That's very true, but for me that's an acceptable tradeoff for not having to write much code or spin up a database, docker container, load balancer, or worry about security. Dark is great for getting a project off the ground and getting to the "proof of concept" stage: for me, there is tremendous value in getting there sooner rather than wasting a bunch of time doing things that are tangential to creating the product.

True, it's great for prototypes, but remember that their incentive is for you to remain on the platform as your project gets bigger.

Therefore you're likely to have extreme difficulty in moving off the platform as your system grows. Which means either a complete rebuild off the platform, or the easier option... Stay on the platform.

So the ultimate options are; either choose open-source flexible solutions, or remain shackled to the proprietary platforms...

That's an interesting point-of-view. I was also worried and had to search in their Slack channel. This is what they have to say:

"so specifically to lock-in: right now, no one concerned about lock-in is going to have a good experience on dark. We don’t have the resources to address that need, but at some point we will be in a company position to deal with the lock-in question. Our plan is to give you tools and resources to move off Dark if you want to"

in b4 all of HN crying posts about dead propriatory software and blogs of how to hack and partially import stuff to "new replacement startup" that is also locked in. When will people finally learn that locked in languages are a very costly mistake.