Summary

Wrote a lot of Mudmap’s new frontend, integrated it with the backend, and published another blog post for the site. I did not get all that I wanted to be done within the month, though.

Made the transition from using Stripe to Paddle after a wonderful exchange of Tweets which led to a phone call with Paddle. They were incredibly helpful and offered some points of refinement for the business. It cemented my commitment to Paddle as I was treated like a valuable addition to their cause, instead of a faceless and numbered money-generating object.

Highlights

• Integrated the new dashboard
• Published a post and sent out a newsletter for Mudmap
• Finished the MVP feature-set for Mudmap minus the payment system

Goal Performance

1. Deploy Mudmap’s application for beta testing

• Appraisal: Did not complete all the tasks necessary to do this 😞
• Rating: D

Why a D if I failed to launch it? Seems pretty binary.

I accomplished a lot during this month and although the goal wasn’t completed, I am reluctant to call it a failure. Having a time box around this really helped to keep me focused, made it easy to cut distractions, and in general, pushed me hard to achieve it. That is not a failure in my eyes.

So what did I get done?

• Added structlog for better logging management (also have Sentry)
• Improved the SSH Client error and timeout handling
• Reworked the device registration flow in the front and back end
• Built the React frontend sign in/out, registration, etc from scratch
• Integrated dj-rest-auth after wasting time rolling my own cookie auth
  • This took a lot of time out of the month but it gave me so much for free that it was worth it
• Replaced fetch with axios on the frontend to make JWT’s so much easier to manage
• Began the Paddle integration

More importantly, what didn’t I get done that prevented this from being a success (i.e. launching the MVP on time):

• Payment integration
• Underlying API tweaks (namely around the installer redundancy)
• Deployment testing (I see some initial difficulty getting Traefik and netlify to play nice)

Will I knock it out this month? I really do hope so!

Demonstrating functionality

Here is a short clip of the dashboard’s device detail in action. It’s a bit neater now, but the rough functionality is the same. It will at a glance give you up-to-date information about the pfSense system, as well as let you start, stop or restart system services.

2. Write a blog post about pfSense

• Appraisal: Done!
• Rating: A

What is pfSense?

A pretty easy goal but important. If I did not set this, I would not have done any content marketing for Mudmap because I spend all my time down in weeds writing code instead.

I like writing, but it is definitely not something that I am gifted at. It is very humbling when you think you know a subject but struggle to write about it in a coherent and concise manner. Though, that is why you have to keep producing and perfecting it as a craft.

3. Increase Mudmap’s marketing

• Appraisal: Average to poor - didn’t prioritise this as I should have
• Rating: C

I wrote a blog post, publicised that on Twitter, and pushed out a newsletter to my audience of zero! Can only go up from here. Now as I reflect, I see that focusing so hard on knocking out code, so I can launch the MVP is also a bit of a trap. The more marketing I do, the more awareness, the more feedback and, the more validation I will get.

On the topic of emails, I did find a great email template generator; mjml. I have not used it for any of my emails just yet, but it looks brilliant. Trying to style an email (in my opinion) is a complete tyre-fire.

I am, however, working on my Twitter game. It’s been a lot of fun so far. Trying to write short 280 character messages that convey meaning is a lot harder than I originally thought. Having to word things carefully has been great for my communication skills - especially as our society’s attention spans continue to rapidly decrease. It serves as my primary marketing platform for Mudmap at the moment as well - something that will need to be adjusted as time goes on. This makes it an important skill to refine and work on which is why I am investing in it.

Recommendations

I listened to Lex Friedman’s podcast; 176 Robert Breedlove. Even if you aren’t into bitcoin, it is a wonderful podcast with some interesting views on the current state of the financial world. I highly recommend it, if you’re only slightly interested in finance.

Not a recommendation. A sad note. I am getting older! I am now working with a lot of people that have never seen Home Alone or Jurassic Park. Thankfully, the guy next to me is my age and we can reminisce about the glory days of movies 🤣

I’ve nearly finished The Three Body Problem. It’s a great book, even if most of the physics is above my pay grade (it is still understandable). I am already looking forward to the next book in the trilogy. I usually read non-fiction but have opted for fiction before bed these days as I need to unwind.

This month I learned of litestream. I think it’s a project that has real potential - not every application needs Postgres or some other server-based SQL solution. For a great introduction, see this Tweet.

One of the best hidden gems in app development is @litestreamio. It streams SQLite databases to cloud storage, giving you easy data replication without standing up a whole database cluster. I used it to build a log uploading service and it's been fantastic https://t.co/4NGDeMbHzV

— Michael Lynch (@deliberatecoder) April 29, 2021

Wrap up

What can I do better?

• Marketing for Mudmap,
• Documentation - I am going to have a documentation fest once Mudmap launches, but it would be much easier to have done parts of it as I developed instead.

What have I done well?

• I’ve knuckled down and accomplished a lot this month. So much that it wasn’t until I started this retro and looked back that I realised how much I have done,
• Went from having never used React to implementing a fully functional application with authentication.

Next months goals

• Launch Mudmap
• Start the documentation pages
• Replace the pricing page, and if possible, embed it in Hugo as a React component

Analytics