Understanding IoT Market Opportunities

Career
1

Good Morning. I’m new to this community and am enjoying reading & listening to all of the great content. I have a couple of questions, but I thought it’d be useful to provide some brief info on my career background for perspective first…

I’ve known @cbrake for a long time and he was instrumental in helping me find my way early on and even let me help him for a couple summers while I was still in high school (which was much more for my benefit than his…). I’ve been in the IT consulting world for the last 12 years in a variety of roles. I currently have a strong grasp of routing, switching, server administration, development (particularly .NET, systems integrations, and API development), and have worked extensively in AWS and Azure clouds. I’m very familiar in working with distributed systems and regularly assist a number of clients with work in this arena. On primarily personal endeavors I’ve spent a fair bit of time working with others on designing electrical & control systems, process piping, and have developed & played with all the usual gadgets such as raspberry pis, beaglebones, and arduinos. I also have an MBA and have worked extensively with clients in analyzing, documenting, and optimizing processes.

One of the things that I haven’t yet done is make IoT a professional focus. As I continue to grow and learn I keep coming back to it as my answer to the age-old existential question: If I could do anything what would I do… I’m especially interested in it when I see how the market has seemingly evolved beyond strictly the embedded Linux work to things like Go, SimpleIoT, easier & cheaper hardware support, cheaper cellular connectivity options, etc. Theoretically, it seems like this world should be full of opportunities to serve & help businesses leverage technologies to maximize their business processes & pursuits…

For those who work in this field regularly I would be interested to hear your perspective on the IoT market demand (positive or negative), how to go about finding customers, and any sectors that may be particularly worth focusing on. Additionally, I’m interested in any suggested learning opportunities (beyond this community) and additional certifications that I might consider. I’ve recently started working on learning Go as it seems like it’s providing a lot of value in this space and thankfully it’s fairly in line with other development languages I’ve used. Thanks in advance for any feedback you might have.

1 Like
2

Hi Zach, great to hear from you again – it’s been a long time! Welcome to the TMPDIR community.

These are great devices and have done a lot to move technology forward and make it accessible.

Although I am still focused on the Linux at the edge, there are a lot of neat technologies coming in the MCU space – ESP32, Particle.io, Zephyr, …

This is a good question – I don’t go through a lot of customers, so I’m not sure exactly where or how to find customers. I started doing embedded Linux, but most people want remote access to their data, so I naturally moved into IoT. To me, it is a huge field – way more opportunities than people to do the work. I’m most interested in solutions with very fast payback ( < 1year). Ex: a system that eliminates the need for a person to check a tank level once a day. Good place to start is problems around you – people you know. Ask, “what would you like to monitor/control remotely?”.

The biggest problem with IoT and technology, in general, is complexity. Most IoT solutions are overly complex and have too many moving pieces. Successful solutions need to simplify users’ and developers’ lives, not make them more complicated. Even home assistant, which is very well done, is way too complex – especially once you want to go distributed, and scale to dozens of sites.

I feel we need to drastically rethink things and invent something that does not currently exist – maybe get out a blank sheet of paper, draw up your ideal solution, and go from there. :slight_smile:

Yes, Go is amazing. It drastically simplifies developing and deploying these systems.

Lots of opportunities to exercise Go here:

SIOT is a generic framework, so the idea is you can add clients to interface with whatever you are interested in.

Great hearing from you again. Please feel free to share trends you see.

1 Like
3

Thanks Cliff. Always appreciate you being willing to lend a listening ear and helping hand. I’m also appreciating all of the hours you’ve spent over the last several years sharing & documenting your knowledge on your various platforms. From my perspective you’ve created almost like an online self-paced trade school. Many would be tempted to hide the things that you’ve openly shared as trade secrets… I’m hoping to schedule time each week to go through your content and the inevitable side tangents.

I think at the end of the day my situation is not actually a problem of finding customers, but rather working with existing customers to understand where IoT may be a fit for them. In particular I’d like to focus on IoT projects that include physical sensors and controls. I’m thankful for a workplace with thousands of clients where we’re deeply embedded in their business processes and leadership has enabled me to generally focus on whatever I want. It seems highly probable that there are opportunities that exist here that are simply getting missed…

It’s often far too easy to overlook opportunities, because we’re hyper-focused on the task at hand and we miss the big picture of how we might be able to maximize value. It’s also very tempting to mentally glamorize IoT to a point where we discount existing IoT projects that we’re already working on (all depends on how you define it).

I’ll work on getting a SIOT environment setup in my lab as I believe having a simple ready to go solution offering will make selling it internally & externally that much easier. There’s also a certain confidence knowing that you’re actually running this in the field. As I progress, I’ll be sure to take a look at your open backlog of issues as this is a great way to simultaneously learn & contribute.

1 Like
4

@zpertee thanks for appreciating this work. It means a lot.

1 Like
5

Well said Zach – IoT is everywhere, it all depends on how you define it. After the addition of the recent metrics feature, I’m working on deploying SIOT to all my servers to monitor them. Of course, there are dozens of other ways to do this (Prometheus, etc), but eventually, I’d like SIOT to be the easiest for the simple (non-Google-scale) use cases.

Feedback is welcome. If you fire up SIOT, let us know what you like and what you don’t like.

1 Like
6

Go might be great for cloud platform stuff, it will be too large for many embedded platforms comparing to using C or C++ or Rust or Lua.

Also Go does not do well with shared libraries on the target board either, so if you have a few Go binaries on the board without using busybox-all-in-one approach, your storage space will run out quickly. In fact Rust has similar issues. I really wanted to use Go and Rust on mid-range embedded systems(8-64MB storage, 32-128MB RAM), but I ended up with C and C++ and Lua still.

1 Like