I’ll Launch As Soon As I’ve Fixed This Bug

Update: I wrote this as a tongue-in-cheek piece, but it did prompt me to actually launch.

I’m this close to launching my new app. I just have a couple more bugs to fix.

I know, I know, I’ve been saying that for months. Longer than I care to admit. But it’s true. It’s right around the corner!

The thing is that I haven’t solved the root cause of the latest bugs. It’s really this whole other part of the code base that needs to be tidied up.

But looking into that part of the code has exposed a whole other series of issues… This feature I added doesn’t even work! Do I still need it? I’ll at least fix it so the damn thing compiles.

What the hell? This library of functions I added isn’t even type-safe. There could be all kinds of bugs lurking under the surface!

How is it that I have two type files with overlapping types… better take care of that. Ah, much better!

Why is it taking so long to build? And what’s that weird build error? It builds, but it just makes me worried — better look into it.

Why is this function its own class? Seems overly verbose.

Why is this class split into three files? I should combine it into one.

Maybe that’s why I keep getting that warning about build size. Let me throw that into the LLM and let it take care of it while I make coffee.

Check out this new UI feature I added! I know, it’s not central to the path to profitability, but it looks so cool… I just needed something to distract myself between debugging sessions.

OK, back to it. Today I’m going to focus on this app actually making money. Better just update the libraries first — no point launching if it won’t even build on my production server… oh damn, why did the API change spec again?

… Wow, it’s because they released a whole new feature set! This could be a game-changer. Let me just incorporate that.

Just discovered a bunch of ghost code!!

Hey, my version of Node doesn’t match the one on the server!

Oh man, it’s crazy how you tell the AI to not use ‘any’ types but it does anyway… so much cleaning up to do.

I’m just trying to get the back-end secure — I took a bunch of shortcuts and it’ll come back to bite me in the arse later.

Look at this feature I added to the front end!

Sure it works, but I mean, it’s a little ugly. I’ll tidy up the UI — that’ll help me debug and improve operational performance anyway.

I managed to get the build time down to 2 seconds!! And the whole package is under 500 kb compressed!

Wait… sure, it’s looking pretty, but the core logic underneath it needs some work. That will get 100% of my attention.

OK, look, I’ve got to launch soon. I’ll set myself a goal — I’ll launch as soon as I have five days straight of normal operation.

Three days.

Those three days might have been idiosyncratic… besides, I got a weird bug in execution; better fix that first.

Anyway, it can’t be this week because it’s already Wednesday. And besides, the whole thing is still riddled with security issues.

Maybe I’ll just fix this UI issue — I can look at the core engine tomorrow.

Still, getting closer. Launch date is around the corner. ????

Ed: These are all toxic things I’ve realised I say to myself, thanks to a friend. Also, if you liked this, you might like this article: That’ll Never Work, and 100 Other Ways People Will Poo-Poo Your Ideas.