IoT is dead in the water right now because of the same reasons smartphones did not take off until iPhone. The dataplans suck. Steve Jobs insisted on unlimited data for a flat fee. If you go look at the M2M offerings for most data providers you will see they do not list prices. That is for a reason. I personally saw many of the actual prices. The ranged from 'not bad' to 'you got them to agree to that?!'

You can piggy back on customers networks to a point on 'home automation'. Everyone is trying to be the next modbus/zigbee on that one but charge a monthly fee for the privilege of using their protocol, or even better their service. The integration between services is kinda crap and the ecosystem is very shaky with no clear winners.

Bespoke systems are doing 'ok' but again no clear winners yet. Most seem to be setting themselves up to be bought out by some bigger player instead of shipping good product.

What we found with IoT was it was not the software or hardware. We had that 15+ years ago. The problem is integration and maintenance and connectivity. The integration phase you can see sales cycles of 2-3 years. Maintenance ongoing cost no one really wants to pay for. Connectivity is at the mercy of someone elses price plans.

The data plan certainly doesn't help but the fact most IoT devices actively put time and money into providing anti-features like demanding an unnecessary connection for basic functionality, an unpredictable sunset period of 'whenever we go out of business, get bought up, or just feel like it', needless privacy issues, and needless vulnerability to hacking and viruses and expect people to pay more for it including recurring charges? Using ubiquitous computing to turn their devices into dumb terminals which must defer the home server?

It is downright stupid runaway greed run amok while viewing themselves as entitled to customers - they couldn't make their contempt for their customers clearer if they had put photos of themselves flipping the bird twice to the customers on every box and giving making every product name a combination of three to five randomly chosen swears and insults.

If they actually started with an appliance and asked "what features would a customer find useful" instead of "what can we charge them for and put behind a pay-gate" then we would have something worth considering. Washer/dryer unit which can text or email you when your load is done and can be programmed in sequence? Great. Smart locks for changing entry codes remotely and phone communication with an app? It is actually made for its niche.

Instead their strategy is like including a remote triggered bomb in every product and then wondering why people rarely buy their stuff.

You hit my sentiment on that perfectly.

I am currently going through something similar on some open source software. Basically the people who run the project are 'done'. But it depends on other services and those people are 'done' too. They got tired of the endless support requests with little recompense. So they bailed out. Yet from my 'customer' point of view now something I used no longer works correctly and no amount of money will fix that. So now I have to spend a bunch of time trying to figure out something that works for me for something that 'worked'. This is the situation many of these companies leave other people and companies in.

Fail usable is what IoT should be. Instead it seems to be just 'fail and you are out of luck'.

Most IoT devices do not need Internet for the home automation features that actually provide value to users. They need a LAN connection at most. They say they need Internet but it's typically just used for exfiltrating user's personal data and analytics, and as you say for implementing needless payment gates and remote-triggered bombs.

I think the issue is that IoT is not really broadly useful for normal people.

IoT devices have been a big thing in the manufacturing/industrial sector for a while now, because collecting telemetry on equipment is extremely valuable. The growth in ML has been a huge value-add in the sector.

I can't really see the value in IoT devices for the consumer market. There are niche applications which may appeal to a small group, for example security systems or climate control. But, for the most part, there's really no obvious "killer app" for consumer IoT devices. The few ideas with potential are already getting integrated into the iOS ecosystem (health monitors).

Agreed. I'm working on an IoT platform and there is a lot of interest from big industrial players (energy sector, manufacturing, fleet managers of all sorts). IoT can help companies save money either through analytics or reduced need for labour, but I don't see too much value added to regular people.

I think "low level" here means "low skill level", not the programming definition [0]

0 - low-level programming language is a programming language that provides little or no abstraction from a computer's instruction set architecture

I would like to see some concrete examples of what is meant by low skill level "programmer" work. In my opinion, programmer, developer and engineer mean the same thing. Is there a market for "Hello, world" that I am not aware of?

I think of “an Excel workbook” as being an actual workable example of the pattern.

I think that gives a lot of solace that it’s not going to collapse the market for devs.

Simple CRUD websites maybe, adding simple things to e.g. wordpress deployments.

This is good to hear, and I've felt that a lot of people are underestimating the IoT explosion that is to come. The automotive industry is just finally on the cusp of it. Buy a new high end appliance, and you can probably connect it to a network. Everyone loves "smart home" accessories. I think demand is going to sky rocket.

Why would I want to buy a fridge in 2021 with 2018-era tablet embedded and expect to use that fridge in 2035 with who knows what WiFi standard changes in between?

I’d pay extra to not have a tablet embedded in an appliance.

Yeah, I'm not saying I personally want that either. I prefer dumb appliances with less to go wrong, but the industry seems to be going that way?

CMS replaced the guy who turned your .doc into HTML (this used to be a job). Many of these people became LAMP CMF specialists. That transition was at least a decade ago. Sysadmin roles have been under the most pressure since.

> So no, if anything, it's going the opposite direction, more code, more software.

That's the paradox, the more we automate, the more work we have to do. All because of scope creep.