One of the greatest misconceptions around IoT today is that it as easy as plug and play to integrate IoT devices into your project. Sure it is not that difficult after the nth project because you now have the knowledge and have built an ecosystem into which you can add your IoT devices etc. Before this stage, however, it is definitely not plug and play.
I think that this misconception is one of the biggest reasons contributing to the failed IoT project landscape. Technical people have an idea that the learning curve is only a few days to learn how to integrate all the devices found in the IoT landscape. In some case it might only be days but in other cases some of the devices have many moving parts that have to come together to form one seamless application that you can just drop into your environment or system.
Take LoRaWAN for instance, This is such a simple concept and to everyone who hears about LoRaWAN for the first time, the first instinct is to say, "It's just a radio that communicates with another radio hence it will be easy to integrate"
In order to intgerate a LoRa device into your system you will need to know the little quirks of LoRa and how the devices communicate to a gateway and how they join the network which in turn will need to communicate to an LNS - LoRa network server, then there are a few LNS providers and this will mean choosing the LNS that best suits your purpose. So will it be TTN, Chirpstack, AWS IoT core, LoRiot, Thingpark or one of the other LNS which are apperaing on the landscape monthly. Each of these have their own little peculiarities that the techical resource will have to find out and deal with.
Suddenly a seemingly easy project has become difficult as your technical resource now has to battle with the LNS and figure out its own peculiarities. Then we have to send the device data which is still encoded, to the backend server where it will need to be converted from Base64 encoding and then once you have the hexadecimal payload, you will need to figure out by using the documentation of that product whatthe sequence of bytes means and which bytes need to be parsed from the hex payload in order to get meanigful device data.
Once you have the device data isolated then there are the mechanisms needed to store and visualise this data. Sure there are many opensource projects which you can use which will do most of the heavy lifting however you will still need to glue it into place within your own ecosystem. This is where many projects get stalled for weeks as teams start to have meetings as to who needs to do what with the data that is coming in.
This all can be easy but you need to have gone through the learning curve already and then it becomes extremely easy to place any of these devices direct into your system.
It is for this reason that we at IoTdc have created NOA Data Service.
It has been said that very much like the law of the conservation of energy that there is a similar law which describes complexity.
Complexity does not go away and there is absolutely nothing that one can do do to eliminate the complexity of IoT, however one can move the complexity to someone elses purview and then it seems as though things are not as complex anymore. It is however deceptive as the compexity is still there in all its glory, however it has now just moved to someone else and they will end up hiding most of the complexity from you and your team so that you dont have to know everything about a certain topic or field.
When using NDS - NOA Data Service this complexity has been moved into our software and system and you dont need to know anything about most of these topics discussed. We will set up your devices in an LNS - LoRa Network Server, we will strip the data into meaningful chunks which your system can consume with ease. This takes the timeframe of setting up a POC - Proof of Concept or POV - Proof of Value from months and weeks down to hours and minutes.
Written by johnkweber - Technical team lead at iotdc.co.za
