@Bortels Phew! It’s not just me. I once tried to estimate the amount of copper deposited on a nail by photographing it and measuring the intensity of the colour.
It’s the converse of the “If all you’ve got is a hammer …”: if you want to bang a nail into a wall, everything you’ve got suddenly looks rather like a hammer.
@Bortels and others - I agree that you can build a wide variety of apps in Codea, and I think Codea is the most fun tool to work with.
But I don’t agree that Codea is fine for business apps. By design, it lacks a number of important features, especially UI, and even with libraries like Cider, it isn’t a natural tool for building form based apps. You are going to spend a fair chunk of time messing with the UI and other issues, when you could be working on the app itself.
When you used your ruler to weigh flour, you used the best alternative you could find, from a range of choices lying around. You didn’t restrict yourself to one tool. In contrast, saying Codea can be used for anything is like saying “I love this pencil, so I’ll use it for everything”. A pencil isn’t going to weigh flour very well, any more than Codea is going to be well suited for every type of application. In fact, your flour example just highlights the advantages of having several alternative tools to choose from.
When I choose projects in Codea, I try to program to its strengths, primarily graphical animation. Why would I waste my time building a UI when there are so many fun graphics projects to pursue? But maybe that’s just me… :-?
@Jmv38 Simenon? Programming by Inspector Maigret? 
I suppose one could use Codea for business apps. COBOL is also awesome for writing a PacMan game, too. 
Codea needs better support for data storage and manipulation, just in general. (Disclaimer: I have not played with the LocalData storage yet to see how useful and scalable it is.)
@Ignatz It’s fairly easy to write a simple UI in Codea and it’s even fun! But you are right–it’s not the big draw here.
I agree with some comments from both sides of this discussion. After a bit of thought, I think that tool suitability depends on how you try to implement the solution.
Have a look at Tydlig, which Simeon tweeted about. http://tydligapp.com/
If you reinvent the concept of display and interaction, unconstrained by rigid placements and the need for uniformity, Codea IMO would be much more suitable than many might have previously considered. It might still need some more capability for data handling, but Tydlig opens up the mind to what could be created with Codea.
I think Codea would make a wonderful MIDI control and algorithmic composition tool. So why is there so little interest - nay, pushback - on adding Core MIDI to the library? It would be really simple to add using the MidiBus library from Audeonics. Think about it. Moog’s Animoog App has been downloaded over 400,000 times already, as just one data point. Not a small market, and they (we:) are used to paying $10 - $20 per App. Lua is already under the hood of several music Apps, but without the Codea touch Apple is not letting them open it up to end users.