Let’s build the Internet of USEFUL Things, not just the Internet of Things. IuT ! IoT
That’s what we’ll be doing over the next five weeks. The second challenge of the 2017 Hackaday Prize begins today. We’re looking for the best ideas we can find for useful connected devices. Twenty entries will recieve $1,000 and move on to the final round to vie for the top prizes ranging from $5,000 to $50,000.
There is no doubt that the future is connected. It has been our future since the advent of the telegraph, and we’re unarguably becoming more connected at a faster rate. The phone in your hand, pocket, or bag connects you to the bulk of human knowledge. But it doesn’t yet connect you to very many “things”. It won’t be that way for long.
Already we’ve seen cameras (security, baby monitor, and everything in between) appear as some of the earliest connected devices, and they’ve brought with them all of the unintended consequences of poorly secured computer gear connected to the wider Internet. At least remote cameras have a purpose; there have been more than enough product launches for things that don’t. Our go-to counter-example is the Internet-connected toaster which is the topic of our wonderful art from Joe Kim this morning. Who needs to toast remotely? Nobody.
Let’s Invent the IoT
Here is our chance to do it right. How can Internet of Things make life better? What things become more meaningful when added to a network and what does that look like? How do we continue to connect our world while safeguarding privacy and being mindful of security. Finding answers to these questions will lead you to Build Something that Matters.
Make it so
https://www.youtube.com/watch?v=RHevGGYuAsw
IOT s/b Internet of Tools
Not the same concept.
The idea behind Internet of Things is not all that different from the idea of Augmented interfaces. The bottom-line goal is that a person can query a thing, and control it. If it’s a shower-head, then it can tell you how many gallons passed through it, and you can set the temperature, flow rate, and maybe even nozzle. The idea being, that thing is an internet thing, all by itself. A “smart house”, or “smart bathroom”, or Augmented Reality interface, or random arduino hacked by the homeowner can then come along and provide functionality on top of that, by managing all of the “things” in a coherent user experience. Is a showerhead a tool? Not really…just a thing.
Is a spanner a tool??? Nah, it is just a thing…..
Well I think of things that get worked into manipulating states of things into a known/wanted state, a tool:
Shower head: A tool that is used to dispense an aquatic fluid consumable* over a subject in conjunction with a separate consumable to perform the task of cleaning said subject.
Like how a circular saw is in a table, yet as it is the active component in the cutting of things, it is a tool, not a thing and not a furniture (Although can be considered a piece of workshop furniture, same as a full blown mill/CNC).
*Well I get billed for the amount of water consumed, thus it is being, as it is able to be, consumed (Composite word: consume + able == Consumable)
I think he meant Internet of “Tools” as an insult, because in most cases you’re running infinitely more risk by having things connected to the internet which could then be used by other people maliciously.
Isn’t ‘internet of tools’ the programmers of current iot kit?
One interesting project.
I may be bias.
https://hackaday.io/project/20960-an-iot-unattended-3d-printer-w-auto-eject-sys
Unattended 3D printers make me nervous, in a “this might turn into an unattended bonfire” sort of way.
Well, then it’s high time somebody made one either reliable enough to leave unattended or at least make it really hard for the thing to fail in a way that burns your house down. Shouldn’t be that hard to detect that the printer is not doing what it’s supposed to and take action.
Unattended as in it prints tons of useless stuff, or that it chatches on fire?
Fire detection is a well solved problem. Add a solenoid-controlled fire extinguisher and you’re good to go.
Have the solenoid cut power to the printer at the same time.
much better print eject:
https://www.youtube.com/watch?v=Viy928RLsdU
Internet of Toasters
Ah, the 8-bit toaster
256 settings as follows –
0 => Off
1 => Warm
2 to 255 => 254 proportional densities of smoke.
You know why techies don’t “get” cooking, they think it is a skill involving your eyes and your hands, when really it is your sense of smell that is the key and technology isn’t even remotely close to a commodity HD camera equivalent of chemo-sensing.
I have raised two kids so I get the cooking thing lol.
I think it’s more like the old (1980’s) analogy of the kitchen designed by a man.
For example – I have an electric frying pan that has two issues.
The first issue is very poor coupling between the heated pan and thermal sensor that causes a significant time delay between the pan temperature and the sensor temperature.
The second issue is that the thermostat has an exceptionally wide temperature hysteresis.
These two issues combined make it the anti-PID from hell lol.
I Just use the temperature control as an on/off switch and regulate the temperature manually.
The old stereo-types of men who can’t cook and women that can’t change a car tire are gone and I am glad of that. If anything they have somewhat reversed for some families.
My (now adult) daughter changed her car break pads last week. Her mind has not been poisoned by any belief that being female is somehow a limitation.
I see other woman changing car tires on the side of the road while a male millennial looks on in amazement.
I know plenty of men who are avid cooks.
And now, post 2000AD, some men can actually be trusted to design a kitchen.
Almost all of the worlds top chefs are men, and always have been. As for the kitchen design thing, well builders don’t design anything they just plonk stuff where it suits them to build it in and as for architects of any sex they should not be able to touch a kitchen design unless they can demonstrate how one works by cooking a complete meal for an entire family. It is amazing how many females can’t cook, my mother was hopeless no matter how hard she tried, I should have got her nose tested, LOL.
I think it would be very easy to determine darkness of toast using a visual algorithm. The hard thing would be protecting the camera from the heat. I put a pi cam in my window as a security cam and it burned out the first 80F degree day, and it wasn’t even hit with any direct heat or sun light, the most direct angle was about 49 degrees.
This is exactly what I was talking about. “But I haz muh RPi camra…”
Can you cook at all, I mean at the level where you can grab a bunch of raw ingredients and produce something impressive without looking at a recipe? It is a nose skill, ask any competent chef.
Oh my, you’ve completely proven me wrong. I should’ve realized my inferior cooking skills would render my facts useless. What does it mean if I ask a competent chef and they disagree with you? They must be incompetent, eh?
All hail , the best chef and arguator on the interwebz evar.
OK then ask your mum next time she lowers a tray of chicken tendies down into the basement for you.
90% of food preparation is exactly that – preparation – it’s only the other 10% where the nose plays a significant roll.
It is actually extremely hard to tell if toast is done with a visual algorithm because you have to somehow account for the variety of bread types. Even assuming you can baseline them when they are put in the toaster, something sufficiently dark like Rye would probably turn out to be toaster incompatible.
I thought about this before I posted. I’m pretty confident that even if it is difficult in RGB spectrum, you could do it from IR or similar. A combination of techniques can account for all bread types.
I can only reply so deep so I am replying to One Thing Two Thing here.
You probably wouldn’t want to use IR to analyze toast because the SNR would be terrible. Literally every surface inside a toaster, except the toast is designed to have a high as possible IR signature. And please don’t take this the wrong way. I’m not saying this is impossible, just much more difficult of a problem than it appears at face value. In fact, as useless as this would be for toast, it is an extremely useful bit of engineering for manufacturing and some very smart people at very specialty companies currently work on this problem both from building equipment that can survive furnace environments to take measurements and the heavy duty signal processing required to get something useful from those measurements. It’s a cool field with a lot of depth and you should do some reading on it when you get a chance.
I have an older system to indicate when the toast is done. It’s not like this IoT system as it isn’t Wi-Fi but it is wireless.
The indicator that indicates when the toast is done is fixed in another room. The signalling method is rather ancient. It uses smoke signals to set off the indicator which makes a very loud alarm noise.
I use the same system!
It sets off all of the alarms, so everyone within 100 feet knows their sandwich is almost ready.
OMG, I love that graphic illustrating this article.
I thought it would look interesting if the smart phone showed “toast status” and have flames coming from the top of the phone!
Here it is in 3D with the hackaday logo on the side…
Because I HAD to…
http://www.thingiverse.com/thing:2288669
http://thingiverse-production-new.s3.amazonaws.com/renders/32/54/2a/ba/37/d9fd05ebe76a424cc0bf2eb586d053a0_preview_featured.jpg
Now, 3D print a bunch of them, and someone will design a catapult to launch them at trade shows and makerfests.
One thing that would make the IUT a lot more useful would be if you had a meta-layer of IOWYA, the Internet of Where You Are.
I was thinking of a device that opened the curtains when you entered a room if it was daylight (and closed them otherwise), a trivial task with IFFT except that we really don’t know where you are in your house.
You could also do other things like slightly bump up the temperature in the room you’re in and down on the rooms you’re not in, etc. Music could follow you around, etc. It could know if you’re in your car and/or in your driveway so it could turn on the lights, etc.
So were I to enter this contest I’d propose a small very inexpensive device that could be deployed to every room and simply detected as much about occupancy as it could. Clearly it would be bad for this to be hackable :-)
then you could expand capabilities to try to detect anomalies, like excessive heat, excessive humidity, combustion byproducts, etc. Even as security with IFFT if it knows you’re in your car and it detects someone in your house…
There is an open source project about 10 years or older that tracks your bluetooth signal strength around the house. Search the web for it.
I assume by IFFT you mean IFTTT.
You should know that IFTTT is basically just how every single computer program works. Only it’s in the cloud. For clueless people. It’s like the “bottled water” of tech. Don’t be fooled.
IFTTT – I stand corrected :-)
you can do the same now with a module interrogating the luci RPC package on openwrt firmware access points, using the excellent home-automation server in python, nothing then required to leave your local lan.
Though I mostly use it on a practical basis to log when my son has sneaked his android phone into his bedroom for some late night usage that he thinks his old dad doesn’t know about, and ask him searching questions about it when it gets too much, I have tinkered with zone presence dependent lighting and things, flaw in it is I don’t always have my phone because I don’t like being tethered so it can’t be relied on.
Bill Gates had similar ideas when he built his mansion years ago. It required each person carry a small transmitter to identify where they were, and priorities were assigned, so that if Bill entered a room, it conformed to his preferences.
(Well, I THINK Bill had the highest priority, I never asked his wife…B^)
Stop putting things on the internet that don’t need to be on the internet!
https://www.youtube.com/watch?v=hMtu7vV_HmY
I only skimmed the first 5 minutes, because I could tell the nihilist ignorant viewpoint of the speaker. The importance of firewalls should be the take-home. Has nothing to do with IoT. Access control is important, whether it’s your banking interface, or a steel mill.
The whole thing is a stupid straw man.
This all depends on what one considers “USEFUL”. I find many of the ‘wonderful’ household devices that are connected to the internet to more of a hassle than anything else. Besides – how many of you that use such devices have ever considered that actually checking on the device(or it’s contents) to be something of a daily ‘brain exercise’.
Personally, I keep a running list of nearly everything in my refrigerator, freezer, and hall closest(soap, tissues, etc.), as well everything currently on my ‘dry & canned goods’ shelves – and I keep it in my head. I do the same with many personal contacts and family, just in case a SHTF catastrophe should happen and I cannot access my tablet which holds electronic records for me(hard copy is kept in a second place outside of my home, in case of fire).
Oh yeah, there’s all the lock codes and other things I keep track of in my head too….
The idea is fairly simple. It’s even been used on a few movies and TV shows – it’s called a “memory palace or vault”. You just learn to remember things by their location in your house, eventually you can expand this into remembering any number of odd numbers and items. It just takes practice and constant use of your brain – something that many people rarely use for things like math, or remembering things in the fridge.
Good for you – apparently you don’t have kids.
I award you one Internet Merit!
I have tons of random things in my home, in dozens of boxes stacked everywhere. I know exactly where everything is supposed to be. My wife on the other hand can’t remember where she put her glasses down 2 minutes after taking therm off. On multiple occasions she hid something so well (so we won’t loose it or throw it away by mistake) we were unable to find it few days later. And when she cleans up the (supposed) mess, she is screwing up my mind a little, because my mental map of everything doesn’t match reality anymore…
All my tubes and wires!
One correction –
Quote: “My wife on the other hand can’t remember where she put her glasses down 2 minutes after taking therm off”
I think you mean … My wife can’t *see* where she put her glasses …
It doesn’t have to be on the Internet, but maybe the home LAN.
I would like to know:
Garage door: Open/Closed
Dryer: Entering Cool Down cycle
Clothes washer: Finished cycle
Water softener: Buy salt
Shop light: On/Off
Furnace: Change filter
Chest Freezer: Daughter left lid partially open
Mailbox: Mailbox door has been opened (Mailman removed/placed mail)
All of these would be great, and wouldn’t cost manufacturers anything to add. But their greed stops them.
Washer/dryer — this is one of the most frustrating. They don’t even have to go to all the “trouble” of adding an IP stack. They could just add a small TTL serial output that spits out SensorML. Hackers will do the rest. But no, unless they can spend millions on creating a “working group” to influence a “IoT standard”/”framework” to heavily bias their devices so that it *looks* open but really *coerces* people into a single manufacturer’s “ecosystem”, they don’t think it’s worth doing. Stupid greed.
“Chest Freezer: Daughter left lid partially open” — greed hits again. They all want to create something that tracks your food, so they can auto-reorder, and get a percentage of that revenue stream. But that’s not useful for the consumer. What’s useful for the consumer is: temperature, door open/close status, compressor on/off status. That’s it. No special monitoring program or proprietary software either please. Just output the damn data in SensorML and let something else handle the rest.
Water softener: probably has the same issues as washer/dryer. They could make it open, but decide not to because greed.
Furnace: all it needs is a simple airflow sensor
Mailbox, shop light, garage door: Switch contact and <$10 microcontroller. Done.
Most of the unuseful application examples were those of a user with single device that push/pulls data to/from the internet. The more useful applications are where a single user has thousands of connected devices.