Apply HN: Connyct – Complete IOT Solution | Hacker News Reader

Apply HN: Connyct – Complete IOT Solution

9 pointssushant2mainali10 comments

What are we doing? Connyct is a complete IOT solution. We will be the facebook and twitter of IOT.

Features: 1. Create a Device: Just define the data structure and the GUI associated with that ( all with a UI interface - no programming required)

2. Follow / Share: People can follow or share their devices. For example a scientist places a weather station in Antarctica, you could follow that device. Or to be commercial, what if Kim Kardashain's weighing scale uses Connyct? Million would be interested in following her weight. Developers of the device just focus on the device, we take care of the rest.

3. Data Analytics: Sensors/Devices just send their data; We perform complex data analytics on them.

4. Interconnection: You can do trigger actions on the devices you own (in real time) - for example, if temp in antartica weather station is less than 30, email me or turn on light at home, etc . All this with no programming.

Use Cases: 1. Companies make internal phone apps all the time. In the future, they are going to make internal IOT devices and they need a platform for that. We are that.

2. You have an idea for an IOT device, use our service to create the device - you focus on the device coz the backend, website, phone app is all done. Scale from 1 to a million.

3. You own a company and need a custom IOT device - you need to hire a h/w designer, backend developer, phone app developer, maintain a server etc. Use Connyct and you only need someone to send us the data - may be an intern? Look at the cost savings.

Saving the best for the last: FREE for developers and users. Only cost for storing data along term.

Youtube links to give you an idea of what we are trying to accomplish: https://www.youtube.com/watch?v=nk5QK7SE2ME

We plan to launch in about 2 month time. Please feel free to drop in your suggestions/questions. We are also looking for a 3rd co-founder. If you like the idea and are interested, please feel free to drop in a comment.

This is cool. Not sure about the market size though. Plus this is the kind of things that needs a lot of engaged people to thrive, how do you kickstart growth?

One of the main feature is that it is free for developers and end users which we hope will attract developers. Also we plan to cold call small-medium sized businesses to get to our first 100 customers. Along with that we will invest time and resource to be active in blogs and forums and programmers circle so that more hobbyists and casual programmers are attracted to our platform.

Love the idea. Having a platform for which devices can be easily routed for accessibility. On the example of a scale being accessed by millions. I'd want to have a simple : PRIVATE / PUBLIC switch for this implemented. The concept of allowing your fridge to suggest food that would help you lose weight might be something to aid a user but shareing any of that with others sounds like a micro, or even macro invasion of privacy. And maybe your friends might not need this information but soon enough, large companies are going to be wanting to know how to better sell to the right customers.

Interface of Twitter / Data algorithms of Tinder / Marketing (ads) Stream of Facebook

Again, I find this idea & business attainable. With a name like Connyct. It sounds right & feels wrong. Its IOT, these off name companies are only a fad. IOTA / InsideOuT / IOTFeedr / idk, & seriously no offense. I have been studying Branding & Design, its become second nature to see a business for both its service & how well the name holds to its service. Cheers & good luck

Thanks for your feedback. I like the private/public idea and that is what we are doing. However, we are implementing it slightly differently and that is subject to change as we go along. You don't share a device with your friends. You become friends with devices (and IOs of that device), which is a unique concept. So each device has its own friends vs friends sharing their devices with friends.

May be National Geographic can plant an Internet Connected thermometer in a volcano and we might have many many friends/followers of that device.

So, if I have 10 devices ( each with 3 data points); each device might have different friends with different levels of access to each of the data points on the 10 devices. The devil lies in the implementation though. If we make an excellent interface to manage the sharing and becoming friends with devices, we will win. And the sharing becomes crucial not for home devices but for commercial devices like may be a weather station at an airport (which the entire city may be interested) or like I mentioned the weighing machine of Kim Kardashian ( lol ) or may be goods on a commercial truck, the real time location of which is needed by 7/30 employees in a company. I want to develop this framework into something more than just for devices inside a home (which is where a lot of IOT focus is right now).

Since we would have the data, we can do stuff with it. Have not thought about what, but we plan to make it to good use.

Thanks for feedback on the name, we love positive criticism. We will look into that and research more.

Nice idea.

How would your service handle poor internet connectivity of the IOT devices? I've done network connected devices for companies in the past, and the trickiest technical part was always how the device/service behaved in the presence of network failure (was harder than making the devices pass IP69K testing).

As of now we here is our plan: 1. Devices can either be always connected ( using websockets / ajax / any other technology) or 2. Devices can define what is the maximum time before which a device is deemed as disconnected ( could be secs / hours / days). If they don't send data in a certain period of time or if they don't respond in a certain period of time. 2.a. For devices following this - the time will depend on internal memory and how much they can store and process, as whenever they have connection, they can send time stamped data in bulk. We need a way to determine how to acknowledge since we may lose connection after getting the data but before sending ack. At that point, we may just discard the data ( but there might be a better solution ) and device can send data ( not very effective). 2.b. The developers can determine the timeout in sending commands and also do they want to queue up all pending commands or just send the latest one.For instance, if I want to turn on something, and I send 100 'on' commands; do we have to send 100 of them or just the last one and for how long is the command effective before it is determined a timeout and we can flag a connection lost. The developer will be able to define all of that and for starters with simple application, they can just leave all that to default as most of the times it will just work.

How the h/w implements it, is up to the hardware developers ( we plan on developing some ourselves using different technologies and hardware ( raspberry PI, Intel Edison, some using sim cards, etc and using different protocols) and others can leverage off of that). The server side: receiving data, sending data, analyzing data, displaying data and detecting disconnection is what we will handle and the phone app / website for all of that.

I think you should add both real time video streaming capabilities and an Augmented Realty interface.

Massive Data streams like this should be carried out by other more serious companies & then implemented into other businesses, much like the 'login with Facebook' button is used. 'login with occulus' or something. Plus I can see a lot of crazy stuff coming from the first years of this transition. Is IP addressing even going to be the standard when everyone's phone, fridge, toaster, scale, stove, mirror, & tweezers are internet connected?

In my opinion IP addressing may still be feasible for some time no (thanks to IPV6). But I clearly see a future where the internet and its protocols that exist now not working for connected devices. This internet was created with humans in mind and now by adding protocols like MQTT etc, it has tried to become a hybrid system. I would love to see a new network just for devices.

And as far as going into interfaces for VR, we do not have plans for that yet. We plan to support data streaming though. As a startup, we don't have the resources required for that. But if there is demand and we grow, we would love to look into that. Business is all about evolving.

We don't have plans for that right now. However we have plans for data streaming. Good suggestion though. We can keep that as a feature we might want once we start growing. Right now, with our resources, we might not have the bandwidth to implement that. Surely a future feature we can look into if it generates enough interest.

Reply on news.ycombinator.com