[Michael Lynch] has been a solo developer for over three years now, and has been carefully cataloguing his attempts at generating revenue for himself ever since making the jump to being self-employed. Success is not just hard work; it is partly knowing when the pull the plug on an idea, and [Micheal] has been very open about his adventures in this area. He shares the good news about a DIY project of his that ended up becoming a successful product, complete with dollar amounts and frank observations.
About a year ago, we covered a project he shared called TinyPilot, which is an effective KVM-over-IP device, accessible over the web, that could be built with about $100 worth of parts. [Micheal] found it to be a fun and useful project, and decided to see if he could sell kits. However, he admits he didn’t have high expectations, and his thoughts are probably pretty familiar to most hardware types:
I questioned whether there was a market for this. Why would anyone buy this device from me? It was just a collection of widely available hardware components.
Well, it turns out that he was onto something, and the demand for his device became immediately clear. He’s since given TinyPilot more features, an attractive case, and even provides a support plan for commercial customers. This is an excellent reminder that sometimes, what is being sold isn’t the collection of parts itself. Sometimes, what’s being sold is a solution to a problem people have, and those people are time-poor and willing to pay for something that just works.
It’s great to see [Michael] find some success as a solo developer, but his yearly wrap-up covers much more than just the success of TinyPilot as a product, so be sure to check it out if you’re at all interested in the journey of working for yourself.
That’s awesome that he cataloged his successes and failures. I suppose we could all do a better job of that!
Whoa!
He did all that without crowd funding?
B^)
Very cool. Glad to see his success to inspire us all.
…Just keep swimming
The reason it became popular is because he priced it at $100 and the competition is priced at “If you have to ask, you can’t afford it.”. A single port KVM switch costs about $1000.
Another similar product is a VPN box that you can plug into any network that offers internet access, to get you into the LAN or to a second network interface from anywhere in the world.
The closest commercial alternatives cost about $2000. You could build one out of a Raspberry Pi and OpenVPN – the reason why people don’t is because they don’t know how and it’s way too much work to build one from scratch. You could easily sell it for $200.
It’s a bit more that $100. https://tinypilotkvm.com/product/tinypilot-voyager (looks like the kit version is discontinued).
But it is still a lot cheaper than something from say Avocent.
Between that and the PiKVM it’s a good time to be someone who is looking for enterprise management features but doesn’t have the enterprise budget not works at enterprise scale.
If you find yourself in a position where the demand for your work outstrips the supply, the natural reaction is to increase the price until the two meet.
“The reason it became popular is because he priced it at $100”
Suggestion:
The reason it became popular is because he originally priced it at $100
The quote seems to imply the current price is $100.
Michael Lynch is free to set his pricing to whatever the market with bear.
Just wanted to point out that in the Raspberry Pi-based KVM space, Pi-KVM is the other contender worth looking into. It has a lot more features, and the kit (crowdfunded) is a bit cheaper, though it doesn’t have the software/support options TinyPilot has. So they’re two different products for two slightly different markets, both a lot cheaper than the proprietary incumbents.
But those who are price-conscious or need features like ATX control would probably favor the Pi-KVM.
I love this but $350 ? PIKVM is completely free… Just saying.
Yeah if you find that your product is in very high demand then you put the price up. Started as a kit and ended up as a commercial product, just how it goes.
I love seeing all the custom builds on here by solo devs, showing how to build stuff with freely available parts.
…. but I’d say 90% of the time, I’d happily pay the guy to provide it already built :)
No FCC testing :( Was hoping to see timelines and costs for someone doing that the first time.
If it’s all made out of modules that are already FCC approved, then no further testing should be necessary.
I believe the entire unit needs to be tested for unintentional radiation with all the individual modules connected.
I don’t think so, hopefully not. My basic understanding of the engineering indicates there is no point.
Components don’t need to be FCC tested – but, final products do.
Thanks for pointing this out! Good read.
I have had done something similar to using VNC, an old notebook, and arduino. The screen capture is done via webcam. Lame, but works. Arduino simulates ps/2 keyboard and ON-OFF/RESET switch. No mouse support (who needs mouse anyway?) It works for some years now and it was basically free.
It sounds good enough for me. Is there anything like a step by step guide?
Thanks for the writeup! It’s been a fun journey for me.
Hackaday’s coverage when I published the first version was a big help in getting the project off the ground, so I very much appreciate you following along with updates.
hey michael, im looking to make some video streaming products for example law enforcement streaming camera. and i wanted to ask you. what are you using for streaming video? is it webrtc?
I use uStreamer for the streaming, which supports MJPEG and H264. I currently stick with the MJPEG option because it’s really simple for browsers to consume, but it’s fairly bandwidth heavy because it’s essentially sending every frame of video as its own JPEG.
While I like the product, I’m pretty sure Tiny Pilot is based on Pi-KVM and uses some of the code from that project. That makes the project not necessarily solo. What this guy did is amazing, but all parties should get credit for their great work.
Yeah, the title “solo developer” is just because I would write yearly updates about my progress working on my own after quitting my job at Google. In the next update, I’m going to call it “solo founder” because a lot of other people work with me on TinyPilot, including a team of three other developers.
I give credit to TinyPilot’s major dependencies on the website and the source repository, and I send monthly financial donations to many of them.
I’ll definitely have a close look at the details! I am trying to start a business making open source hardware, too. It’s electromechanical rather than electronic. It’s an energy recovery ventilator that goes in your window. Similar price issues, to buy one is more than a thousand bucks, but I can build one for $100 or so in parts. Hopefully that differential will be enough to pay a decent wage once all the marketing, design and development, packing and shopping, is factored in. I’m not expecting a get rich scheme, just self employment.
I do worry that other people, especially rich people, might just clone my design and undercut me. I am starting to see the logic of IP better these days, if I’m honest. But it doesn’t have much application here. My hopes lie in operating in the self employment zone; if profits to be had were too great, rich people would come in and try to steal the business. If the margins are more working class, they won’t bother. Sigh. So there is a ceiling, a glass ceiling. All I really want is to be able to afford a house, but that means about a million dollars these days. That requires a get rich scheme, which my little undertaking can’t give me. So I wonder if it’s really the best idea sometimes.
But it seems fairly reliable, just a sounder bet than the other ideas I have. I am hoping I can get it started and then become a part time manager of the whole thing, filling the more profitable role, and paying other people fair wages to do things like churn out the parts etc.