If anything if you just go with “got good enough for the average user years ago”, that works.
I’m on a cat s62 pro with a 5 year old Snapdragon 660, and, while it shows its age, it functions just fine and will for the next few years.
If anything if you just go with “got good enough for the average user years ago”, that works.
I’m on a cat s62 pro with a 5 year old Snapdragon 660, and, while it shows its age, it functions just fine and will for the next few years.
I think they mean the software support lifecycle from the time of release, not how long the physical hardware lasts.
I also have relatively positive experiences with Chromebooks but also you can buy a Dell Latitude used on ebay and get the same experience. My $200 shop laptop from 2015 (Latitude 7450) which I only stuck an SSD into is going strong despite being kicked, stepped on, dropped, having oil and acetone spilled on it, melted a bit, and subject to other such garage abuse. (Though it is on its 3rd $15 keyboard)
The big car companies are aware of this need and have shifted their focus to become “mobility” companies (you can read company goals for shareholders, it’s public), so you’re not alone on this fact.
Unfortunately, this also means that they’re all pining to make mobility a subscription service and vertically integrate as much as possible, which means to have their own fleets, repair shops and refueling / recharging stations to cut down cost. This would net them a good recurring source of revenue - which is golden - and reduce operating costs.
In theory this is more efficient for mankind, but this also means you can’t fix your own cars and a lack of money means that one of the last (unofficial) safety nets is torn from you the moment you’re out of cash as you’re no longer able to just sleep in your car.
It also means maintaining the status quo when it comes to car-centric infrastructure, as cars are still cars. There is no real incentive to build separated bike lanes or walkable shops if the amount of people using cars is still the same.
This would in theory all work without issue for city-dwellers in warm climates, but the same issues that plague(d) telecommunications seem to apply to shared self-driving cars; Electric cars happen to be the most efficient for fleet use as you can rotate them easily with minimal maintenance, but their limited range and need for service techs would mean that you would have rural areas that would be less profitable to serve. Reducing the supply of manual cars would impact their economy of scale, making them even more expensive to own outside cities.
The only way I see mobility companies not ramping up the prices to intolerable levels and ignoring poor people is if we did the same thing as telecommunications in 1934; by making fleet operators common carriers, making them responsible for their customers, and allowing smaller operators to openly use the larger operators’ infrastructure (charging stations, repair manuals, parts and technicians) so smaller mobility companies can exist.
I’m surprised they’re bothering to focus on consumer devices instead of just going all-out on enterprise and business.
Cloud workstations make a lot of sense for when you need the extra grunt occasionally and have a rock-solid internet connection, but about the only reason the average consumer would want to use them on a portable device is gaming. Everything else you can do locally or as a web app.
And even gaming has been a bit rocky, though it has its small cult of followers.
I’ve had this thought in the past, but upon further research, on a smartphone scale it runs into a few issues.
A PC has relatively divorced components because it has the space to accommodate slots, but that already goes out the window for laptops; the best you get is RAM and maybe MXM. People generally want slimmer and more battery-dense products, so motherboards with everything soldered can manage to be absolutely tiny (see: newer macbooks) compared to ones where everything is slotted. Newer smartphones don’t just last longer on battery because they’re bigger; much of the internal volume of a smartphone is now to hold the battery as everything around it shrinks. Modular connectors cost volume and money, for a use case where both are at a premium; most people would generally rather buy a non-modular phone that gets better battery life and offers better water resistance.
As evidenced by the Note 7, batteries also need a decent bit of structural integrity around them to be safe, and the best way to accomplish that is just a 2-part box that’s been glued together. This has the side effect of making water resistance easier and slimming the phone down. Any swappable battery inherently has worse water resistance and volumetric efficiency than a pouch csll in a glued box, and manufacturers value both good battery life and things not blowing up.
Smartphone CPUs (or rather SoCs) contain not only the CPU bit, but important parts of the radio, camera processing, location processing, and GPU. Therefore Android ones come with driver packs from manufacturers that make it all work, but if you’re swapping components around you need to make sure your sensors like cameras have a way to install drivers (and a general compatibility standard); which means at best you’re swapping pretty big chunks of a smartphone at a time.
Even without forced obsolescence, smartphones have a relatively fixed lifespan of ~5-8 years due to software upgrades, how carriers buy up new radio frequencies, physical damage, battery degradation, vibration motor wear, camera focusing mechanism wear (from vibrations) and in the case of AMOLED display wear. This isn’t a knock on standardization, but you can already achieve this lifespan out of a smartphone if you use something like Advanced Charging Controller to limit battery charge to 80%.
If you look at Project Ara (which is the most extreme case of modularity), it failed because of all of the above. The battery life, structural integrity, cost efficiency, compatibility, and waterproofness would all be awful compared to a standard phone.
All you need to run a lemmy instance is a bit of time, a computer and an internet connection though, same as running any forum. The costs of all those are rapidly decreasing over time and are well within “eh who cares i’ll just pay it” ranges; a smaller instance can easily be run for less than 5-10 bucks a month on the cloud, or for the cost of electricity if you have an old PC lying around.
A lot of costs of keeping servers running is employing people to do it; if its not too much hassle to keep one maintained, people will keep servers online.
Source: I run datacenters and support various cloud apps