In theory, this is a great idea. In practice, it's been so horribly implemented as to be nearly worthless. When you try to use Notes development for web applications, you discover that 90% of the functionality available in thick client applications are unavailable to you, and that it's very difficult to work around these limitations.
I just spent 3 hours trying to get a function working, before finally finding a small footnote buried in the Notes documentation telling me that something I used in my function is unavailable to web clients. Making the whole thing, and everything I did in the past 3 hours, a waste of time.
The more I work with Lotus Notes, the more I realize that there are some truly great ideas here. And, the more I realize that they are implemented so poorly as to be worse than worthless. I keep getting the feeling that someone came up with an idea, then had someone else try to implement it, without fully understanding it. Only, it doesn't seem like just one person implemented it, it seems like 10 different people tried to implement it at once, all without ever talking to each other.
There are so many inconsistencies, bizarre things that work one way in one place, and a different elsewhere, or not at all somewhere else, that it's a royal pain to understand. You can't help but wonder if the program had a dozen different development managers, each with a very different vision for the program, and each trying to make his vision happen at the same time.
If it were rewritten from scratch, and implemented correctly, I think Lotus Notes has the potential to be one of the most powerful, useful, and impressive tools for application development I've seen. Unfortunately, I don't think it can realize that potential without a complete, or nearly so, rewrite. There's just too many fundamental problems and design flaws in it.