Exactly. The whole "modular phone" and related phenomena are effectively nerd porn and against 50+ years of industry progress towards cheaper, better, denser integration.
I’m glad that the nerd porn exists because I do care about sustainability but operating on Apple’s scale means those devices are a fantasy for the majority of people. Apple’s approach seems more sustainable considering how many claims they can continue to make. Maybe it’s all marketing but I can’t find anyone that invalidates their claims without mischaracterizing them (as this article does a ton).
One way, is to make it non-user serviceable, the other, to make it user serviceable.
Now, which do you spend immense sums researching? Which people do you hire for your company? And after you spend years down this path, tooling, hiring, designing, someone says "the environment counts, it should be user serviceable".
Well of course, with billions spent designing, and predicated upon current methods, AND with all the people you hire experts in closed, non-servicable design?
What sort of answer will you get?
Apple, and others, the entire industry, has created this industry to be like this.
If the same R&D was spent on user serviceable, it would happen. Cheaply. Easily.
So of course it's "very hard" to do user serviceable, because no one knows how, and no one has the experience, and no one is researching it.
And no, these little firms working at it, don't equate to Apple working at it.
It may not be on purpose, but to claim it isn't possible is unfair.
You know if the auto industry was left to its own devices, it would still be claiming electric cars weren't feasible too, right? And from their perspective, they were not! Because without billions in research, and iteration, it wasn't.
Just like with Apple and user serviceable parts.
Just as with cars, and bags, and everything else, we should legislate such requirements. So all players in the market must comply.
This seems a little disingenuous. The entire reason that user-serviceable devices aren't possible is that the tooling required to service them becomes more and more specialized as the devices get smaller. It's not a matter of whether or not they can make user-serviceable parts (which why I replied to a comment about things like Fairphone and other modular/serviceable phones) but whether they can scale that and the reality is that 1) it doesn't scale without being so expensive that users can't afford the devices and 2) most users just don't care about user serviceability. So the idea that there are 2 ways to build is a false dichotomy.
It's not a question of possibilities. It's a question of trade-offs. It just does not sell (or at least hasn't.) Integrating everything often equates to less cost, less weight, less materials, less power use, less thickness, more reliability, more water resistance, more performance etc. for the vast majority of the users who have voted with their wallets. And to a nontrivial degree there are physical constraints that dictate this not just marketing and R&D spending (do you want to have replaceable SODIMMs in your iPhone?) I bet the average user likes the idea of modularity and user serviceability but wouldn't want to pay $100 extra for modularity in their $1k iPhone. They are likely more willing to purchase peace of mind through insurance, ala AppleCare+, than to purchase serviceability.
But sure we can keep pretending modularity is free and therefore of obviously a nonnegative option value to the user.