[drdaeman]

That would be ideal, but this sort of changes the topic from reliability to interoperability.

I’m pretty sure this is merely temporary state of things, as IoT is still a novelty and most people are unaware about new failure modes.

I hope - as the realization becomes common - the demand for devices that will keep working until they mechanically break is going to eventually become a marketing point.

[BrenBarn]

I suppose we can hope, but I don't see the trend going that way. Without some kind of regulatory support, there's not much incentive for anyone to pursue that. It will be more profitable to produce tons of crap that breaks quickly so that people have to buy another one.

[drdaeman]

> Without some kind of regulatory support, there's not much incentive for anyone to pursue that.

I’m not sure there isn’t an incentive. For me it’s pretty clear one - I’m trying my best to avoid buying products that would likely fail to work. I did that couple times, I recognized and understood the issue, now I try to avoid repeating the same mistake.

Right now I’m in a minority, and my choices are limited or even nonexistent. Most folks out there haven’t yet tried and haven’t yet experienced the failures. Or had too few of those to mentally register. Either way, only a few vendors cater to our needs. So in the interim I have to settle for lesser evils, hunting for products that I can hack.

Yet, when I’ll - inevitably - become a part of a majority (as more and more people have similar experiences), I’m sure vendors will stat recognizing the demand and try marketing on reliability and independence.

Or so I hope.

[BrenBarn]

I don't think you're going to become part of a majority, or at least not via that path. Companies are competing on obfuscation. Most people don't have the energy to even attempt to determine which companies are reliable. People will just drift from one provider to another, hoping each will be better than the last. Also, there's no way to know if they're going to rugpull until after they do so. Unless there are some consequences for doing that, it's too easy for people to set up a company, sell crap that will stop working, shut down, and then just start up a new company. There has to be some kind of force directing customers to the companies that will do a good job.

[drdaeman]

Maybe you're right, but I hope you're not. Only time would tell for sure.

I believe you're underestimating "most people". Yes, sure, today it requires a lot of specialized knowledge to determine reliability. But if you try a lot of times you eventually notice the pattern and start asking more and more meaningful questions. Such as "does it work without Internet" or "can I disable updates"? That's the whole history of humanity - we try various things, learning as we go, refining our search directions as we learn.

Indeed, some niches, particularly those with high entry barriers, are not healthy. They lack proper competition and feedback mechanisms are broken by severe disparities. Those niches can be disrupted either through regulation (feedback through an alternative channel) or mass consumer action (amplified feedback).

I'm optimistic because I'm seeing signals that suggest companies explore all possible directions. For example, GE appliances have pretty much open control and diagnostic ports, with official SDK and corporate blessing to make enthusiast projects to extend appliance capabilities (with some caveats, e.g. SDK is not fully complete). When I was looking for a new washer and dryer machines GE became my obvious choice. And then there's also example of Kagi that sort of brought my faith in humanity, showing that wherever there's a demand, even if very niche, even if not fully realized by many, good things can and actually do happen. Sure, things are progressing very slowly, but as long as people can think and say "no, I'm unhappy about this and I don't want it" I'm sure we'll be eventually fine.

In the meanwhile, to accelerate, I guess best we can do is spread the awareness.

[BrenBarn]

Well, I hope you're right too. :-) I just see the trend as towards more of this, not less.