Just looking around me there are 45 wireless SSIDs, crazy interference and nothing but crap flying around my airspace. Half the time I can see people's bluetooth devices wide open in nearby houses as well.
Add to that the impending 9000 useless little devices that will be broken and this is the RF equivalent of space debris. It's got to the point I can barely keep a WiFi connection up for more than a few minutes already. Even on my 4G connection I'm lucky to get 4mbits due to the contention locally. I can upload at 40mbits though which is crazy
So literally last weekend I cabled my house with 5 runs of 100 meg ethernet and a couple of long patch leads in the living room and stuffed in a cheap 8 port netgear switch from ebay. Plugged the whole thing into my router and turned off the WiFi entirely just using it as a switch + NAT combo.
I can now transfer files between my desktop and laptop at 10Mbytes/second rather than 500k/second with dropouts. It is simply bliss. Even my wife is no longer annoyed about the prospect of dangling wires because stuff is working again.
It has got to the point that I'd rather have a broken CAT5 plug that falls out every 5 minutes than WiFi. I really don't know how all this tech is going to work when there is so much contention and noise already. I think it will just contribute to the pain.
"So literally last weekend I cabled my house with 5 runs of 100 meg ethernet and a couple of long patch leads in the living room and stuffed in a cheap 8 port netgear switch from ebay. Plugged the whole thing into my router and turned off the WiFi entirely just using it as a switch + NAT combo."
Exactly. I don't give out a lot of tech-guy advice to folks these days, and most people don't even think of me as "the computer guy" in their social circle ... but any chance I have I implore people to hardwire their networking connections.
No matter how good your AP is or how nice your Sonos mesh algorithm is or how clear your airspace is ... you will deal with random, annoying, difficult to diagnose wireless comms issues. It's very easy to make that entire class of problem disappear from your life. It's also a lot faster.
EDIT: actually, this post will seem a bit disingenuous when we (rsync.net) officially announce the drive-up feature at our datacenter locations. We're stacking the deck in our favor by using Ubiquiti long-range APs ...
Yup. When we built our home I had them run two CAT-5 cables to the center of every wall in every room in the house. Yes, lots of cable. Cable is cheap. They did this before insulating and closing the walls, which made it super easy. All cables go to the garage. The garage itself has four cables per wall (hint: never in my life have I parked a car in my garage).
The cables are not brought out to connectors at all. When a need for connectivity appears it is easy to tone a cable through the wall, find the wires, cut through and install a box with connectors.
Thanks to this setup wifi use is kept to a minimum and security is enhanced greatly. It's always fun to watch the reaction of phone company technicians the few times we've changed internet connectivity when they walk into a garage full of computers, lab and cnc equipment and an EIA rack with servers, a large switch, hardware firewall and patch panels. They usually say something like "OK, here's the cable, it might be better if you wire it."
I just don't understand this wireless craze. I get that normal consumers buy into it, but for anyone mildly technically inclined, nobody should be using wireless anywhere they can avoid it. Wireless is just a messy solution. The only wireless devices I use at home are my phones.
The last two places I've worked have gone completely wireless - and gotten mixed results.
The first place was a small business and their network guys were nails. They had the whole system set up in three days, and we were screaming (close to a 100MB both ways) and since we didn't have a ton of users, it was rock solid and really dependable.
Just landed a few months ago at a huge corporation who did the same thing and it's horrible. The connection drops in and out, the speeds are anemic (around 2-5MB during the day, both ways) and in out project room, we've had so many problems, they're going to turn the LAN outlets back on so the service will be at least reliable.
The crazy thing is I've spoken to other people on my floor and they've told me before they went wireless, the network was far more reliable and much faster (closer to 40-50MB) both up and down.
Sometimes I'm amazed how much better consumer hardware seems to perform over enterprise wireless hardware. I just haven't seen an enterprise wireless device I was impressed with.
Mind you, I assume enterprise hardware prioritizes security over performance, whereas consumer devices prioritize performance over security.
IDK, that sounds like a false dichotomy. Enterprise security just has you look up the credentials in LDAP on initial connect rather than having everyone use the same credentials last time I checked.
It's not meant to indicate it's impossible to have both fast and secure, but in terms of where a company's focus is on improving their product. Many more resources are invested in improving security in an enterprise product than performance.
That's what I'm now doing. I duct taped a couple of ethernet cables to the back of the couch and you just stick your hand down and grab one if you need it. I terminated the cables slightly earlier and used an inline coupler so you can replace the last 1m when the plastic clip inevitably breaks off after a couple of months without having to rerun 25m of cable.
I mostly use a desktop so this is primarily for the wife and kids.
This is crazy. Duct taping cables to the couch? Too funny.
What about tablets/iPads for wife and kids?
I don't know why you were getting so many dropouts but it's not because "wifi is bad". I live in a house with lots of noise from neighbours wifi, but no problems here. Wifi will be solid for months at a time.
I must have a good wifi router (Billion 7800NL). Not all routers are created equal, I've gone through a couple of duds before the Billion. Some deal with interference better than others. Let the router choose the channel automatically too, very important.
My house is old, quite large, thick walls and so on. But everyone uses wi-fi here from different rooms without any issues. A lot of simultaneous connections going on too.
"...turned off the WiFi entirely just using it as a switch + NAT combo."
I took this angle early on and have never had second thoughts. As they say, "It just works."
Glad you are seeing the benefits.
However, I think "WiFi" is a strong marketing signal. Even if it does not work as well as Ethernet at transferring data, it appears to work well to sell products and services.
"I really don't know how this tech is going to work..."
It may not have to work. Consumers of computing devices have developed a high tolerance for stuff that is either ridiculously slow or does not work. Many do not know any different: when none of the devices people use have an Ethernet port, they will never know the speeds they are missing. The choice of using Ethernet has been removed.
I am from the industry on the B2B side and this article rings even more true and much more strongly so on the B2B - industrial internet side of things. Go to any 'messe' (as they call it in Germany) or salon or expo and you see tons of connected/connectable industrial products but when you dig deeper into how those machines/components are going to be integrated, installed, commissioned, used and maintained, it becomes quickly evident that the vendors haven't thought those things through.
The dominant discourse when you talk to many OEMs is that this is a defensive play to be prepared for all those secretive inventions that Google and Apple are making in their X-Labs. They don't know what's coming so they are throwing everything at it to "cloud-wash" and "IoT-wash" their offerings even when their customers are justifiably scared about the risks of cyber-attacks that connected machines bring into their factories and plants.
In short, not enough RoI evident for the investments and changes in processes IoT mandates in the enterprise - especially in the production plants. However, based on my discussion with all the different actors, its likely that we'll find use cases in either Operations Optimization or Asset Optimization.
I actually think the industrial side of the story is where real value will be created. I'm also working on an industrial iot application, and would really like to chat! Is this your company?
No it's not my company - thanks for pointing it out though - I am cofounder at a startup and we call ourselves MachIQ. Looks like a Trademark purchase will have tp be made eventually :-)
you can reach out to me via LinkedIn/Google - rchikballapur.
As a meta comment, I can't get to the forbes story on a direct link because I adblock and scriptblock. I googled the title and got in via the google search free pass and I now see I'm blocking 19 social scripts/ads which provide zero to negative value for me. The point is no one wants spam or pointless interruptions or endless big brother tracking. The future of "smart devices" is your overhead lightbulb connecting to an absolute minimum of 19 social media and tracking sites and government offices to permanently record your every activity and no one wants that except spies and corrupt people.
If there's no end user benefit unless they can get the government to force the population to use them, most "smart" devices will end up doing the equivalent of the 80s/90s VCR clock flashing 12:00. Maybe the strategy to work around that is light bulbs of the future will refuse to turn on unless you give the home factory in China full access to your complete facebook, twitter, and linkedin profiles and perhaps your credit card number so you can purchase "upgrades" to run beyond 100 hours or beyond 50% brightness levels.
I'll be using dumb bulbs and laughing at people stuck with inferior smart bulbs.
Careful. If you don't source your candles properly you might end up with lead wicks, even though they've (shockingly recently) been outlawed in the US.
I've also got concerns about the material used to make the candles themselves. I started burning beeswax candles, and now when I'm around the more common sort it smells like factory waste, or like someone took all the chemicals under a sink and mixed them together. Unpleasant and disquieting.
I take a contrarian view on this. Such advancements / toys / useless things are vitally important to forward progress. What this article fails to recognize is that innovation is often built on the current "pareto (efficient) frontier" ... that is, it is built around the universe of available devices/applications/protocols/technologies/useless-things ... rearranged and consumed in new and novel ways. While the final application might not be a smart oven, as it might indeed be a useless item, that final application could very well be an application that requires the availability of a smart oven from which to build on, or one whose innovation was triggered by the availability of a smart oven. In other words, a smart oven plays a role in innovation and society as a whole, and shouldn't be dissed.
I agree - with the caveat that it should be "done responsibly" (which is a tough nut to crack in this context).
I'd love a "smart house" - the fewer things I have to do when I get home, the more time I have to relax or do my own thing! I know the promise of the smart house and the execution of the smart house are years apart, but I feel we are just now getting to the "promise" part.
Remember in the 90s when everyone said virtual reality and smart TVs were the next wave of high tech? And how all the big companies started putting money into R&D for said technologies? Remember how it never materialized? (Well, it's being realized now, 2 decades late.)
The "internet of things" is the same shit. Something people like to mention at TED conferences. But do we want it? Is the market ready for it? No one's asking that. If you ask me, we're not ready. At the moment, the tech is too expensive and too useless. A few things will win out with affluent customers (thermostats, audio systems). But putting chips in everything we own is a far off dream.
Funny how these things are. My reading of the last 20 years is the exact opposite.
We've gone from talking about how devices will change how we live and work to having devices that are changing how we live and work. Ok they weren't VR or TVs, they were sensor heavy mobile devices. But still.
And they're not completely pointless. They may not have solved world peace but they've changed comms in quite positive ways; I have to travel for work and video chat to my family is a wonderful thing.
Also, over the last year or so, I've seen a noticeable increase in people taking an interest with their health due to Fitbits and the like. Having that feedback loop, although a small thing, can make quite a difference.
I can see this sort of thing having logical extensions into the home because we now carry sophisticated control systems with us. It wouldn't surprise me in the slightest if we don't start getting QR codes on menus/food packs with nutrient content, for example.
Don't get me wrong, I fully appreciate that what I'm talking about here are all first world solutions to first world problems. But the idea of connected stuff affecting ones life isn't in the future.
Yeah but in order to reach that conclusion you have to abstract too much. We can all agree that devices are and will change how we live. But now we're not talking about VR, Smart TVs, or the Internet of Things anymore. Because the truth is, we never know what form the next wave of technology will take. And that's why I'd rather experiment than prophesize.
It only requires a bootstrapping application. Enough chips get made for that niche, the chips get cheaper. That opens up other applications. And so on. It's a 'natural process' of economics, with a time-constant related to marketing and chip manufacturing cycles. Used to be 20 years, but its getting shorter all the time.
I agree that the chips could be cheaper with the right killer app. It's a matter of time. Tech "thought leaders" think that time is now. I think it's still a few years off.
But what solves the issue of being useless? And what is that price-lowering killer app gonna be if IOT means shoving a wifi adapter in your blender?
No need to be very useful at all. Just convenient. We got remote controls for our media devices, to save us getting off our butts. Now its a way of life - channel surfing.
Yeah but that convenience is very useful.
I think we're just talking past eachother. Remotes make watching tv way easier. The question is whether some of these internet enabled devices make things easier or not.
Yup. And we won't know until someone shows us how much we were missing by not having them.
An internet-connected coffee pot or dishwasher is hard to appreciate now. But what if they alerted and called a service guy if they failed? That would be handy. We'd come to hate the non-connected device that didn't appear on our 'household service dashboard' app. It'd be like having a Victrola in the house.
As a coherent story this chair is terrible. The parents start off saying they couldn't find a good chair for their kids. So they designed a decent looking chair and it seemed like a great idea that it would work for kids until they were about 10 years old. That's a great solution, adult chairs can be pretty bad for kids at the dinner table. Anyway.
Then inexplicably, halfway through the video this crap about measuring the child's weight every time they use the chair. Nothing in the original problem says anything about tracking weight? Ridiculous leap from "can't find a good high chair" to "high chair with built in wireless scales".
Is there something in the kickstarter ts&cs that says "projects must have wireless connectivity and an app" that I'm missing?
Here's my litmus test for taking an IoT thing seriously: If nowhere in the entire ecosystem of that thing does anyone try to replace light switches with taking out my phone and fiddling with an app, I'll at least listen.
It demonstrates to me how fundamentally unserious the IoT is that people keep coming back to trying to replace a light switch. Light switches are really good, guys. They're fast, everyone understands them, and they're reliable. Don't get me wrong, anything can be improved, but:
1. Every IoT lightbulb thing I've ever seen has been strictly inferior to the light switch.
2. Even if you did manage to actually improve it, how much are you improving it, at what cost? I'm not going to pay $100 to make it very very slightly more convenient to turn on my lights.
Many of them aren't as 'smart' as they pretend to be, and can easily be MitM'ed. No subscription fees, local data storage. Depends on the functionality of whatever server it is connecting to, of course. I did this with a body scale - the online component is only storage.
Then again, most of this 'connected' crap is good as a hobby only. I like playing with it, but the actual added value is very low.
Because the device is just a neat bauble and deep down you know you almost certainly won't be using it 2 years from now no matter what the company does?
After two years either you'll realize that you didn't need that sort of thing at all or you'll realize that that sort of thing is exactly what you needed and you buy a a newer and better one.
There are many pointless devices. But there are also devices that actually solve peoples problems. They will be the ones that emerge as real companies.
I'm pretty sure whoever wrote this article doesn't cook.
I'm a huge believer in data and its ability to make things better, whether for companies (seen for years with Business Intelligence software and whatnot) and now, with IoT, for individuals.
Problem is, now that we can gather thousands of data points about an individual (what exact time I woke up this morning, how well I slept, how long I brushed my teeth, the calories in my breakfast, what I had for breakfast, what time the sun came up, my self-reported mood, the length of my commute, how many stop lights I encountered...) we have no idea what to DO with that data. The state of IoT is ridiculous right now because every company is trying to cash in on it using the old model of marketing ("Surely if we know enough about our customers we can create ads that they will want to action on!") and this old model is 95% dead now that the majority of people get their content served up via the internet instead of newspaper/magazine and streaming instead of cable.
I believe that once the marketing field catches up with the way people are actually living and works in conjunction with the makers in technology companies coming up with products and solutions that actually make people's lives better then IoT will start to make sense. I have hope that this will happen... but am pessimistic about the timeline.
IoT means data, and to me "more data always means more better", but data is only as good as the questions you ask it and right now we're really, really bad at asking the right questions.
For me the environmental impact is also a factor.I don't need a bunch of items sitting around wasting energy because they're all pointlessly wi-fi connected, have sensors, or are "always on and ready to go". And with more complicated parts things are more likely to break, so that's more likely to go in a landfill.
I think what we need is a nice standard icon representing clearly that a product does NOT communicate wirelessly at all. Then let consumers decide which devices are useful to have connected as part of an IOT and which they would rather not.
Just looking around me there are 45 wireless SSIDs, crazy interference and nothing but crap flying around my airspace. Half the time I can see people's bluetooth devices wide open in nearby houses as well.
Add to that the impending 9000 useless little devices that will be broken and this is the RF equivalent of space debris. It's got to the point I can barely keep a WiFi connection up for more than a few minutes already. Even on my 4G connection I'm lucky to get 4mbits due to the contention locally. I can upload at 40mbits though which is crazy
So literally last weekend I cabled my house with 5 runs of 100 meg ethernet and a couple of long patch leads in the living room and stuffed in a cheap 8 port netgear switch from ebay. Plugged the whole thing into my router and turned off the WiFi entirely just using it as a switch + NAT combo.
I can now transfer files between my desktop and laptop at 10Mbytes/second rather than 500k/second with dropouts. It is simply bliss. Even my wife is no longer annoyed about the prospect of dangling wires because stuff is working again.
It has got to the point that I'd rather have a broken CAT5 plug that falls out every 5 minutes than WiFi. I really don't know how all this tech is going to work when there is so much contention and noise already. I think it will just contribute to the pain.