You can still sign and notarize your .dmg without distributing it via the app store. Users who download from your website won't see the "unknown developer" warning as long as you do that.
Your second point about your target market is much more salient. Tough call.
What benefit do you hope to realize from distributing via the App Store? They aren't going to handle payments for you. They won't market your app for you.
I don't expect you'd see a difference. For the simple case of iterating on a diagram you describe to ChatGPT, 4o is perfectly fine as-is. I can't see how you'd meaningfully improve on it.
However, for the advanced case of generating a diagram from a repo, AI isn't even close. You'd need practically unthinkable improvement for it to be useable.
The consequence, unfortunately, is incremental gains won't help in either case.
Programs that produce diagrams should be (and are) written in a programming language. They accept input, execute, and produce output. Diagrams themselves do not execute and have state (unless they are interactive[0]). There's a difference.
Yeah, just about every diagram you see online commits these sins. Probably because companies that create detailed diagrams don't make them public. What's left is mostly diagrams by tech companies (e.g. cloud providers) created specifically to promote the technologies themselves.
You can do it, certainly, but it is overkill. It also limits your potential editors to people who know Python and are willing to install it on their machine. In contrast to Python (or other programming languages), a language like YAML or JSON can be learned in minutes by just about anyone.
I'm biased, but I think it's a bad decision. Programming languages are for writing, well, programs. Programs execute, accept input, have state, and all manner of other things that just don't apply to diagrams. Diagramming in a programming language is just weird.
Sometimes. In the face of entrenched competition, you can't be just a little bit better. You have to be a lot better. And in a way that said competition can't easily match (or exceed).
> To address this problem, we built decisioning models that take into account 100’s of datapoints, such as customer and payment metadata as well as internal classifiers of error codes and messages. The models determine the optimal time to retry a payment with current or backup payment methods and when to communicate with a customer.
All well and good, but doesn't Stripe have much, much more data on this? Once they start doing all of the above (assuming they don't already), their data will give them a gigantic advantage. And that's only on top of being built-in.