What is holding MPS back from being a wild success?

I recently saw this question being asked in a private chatroom. The answers were mostly concentrating on missing features, such as MPS not having a ready Web story or being too IDE-like for non-technical users.

My take is that these missing features are not rocket science (apart from collaborative editing, perhaps), so there must be a deeper reason.

Implementing the features is mostly a matter of money, time, and focus. And focus is what I think MPS has been lacking lately. The focus of MPS has long been software developers, much like other products by JetBrains. The idea was that developers would build more powerful languages for themselves. Later it turned out that this use case was not so hot, and a much better use case for MPS is for developers to build tools for technical subject matter experts.

Subject matter experts do not think of themselves as programmers but their work is formal enough that giving them tool support makes sense. However, they need different UI and UX.

So we now have not one but two categories of MPS users: the tool developers and the tool users. Unfortunately, MPS does not recognize this and maintains a “meta-circular” approach: you use MPS to develop a tool that is a modified version of MPS. This is not how most developer tools work. If you develop a web application in IntelliJ IDEA, the users of your application do not get IntelliJ IDEA bundled with it.

I don’t think it would make sense for MPS itself to break its meta-circularity. However, the new generation of MPS-on-the-Web tools that I have covered last week have abandoned this metacircularity. With all of them, the tooling you use to develop your languages is not part of the application runtime.

In fact, the tooling you as a developer would use with most of the upcoming projects is going to be JetBrains MPS for many years to come. But having a new runtime environment for MPS-based languages will hopefully breathe new life into the whole area of languages for subject matter experts.

Do you have your own take on what is holding MPS back from being a wild success? Reply to this email and let me know.