Picture of the author

Saša Šijak

Saša Šijak

“All grown-ups were once children... but only a few of them remember it.”
― Antoine de Saint-Exupéry, The Little Prince

Live man switch weekly update 1


During my career as a Software Engineer, I was always starting new side projects. But the real problem was that I was not finishing them. Either life would get in the way, or I would travel somewhere, or I have too many obligations in the salaried job. In any way after I stop working on a project for a few days or more I would lose motivation to get back into it and it will be added to the never ending pile of dead projects.

That started to bother me at some point so I decided that I wanted to change that. So to be able to do that I really wanted to focus on finishing. That would be my main goal. Launching stuff on the web and not to the graveyard. So I started the pursuit with several ideas which I found would not take more than 2-3 months to implement per idea.

At the and of November in 2020 I started working on my side project React UI Boutique. I was working on it non-stop for about two months and successfully released it in early access on the last day of January in 2021 which was my target date.

Even though it was a success and I managed to do it as planned, I felt exhausted. But I also wanted to start on my second idea and keep the momentum going. At that exact time Sven from CoStarting contacted me via Twitter and introduced me to the concept which after initial pushback because my introvert part was fighting against it I agreed so I am building the second idea through CoStarting.

Live man switch

This second idea is a service that I am calling Live man switch. It is an alternative name for Dead man's switches, making it sound not so grim. But what exactly are they? To quote Wikipedia "A dead man's switch (see alternative names) is a switch that is designed to be activated or deactivated if the human operator becomes incapacitated, such as through death, loss of consciousness, or being bodily removed from control. Originally applied to switches on a vehicle or machine, it has since come to be used to describe other intangible uses like in computer software.".

I will just implement that concept for the web. It will be a simple service and the flow will go like this:

  • User configures the frequency at which the service will check-up on him (eg. every 10 days by email/sms)
  • User configures the trigger condition (eg. no reply to the check-up message for 5 days)
  • User creates email messages or configures webhooks that will be sent if the previous trigger condition is met
  • Service checks all configurations each minute and if trigger condition is met, sends the messages

What I planned and did for the week

For the first week of working on this project I planned to setup basic things like repo, deploy process, different accounts needed (firebase, analytics, payments, error collection...). I also wanted to create the default landing page, implement supporting libraries (payments, mixpanel…), implement account creation and implement the design of the dashboard.

At the end of the week I can say that I've done all of that and more on top of it. I've implemented both credit card payments via Stripe and crypto payments via Coinbase commerce. One of the target audience for this service would be crypto holders, so I wanted to enable them to pay with their cryptocurrencies. Design is done, auth works. Other services like Twillio, Mailgun, Mixpanel, Google Tag Manager/Analytics, Firebase are integrated. I even started on the logic for building the actual triggers which was not planned for this week.

Apart from coding I was looking into best services to generate me some nice TOS and Privacy Policy documments. I think I will use iubenda. I was also writing help documment and was thinking of writing different copy for different target audiences. Event tried using GPT3 tools to generate that for me 🤪 That failed for this usecase, but on my previous project it actually worked and I used some computer generated copy on my projects website. I also spent some time reading about the taxes again, and if should I just use Paddle for payments and dont wory about taxes at all. Problem was that if I would go that route I would not be allowed to use crypto payments on the website.

Here are some screenshots of the current work.

Challenges I faced

For some reason this week was not as easy as I expected. I struggled with motivation and slight depression. It is still winter here and like a clockwork for previous several years I start to feel the same way during January and February. But spring is around the corner and my mood improves alot then. Also, being in costarting group helped me to stay on track. All the small nudges keep you in the mindset of finishing.

On the technical side I did not have much problem. Worst one was finding a way to implement crypto payments in a way that could not mess up Stripe subscriptions and vice versa. One thing that I did struggle with was with copy on the website. So for that reason I paid someone on Fiverr to write me some copy, we will see how that turns out.