Google shuts off IoT Core services shortly after announcing API stability commitments Google Cloud is turning off its IoT services as of August 2023, leaving customers with less than a year to find alternatives. In a missive to customers, Google Cloud said: "We're writing to let you know that Google Cloud IoT Core Service will be discontinued on August 15 2023 at which point your access to the IoT Core Device …
And about 6 products, ever.
I'm always looking for a thriving ecosystem, and then deciding Eff that, I'll build my ventures future in the middle of a ghost town run by a predatory monopolist which not only has a reputation for pulling the plug on platforms but also owns a directly competing company in the IoT space.
What were they offering? An MQTT server + done tie-ins to other cloudy bits like graph and analysis engines?
If that is the case, then at least migrating elsewhere should not be an insurmountable problem.
That said, that's why everything I design is based not on products but on protocols, the idea being that there is always a plan B + C to ensure that the product / service remains available throughout its entire operational life.
trevorde,
The missing bit on the end of the quote is .....
" ... But it was totally predictable .... it's Google after all !!!"
:)
P.S.
I continue to be 'Gobsmacked' by Star Wars being so deeply ingrained in the Psyche of the world !!!
The original book was/is dreadful .... it was offered to me by someone who knew I read SciFi.
Yes, I know I am in the minority ..... But my opinion still stands !!!
:)
"I felt a great disturbance in the Force, as if 72 voices suddenly cried out in terror and were suddenly silenced. I fear something completely expected has happened."
Fixed this for you. From the story:
Customers may be disappointed, but there also may not be that many of them. One measure by marketing analysis website enlyft put Google Cloud's share of the IoT managed service market at less than 1 per cent. The Google service claims 72 customers, it said.
Google drives the Valley mindset in that they develop something and dump it on the table, and expect people to put in a bunch of time to figure out how to get it to work for them, all while keeping in the back of their mind it is a Google offering so the plug could be pulled on a whim.
Azure and Amazon have more robust tool sets that users have gravitated toward, because they work for the use cases and they're supported.
So after a few years of Google having this half-hearted semi-product offering out there, they get annoyed that they haven't attracted the majority market share, and they kill it. They were never really behind supporting it as a product, just dabbling in IoT as a me-too portfolio. Google's stuff is cheap but involves a lot of DIY to make it do what you need.
IoT stuff is hard, it is such a broad concept that every vertical or application has a completely different set of requirements, necessitating vastly different solutions.
Whilst I agree on one level, Microsoft are forever deprecating features, or commands in Azure (and on-prem) in tools like PowerShell so there is no guarantee that what worked previously continues to work.
We have just hit this with Azure Kubernetes where a command has stopped working.
I think its the 80/20 rule. Googlers, and google, love to play with shiny things, love to announce pretty things, and love when colleagues and others say 'oh neat!'. 20% of the effort gets you 80% of the pretty shiny thing, but then it gets boring finishing it off... because effort. Without incentives - google prints money with the ad engine, so nothing else needs to generate cash - to finish things to a point where its a business, people get bored, can't be arsed, move on to other things, and things get abandoned. Simples.
Yes, I'm jealous, because that sounds really fun.
Personally I think it ought to become law, if only for ecological reasons (IoT gear not instantly becoming landfill) that any company deciding to close shop on a device is forced or at least encouraged to publish how the gear works. That way, others can choose to continue it or at least hack their own support for it.
I don't know what exactly we mean by IoT so I'll go with my own definition according to which, most products use ZigBee, Z-Wave, LoRaWAN or wi-fi or some kind of simple signaling over 433 or 868 MHz. Most of those protocols are open (those some are proprietary and require a licence).
One layer up, at the aggregator, popular choice are MQTT, JSON over HTTP or XMPP. Those are all own protocols.
The only real issue, appart from appallingly badly written implementations, appears to be that the shits have hard coded servers (and sometimes TLS certs) in firmware *and* that the cloudy server side of things is closed and proprietary.
Personally I don't buy or use anything that uses proprietary components. OpenHAB is a pretty good solution.
What you said + the massed floggers of dead horses trying to bullshit any prospective marks into parting with cash. And them internets are just full of sensors bulk manufactured in around 2017 with no homes to go to. The whole idea should just slink off and die. I'm biased... I had a "discussion" with the LoRaWan bunch a few weeks back about pricing.
... for Google.
With each passing product it becomes ever more abundantly clear that the only purpose of those products and services is to Vacuum up data about people, things, places, and the relationships between them, to feed the Google advertising engine.
Once they've wrung the good stuff out of a product or service they cancel it - it was never about providing service to a user (or possibly - gasp - even a customer!)
If we step back and take a look at who their actual customers are, it becomes clear that they’re incredibly successful…
The advertisers are getting (what they perceive to be) a great service. The “users” (if a service or platform on the Internet is free, you are not the customer: you are the product) get shafted in the process, but that’s just one of the costs of doing business.
The only items in my house that are aware that there is such a thing as an internet are my router, laptops, phones and TVs. Nothing else. I don't need my house connected to anything. It's not like twiddling with the thermostats by walking to them is hard. And this just proves why it's a good idea to avoid - as soon as your IoT service decides there's no profit, they pull the plug and your house stops working.
Same here. While I do have a so-called Smart TV, it has never been connected to the internet and never will be.
I've been shopping around for a new car and after a few test drives, I have eliminated those that run Google-originated software in their core(Volvo/Polestar etc). Who knows how much data that is slurping and sending back to the Mothership?
I only wish that more people would start to see the likes of Google for what they are... Ad slingers and Data slurpers. Avoid them and have a better life. (if you search on Google for my real name, I'm not there and long may that remain.)
I have an oven and a hob that have a wifi connection - but we never use them. The IoT device we _do_ use remotely, is the pellet-stove: it's nice to be able to turn it on in the winter a couple of hours before we get home. The app with our e-Bike is also useful (the bike computer on the e-Bike has a tiny screen in comparison to the phone). The TV is slightly too old to understand the Internet.
I do believe that there are certain items that benefit from remote analytics and control (SCADA style), such as solar arrays w/battery packs. In my view, this list is exceptionally short, hence my solitary example.
Unfortunately, shoehorning this technology into *everything* seems to be all the rage, regardless of any actual value. Hell, I can’t imagine *any* outside of a marketing fever dream.
> It's not like twiddling with the thermostats by walking to them is hard.
Saving 30% on last winter's heating bill says otherwise. And that was before the gas price doubled.
There's nothing wrong with IoT. A different story is *what* IoT you buy. If you go with the cloudy stuff (which is the majority of the offer) then yes, of course that's insane. You want to pick the consumer equivalent of industrial automation stuff instead, which may offer external access or talk to external services but does not rely on them to function.
Btw, do non consumer environmental sensors (usually lorawan stuff) count as IoT? A lot of industry (farming, natural resources management) rely quite a bit on those nowadays. I doubt they use Google though.
Which is an important point - it's not the "intelligent" devices that are the problem, it's the systems designed to require the use of a benevolent (or otherwise) vendor's systems to stay functioning, and the marketing driven fad for adding "with internet" to everything regardless fo whether it actually adds anything of value.
The Register 
Biting the hand that feeds IT
Copyright. All rights reserved © 1998–2025
