Comment by sonnyblarney | Hacker News Reader

Comment by sonnyblarney | Hacker News Reader

Parent Full thread

"5G, however, is not bullshit for IoT"

I humbly disagree because IoT itself is BS.

We've been talking about it for a decade, and I don't have any use for it, nor do most folks. The ideas are nice, but practicalities are not there.

There's a lack of standards, no consistent 'home hub', not enough use cases for expensive gear, security problems galore.

The 'beachhead' for IOT has thus been Google/Alexa, which are really Cloud services and don't integrate very well locally.

Theoretically, I can see my fridge telling me what to order, my microwave knowing how long to cook etc. etc. but I'm wary that any of it will come to fruition for a long time. But I'm not that concerned because I don't care.

I think IoT needs a 'killer app' that everyone 'needs' which paves the way for everything else.

Nobody was building anything for the data networks when BlackBerry came along with an addictive product. It was not until BlackBerry was printing money that Apple and Google took notice and thus justified massive investment in their mobile offers.

Where I see IoT going is the same place as apps already have. Sending off random telemetry data.

You may or may not want a fridge that tells you what to order. But advertisers want a fridge that spies on you so they can tell you what to order. Or the power company wants you to have a fridge that uses power when they tell it to.

What is coming is "You Are The Product, Act 2".

ps. If you upset the fridge company they will terminate your account and your fridge will fail-open-and-warm.

I sometimes wonder if it’s time to make a company that makes high quality dumb devices and hammers on competitors in marketing, calling out their specific privacy abuses, and the long term potential implications. Really play up the 1984 angle, private conversations being subpoenaed from Amazon, that sort of thing.

Or a smart appliance company that works with local hardware and keeps all data on premises.

> Or a smart appliance company that works with local hardware and keeps all data on premises.

That assumes there is a use case for data-collecting appliances that brings enough value to people that such a thing would actually be preferable over an equivalent appliance that doesn't collect that data at all.

I don't think that there is such a use case for common appliances, though.

IoT isn't about making 'things' for consumers like you and me, it's about embedding sensors in public infrastructure and creating networks of machines and sensors that interoperate. You'll never 'see' IoT or buy IoT devices, it'll just be embedded into whatever product you buy and it won't be opt-out. That's why we need public protocols for IoT sensors, because otherwise all of that data will be in the hands of giant corps and then we'll truly live in a dystopian surveillance state. This is the end game of surveillance capitalism; as detailed by Shoshana Zuboff in her new book.

> it'll just be embedded into whatever product you buy and it won't be opt-out.

It will be for me, because I'll disable it. With sidecutters, if necessary.

I think you're assuming that IoT = consumer home devices, which is a small subset of the market. Most home devices are mostly useless, but outside the home they're a big deal. The kind of IoT I've worked with has been for industrial/manufacturing and WAN-based data gathering - things like distributed sensors across a manufacturing floor, wetlands monitors across a few dozen square miles of watershed, road and vehicle low-bandwidth communications, etc. Practical applications for those kind of IoT devices vastly outnumber practical applications in the home.

Yes that's fine but the problems remain.

The carriers will utterly not present a viable solution to IoT. Neither will industry.

There will be 'application specific' things that use current networks to achieve IoT, but as far as the carriers are concerned, it'll be like 'low bandwidth mobile phones' or something along those lines.

I have no faith in carriers ability to innovate, and industry is not good at presenting 'grand magical architectures'.

Something like this could happen: auto companies get together with carriers, and ask for some kind of special device/rate/services. They get special gear made for cars. And then other industries start to use that architecture. Reason being, only Ford/GM etc. are big enough to have real impetus with carriers, and Ford/GM will have specific requirements + the 'long term horizon' enough to guarantee revenue stream. Then other industries hop on board and we have IoT that's used for a lot of stuff, but that was originally designed for cars. Something like that.

The carriers are their own worst enemy. Remember when they charged 10 cents per 'WAP' page with a little bit of text and a few links?

Mostly just because there isn't much urgency to improve in the west. There is plenty of IoT in China. I wouldn't say 5G will be amazing in this regard though.

I don't disagree with you. The carriers in my experience only care about making as much money as they can off data and even their "IoT" package offerings tend to be a joke in terms of actual cost if you want to do more than SMS between devices once in a while.

On the vehicle side, I've sat in negotiations between the big carriers and big automotive manufacturers several times now and I can assure you the autos are not able to exert nearly as much pressure as you would think. Data prices have actually been one of the big causes of crappy head unit experience and lack of OTA updates for the mass market.

Yeah, I think IoT is mostly bullshit that doesn't add much/any particular value. But you know what I do need? Windows in my house that either close automatically when it rains, or when I tell them all to via an app or something. Then I can open all 20 of back up when I want. or maybe open them 5%, or 20%, or all the way, or drop half way, and open the bottom halfway. Know why this will never happen? It's expensive to manufacture & install, the profit margins are way low, and it needs to be secure, meaning it should never call home to google. The opportunity cost of doing that is too high vs. just making some useless gimmick that sends user data back to Big Ad Tech so they can farm it in their high profit margin ad business... That, and maybe I'm an old fashioned edge case that just likes fresh air when the weather is nice. And I live where I could use this for most of the summer, as it's not usually too hot, and it rains all the fucking time.

I’d love to buy something like this. I’d also love window shades that opened shortly before my phone’s current alarm. These are both projects I have planned for whenever we end up owning a house, but I’d love to be able to buy something off the shelf if it had an open api, and preferably didn’t require an internet connection.

There are a couple window shade options out there, but they didn’t appear to be very open when I looked briefly. Maybe there’s some older-style zigbee stuff that I missed.

this already exists

I agree. I think it may be a dollar value thing. I recently doubled my income and still can’t see the value in expensive Hue bulbs. Their price for wirelessly dial-able colored light far exceeds my willingness to pay and deal with hub generations. I might budge if they were half their cost, or if I doubled my income again. But probably not because of the technicalities.

Seriously, what benefits from perfect information? Human stuff...healthcare, weather, investment choices, etc. the information I would like on my “things” are things like when things will fail or what I need to do to keep them. I don’t need nor want a fridge suggesting merchandise for me. The other use is wireless control-that’s also got limited appeal.

i got hue bulbs (adjustable white) for: - lights turn off at night/when i leave - lights turn on when i get home - i can ask google or siri to turn them on or off as i am doing other thing

as for the colour ones? I have on on my deck, its blue. sometimes i change it to another color if i'm asked but otherwise i can't find any use for the color ones.

Iot is not about you or me. It is not about phones or alexas. It's about "smart cities" and surveillance states - why do you think the chinese are so good at it? It is not long until we will be unable to escape 24/7 surveillance. England is a predictor of where we will be 10 years from now. And it won't change, no matter how much we scream. Classic tyranny of the majority. Or apathy of the majority. Every one is understandably upset about private data mining, but public is way, way worse.

IoT's killer app is V2V road safety, collision avoidance. All cars with OBD-II could be upgraded.

Reply on news.ycombinator.com