We’ve been building Merrilin for a little under four months now, and we still do not know where it leads. I’m building it with Abinav, who writes at Lost Stoic. We started on Feb 7, and by day 90 there was already a post I meant to write and could not, because we were too busy building.
That is the strangest part of building something in the blind: you can love the work, believe in it deeply, and still have no light at the end of the tunnel. No proof that anyone will care enough, no proof that it will make money, no proof that the obsession is wise.
For me, it began a few years ago, when I told myself that the Kindle wasn’t good enough. I tried Moon Reader and it felt like something was missing. I tried Readera and KOReader after that, and that feeling never went away. At some point, my dissatisfaction stopped being a complaint and became a problem we could no longer ignore.
We have had a laundry list of features we wanted in an app that’s meant to empower someone who reads a lot. I read at nearly 800 WPM, perhaps faster if I’m in the zone. I do best when I’m not counting my reading really. I have a hate relationship with Goodreads. I want to move out of it.
We wanted EPUBs and PDFs in the same app without one feeling like an afterthought. We wanted progress, highlights, and bookmarks to sync across devices. We wanted offline reading and a version of the app that did not assume the cloud, or even a login. We wanted to ask questions of a book without being spoiled for what came later. We wanted code snippets and technical books to feel usable. None of these wants felt outrageous to us, which is perhaps why the absence of them bothered us for so long.
When we wrote the prototype, it did not look like the thing I had in my head. It was rough, unfinished, and obviously far from what we wanted. But we saw glimmers of what this could be, and that was enough to make us a little awestruck. Ira Glass once spoke about the gap between your taste and your ability, and I think that is part of what keeps you going as a beginner: your taste runs ahead of your abilities, but every now and then you catch sight of the thing you are trying to make. And that was the last day I used KOReader. I have not used it once since the first Android build of Merrilin was ready. I’d rather not read than read on another app now.
I have written before about how hard dogfooding can be when the thing you are building is still rough around the edges. It asks something of you. It slows you down. It irritates you in fresh and inventive ways. But it has been getting better. This week, I am reading The Bourne Identity on it, and Chip Huyen’s AI Engineering as well. I struggled to read Ludlum as a child; I was more into John Grisham back then, and I have started The Bourne Identity at least five times in the years since. I wanted to use this chance to finally read it. Perhaps one day, I’ll best the Malazan series too.
That means more to me than any roadmap ever could.
Feature creep is a thing though, and we don’t want to turn Merrilin into Goodreads for now. We want to build the best reader app you can find, so much so that I bought the cheapest iPad I could find to do this. I don’t enjoy using an iPad. I can’t explain how much I prefer Android.
This is a labour of love. If you install Merrilin in the upcoming days and see a bug, know that I know about it and it annoys me more than I can ever explain. We don’t want to take your money if there are bugs that annoy us. We want this to be the best damned reader app you will ever use before we even think of asking people to pay for it.
Right now though, I don’t even know if this will make us any money. I am anxious about whether it will ever pay my rent, whether it will help keep my family afloat. Honestly, I would be happy just to break even on the cloud and AI costs. I think twice before paying the cloud bill, before assessing infra costs. I self-host our GitHub runner to save money. I do not even know what it means to approach a VC with something like this, or to say that we need sponsorships.
We’d still do it anyway, since this is a problem that has possessed us. It is the great problem we have chosen to tackle.
We love this app because we see what it can be. We aren’t setting out to make a Kindle competitor, we are making something that didn’t exist before. We wanted it to be an app to read on. You don’t even need to pay us or sign in, in fact. The app will sync your files between devices without a hiccup. We want to provide the best damned experience for syncing a library of files across devices without relying on Google Drive or Apple Cloud. Just over your local WiFi. That kind of thing matters to us because reading software should get out of the way, not make you negotiate with an ecosystem before you can open your own books.
And themes? We were tired of book apps playing second fiddle to coding IDEs. We have so many themes in VS Code that no other ebook app has. Which is why we decided to bundle all of them in.
Ask me why I use the Red theme most of the time, I kid you not. I even use Merrilin when I am reading physical books, although that means buying the book twice: once as a physical copy, and once from ebooks.com so I can get the DRM free file. That is how I am reading The Emperor of All Maladies right now.
I have seen people tweet that they no longer fully read books, and instead just ask AI what a book contains. That is not what this is for me. This came from a need to read deeper, to understand books I would otherwise have struggled with. I do not know enough biology to read The Emperor of All Maladies effortlessly. And when I first prototyped this idea with ChatGPT in 2024, one of the things I kept asking about War and Peace was why the Russian aristocrats were speaking French.
I’ve said a lot but I haven’t said why we’re building this without funding, and without users paying us. To paraphrase Richard Hamming, this is the toughest problem we’ve chosen to solve. It’s our problem to solve, something we are deeply well-equipped to do. Both of us read, and have read for our entire lives, but I was always the more opinionated one when it came to reading apps. I could tell that reading apps were not written by people who used them as much as I did. That may be the simplest explanation for all of this: we always built Merrilin for ourselves first. I wanted to forge my own path, my own sword.
Maybe that is what this really is: building in the blind. You work because the work feels necessary. You keep going because you can see the thing in your head, even when you cannot see the market, the money, or the finish line.
We have seen enough in brief glimmers to know we cannot leave this problem to someone else.
It is and always will be a labour of love, although there is every chance this labour may yet be lost.