Subscriptions, collaborators, content ... and a book!

As mentioned before, I've been looking for ways to make things sustainable for this site. With that, I am happy to make a few announcements. In short:

1. Calmcode now consists of two collaborators.
2. We've added payments and accounts to the site.
3. We're going to make a bunch more content again and even some maybe even some learning/dev-tools.
4. Calmcode now has a YouTube channel
5. We've also started writing a book, which you can start reading today.

Having two collaborators

As the site matured over the years, I've personally reached a point where it would simply just be "way more fun" to do this project with a collaborator. It's nice to be able to have somebody to bounce ideas with, especially if that collaborator has different specialities.

At the moment the collaborator in question prefers to remain anonymous, but it's been tremendously helpful given that in the past few weeks we've been:

• adding new courses
• migrating the site to Django with a proper PostGRES setup
• replacing the tinyletter email service with an internal tool
• adding accounts/payments to the site
• fixing bugs along the way

It's a bunch of work, especially when you're only able to spend a few hours a week. So I'm very happy with the both of us working together.

Payments and accounts

As I was looking for a way to make this site substain itself I originally hoped to attract some sponsors that might fit the project. While my recent employers have been super supportive, the main replies we've attracted for the ad slots came from online gambling affiliates. That didn't feel like a great fit for what this site is all about.

We're still keeping the current ads even if they don't bring in the revenue, but we figured that it was time to add subscriptions. Not to worry, we're not going to paywall the whole thing, only a small part of it. It's been super exciting to hear stories of folks being able to learn from our project and the free nature of the site was a large part of enabling this. Going forward, and we think it's a nice balance, the goal of the site is to become "mainly free, just not 100%". That means that if we have a long course, with 8 lessons, that maybe one or two lessons of said course are for members only. This should leave ample opportunities for folks to support the project without pressuring anyone.

We're eager to have most of our content be free, and only handful of the 90 courses on the site will have paywalled lessons today. That said, after maintaining the site for free for 5 years, it also feels fair to ask for a small stipend to help it move forward. I want this site to remain a rewarding hobby project and I am hoping that a small regular stipend can help us find folks worth collaborating with. The memberships will also start at a very fair early bird price to reward folks who are eager to join today.

The long term dream is to be able to have, say, a Rust person join the project for a few months on some Python friendly beginner lessons. I can't expect folks to do this for free in their spare time and I also want to be able to reward people for their effort.

Non goals

With the advent of a subscriptions I have often gotten the question if I would like to do calmcode full time. The answer is: nah. There are many reasons for this, but the main one is that it would be detrimental to the project to work full time on it. We don't want calmcode to ever become a content farm. Calmcode should be a place where we celebrate the calm nerd, not the influencers.

The benefit of having an actual job where you actually have to write production code once in a while is that you are also keeping one leg in reality. We can talk about all sorts of tools on our platform, even the super new ones, but maybe the site can remain a lot more calm if we only have to worry about the tools that we actually enjoy using ourselves. If all we did was make content all day, there is a risk that we optimise for attention, which inevitably defeats the purpose of calling your site "calm".

The long term goal is to not spend more than 4 hours a week on this project while still providing the occasional course on a tool that deserves to get more attention. If we're able to do just that, we hope that'll be a boon to a lot of developers out there.

More content and tools

We're totally planning to make content going forward, but we might alter our focus a bit. When I started the project, the world was different and my view of what might be helpful content has also changed. Originally I thought that the most helpful thing to do was to focus on syntax of some useful tools such that people could quickly get the ball rolling. Some of that is certainly useful, but most of the feedback that I've received doesn't mention it at all.

It seems people mainly appreciate that:

1. Calmcode tries to be opinionated in what tools matter.
2. Calmcode doesn't just do syntax, but also on how to think about how a tool fits in the bigger picture.
3. Calmcode tries to save you time by keeping this short and to the point.

Given this feedback, and with the advent of LLMs in mind, we figured that while syntax is useful we should probably not focus on it in our courses. Syntax needs to happen, but it doesn't help explain why a tool is so interesting or where it might fit in a large project. So we are going to try and emphesize that some more going forward. There is still going to be some syntax in the mix, without it we cannot talk about code, but we also want to be able to take steps back and talk about bigger pictures.

That said, we have also been thinking about effective ways to learn syntax. While videos are a pretty neat medium for a lot of things, if somebody wants to learn the syntax of a new tool then there may be better tools at our disposal that we might integrate on the site. We can't announce anything just yet, but we do want to hint on a new feature for the site that will deal with learning syntax. That way, the videos can talk more about the ideas behind a tool, which often deserve plenty of attention.

While working on said feature we're also going to try and bring back a regular content schedule. We're happy to announce two new courses, one on ibis and another on tailwind.css. More is on the way!

About YouTube

Calmcode now also has a YouTube channel, and it feels like I should explain how we got here. YouTube hasn't been the platform of choice for the simple reason that it's too distracting for a proper course. When you just got into focus mode, YouTube can really distract you as it recommends you other things to watch. Then again, there is some content that might YouTube very well, if only as a platform that helps spread the message.

So how did the YouTube thing happen? Well ... I really hoped to have more time to work on calmcode, but unfortunately RSI got the best of me this year. This led me down the path of ergonomic keyboards which led me to start a small YouTube channel where I discuss the pros/cons of programmable keyboards. Some videos are more like a review while others are more educational and the channel has been fun to maintain.

Besides keyboards I've also started posting the occasional progress on some personal projects. Here is one that involves improving a neural search engine by leveraging the semi supervised submodule of scikit-learn. It's this kind of hobby project over time that we would like to eventually develop into a course on calmcode as well. The YouTube channel can be a nice place to post about progress and when a milestone is reached it can be turned into a course where we discuss design decisions.

You can subscribe to the YouTube channel if you like, but any serious announcements will mainly be made through our mailing list.

A book?

And that brings me to the last effort worth mentioning here: calmcode will host a book titled "data science fiction". If you're eager to dive in, you can check it out over here. The book represents a collection of essays that each describe how the expectations of a data science project might clash with reality. There will be a lot of anekdotes, some cautionary tales, stories that actually happened but also some fiction.

Eventually we may ship an actual physical version of the book, but we're in no rush. The goal is to end up with a collection of stories that will help the field reflect on its practice and we really hope people will read this and ask better questions. Things like:

"Will this actually work, or is this data science fiction?"

The book is only just getting started. It will be written calmly, chapter by chapter, and mostly in public. We are also going to reward the calmcode members with an early preview when we have a new chapter and allow them to give feedback before sharing it with everyone else.

The future

We're pretty excited about all of this, and we hope you are too. The site will remain a calm place with short and simple videos to learn about tools, thoughts and techniques that might make your day to day a bit nicer.

We hope some of you will help us by becoming a member, but no worries if you prefer to just explore the project for the time being. There is no rush.

ps.

If you want to be kept in the loop, remember that all of our announcements are made via email. You can sign up for that here. If you have any questions or feedback about these announcements then we gladly discuss them on our feedback repository on Github.