is mainly aimed at using 5G for things like self-driving cars. But also things like long distance remote control. This is where people see potential for innovation without being clear what the exact innovation will be.
The third is mMTC: Massive Machine Type Communications. This is meant for IOT but also for factory control. The IOT thing is mostly allowing extra low battery useage, low speed, cheap connnectivity. The factory control thing is about getting the advantages of 5G (and e.g. URLLC) and allowing a factory to quickly set up their own private 5G network.
This is on the consumer facing side. On the operator facing side, infrastructure is moving more towards virtualization and decoupling. Trying to make it easier to use multiple vendors, and stop requiring custom made hardware. And in general, moving towards commodity hardware and something closer to 'infrastructure as code'.
This also helps roaming and virtual operators (for e.g. the factory control). It also helps a bit with the ultra low latency part by decentralizing the routing part and moving it closer to the devices.
So "what is 5G gonna do for me" is mostly the 'faster internet'. But the idea is that it will enable widespread innovation that you can later use. With some luck (governments are thinking) being ahead in deploying 5G might also help boost your economy by boosting innovation.
Having a new technology to bring to the market is good for the industry. Service providers, hardware providers, salespeople, etc. Most consumers love the hype!
I needed a new test phone the other day so I bought a new iPhone 11 at a Verizon store. The salespeople could not wrap their head around my choice because it doesn't support 5g. I gave them some great reasons and they relunctantly took my money.
The problem with Fairphone is that it relies on binary blobs which are not supported after a short time. Planned obsolescence is there, even though it is not the company's fault.
Android was explicitly designed to make it easy for component manufacturers to keep their drivers closed. Now that Qualcomm essentially has a monopoly on Android SoCs you can't really build a modern phone that will have up to date software in 3 years.
This is why projects like the Pinephone and lebrem5 use such weird SoCs. Open source drivers are absolutely the only way to know that the phone manufacturer will even have the ability to maintain up to date software.
Maybe I'm wrong, but I think the only reason you need a SoC is if you're trying to make the phone small. If you don't care about size you can have discreet components. There are millions of people with big enough pockets (literally and metaphorically) , who wouldn't notice the extra weight of their phone.
edit: discreet might be the wrong word because they would all be ICs but not quite a SoC.
The other issue is that none of the SoCs on the market with open source drivers available are remotely competitive in performance with today's Qualcomm/Exynos/Kirin.
The Pinephone and Librem 5 are both using chips that are competitive with low end Snapdragons from 5 years ago.
I think the RK3399 paired with an external modem is about as good as you can do today without needing blobs, and while it performs a whole lot better than the current Pinephone/Librem 5 it is still quite far behind today's SoCs.
In most cases the open source drivers for these mobile SoCs are reverse engineered rather than released by the vendor, which is why they typically only exist for chips that are at least a few years old.
SoC vendors like Qualcomm do not support their chipsets more than a few years. Without new drivers, no updated Android versions. Without new Android version, no security patches. How could a company support an unsecure device that might be hackable by any Play store app or script kiddie?
Software and hardware support (from the manufacturer) are tied closely together, and the industry makes it really hard to achieve this.
Have you seen the fairphone? It's exactly this. The first one let you replace all components without any tools. The latest one needs a screwdriver (which it comes with).
You can upgrade a fairphone 3 to a fairphone 3+ yourself just by buying the updated components. It's pretty cool! But it isn't cheap.
"I'm really surprised one of the android vendors hasn't focused on the niche of easy to repair phones."
With the right marketing, this could be very successful.
"A repairable phone built to last," or some other creative slogan and marketing campaign centered around how wasteful and expensive other phones built around planned obsolescence are.
Caring about the ecosystem, recycling, and reuse is mainstream now (witness the BBC right-to-repair article itself, published on a mainstream news platform), so a company showing that it's sensitive to these concerns should do well.
If a phone's hardware lasts 10 years, it's going to need 10 years of updates. This shouldn't be a big deal since Android is open source and Google is the one investing in development; hobbyists regularly port modern Android versions to ancient hardware in their spare time, so even a small team of full-time developers at a manufacturer should have no problems doing the same for a company's devices.
I wonder if the reason is really just that these manufacturers are all blindly following Apple's lead? To an outside observer, a lot of Android manufacturer's seem to be really stupid, however the reality might be different from their point of view. I don't know, but it's frustrating to see how little of a shit all Android phone manufacturers give. It's why I can't feel sorry when I hear that a company like LG or HTC are shutting down their smartphone division; it's their own damn fault.
It's basically because of drivers for the SoC (system-on-chip, the integrated chip from e.g. Qualcomm that has cpus, graphics, sound, camera, power management, etc). They're too messy/hacky/low-quality to be "upstreamed", it is mostly impossible for anyone but the SoC vendor to port to newer versions (closed-source binary blobs), and the SoC vendor is really not motivated to do so.
I think current phones are just about enough performant to withstand many years of use. I think such proposition using some older tech than current generation just wouldn't sell. People buy new phones not only because new ones look "better" but also perform better than older phones. If I think about my previous phone, the user experience was far from satisfactory. The current phone I have also isn't great, but acceptable. However, I'd love to change to get something more performant - I missed so many moments, for instance, when I wanted to quickly take a picture, but the phone just wouldn't respond for seconds.
I have a Mophie for my 2016 SE, perfectly acceptable tradeoff - when I want the extra power I compromise with a larger phone, but most of the time I don't need the extra power, and don't use it, thus have a smaller phone.
Is there a market for it? $200 Android devices with a big battery and decent performance and so on are out there, so I wonder how many people are worried enough about a swappable battery.
Jitterbug makes the phone to sell their service, so it's not directly comparable (it does demonstrate that the devices aren't that expensive to produce).
I was using a Moto G8 Power for exactly that reason, but I still think 3 days of battery life is just not enough. The phone could have easily been 4 times the size without being too big for me. Then the screen cracked and I made it worse trying to fix it because I dont have the patience to properly melt glue. Now I'm using a CAT phone which has cool features, and is supposedly stronger, but mostly the same problems.
I know the ostensible reasons waterproofing, planned obsolescence, looking cool, and being light weight. Though I suspect the real reason is they're just chasing the big players and are afraid to be different. It just doesn't make sense to me from a business perspective, if you know you can't compete in the general market, why not carve out a smaller market and serve that one well. An example of this is the Jitterbug phone.