> Moreover, there are far more powerful ways to interact with code than at the text buffer level.
This is actually one of my core reasons for using Vim. I agree, but all of those "more powerful ways to interact with code" happen in my head, not in an IDE. And my head is far more flexible than the IDE. In pretty much every IDE I've used you start thinking according to the patterns of transformation that the IDE supports, which can be helpful for a beginner, but is ultimately very limiting. New languages start with no IDE support. This is mostly for practical reasons I concede, but it is usually also often the case that the new language is being written because it represents a new paradigm of thinking. It's not until the language matures and the common patterns of thought become known that they then become encoded as "helpers" in IDEs.
The other main objection to this, I think, is to point out that linear textual encoding of code is itself a bottleneck, and that a strict text editor like Vim will never let you move past that. Implicit in this assumption, however, is the belief that there is a better way to represent code than linear text. However, I would point out that throughout all of mankind's history, language (and text as it's persistent form) has been the highest form of communicated logical representation. We have had access to pictography and other forms of communication for millennia, but none has had the same ability to encode and transfer knowledge and logic like text. I think we are primarily adapted as humans to think in terms of language. My bet is that there isn't a more powerful abstraction than this given the structure of our brains. I'm happy to be proved wrong though.
Given both of the above, personally I find the most productive means of working is to work primarily at the conceptual/abstract level in my head, model the problem and solution in thought, then translate the solution into the most natural form for expressing thought: language (as text).
So most of the tools that the IDE gives me (especially regarding code transformation, etc.) are hard-coded, less powerful versions of the kind of mental processes I am already doing. I don't need those. What I do need is a way to encode my thoughts in text as quickly and efficiently as possible. I don't want to have to think about that at all. I want it to flow naturally from my mental model into prose (though not English prose, obviously). For that an IDE again gets in the way. It's slow and full of distractions. The Vim keybindings have become so ingrained that it's muscle-memory. And I've yet to find another paradigm that lets me manipulate text as effortlessly.
I think you are right, but: how about tools like Excel, which is very non-linear, and yet it seems to be easier to pick up, and less intimidating, for complete newcomers compared to traditional programming?
Excel is, to me, an example of an alternative means of arranging logic that is very well suited to a specific subset of problems. You'll notice that people aren't typically building general-purpose applications with Excel. It's clearly not a replacement for a general-purpose language. But it is also clearly a more "natural" fit for it's niche of problems.
Actually, I think that this line of thinking holds a lot of promise: building specific tools well-tailored to solving specific types of problems. I think we could easily find models and alternate modes of thinking that outperform text for given domains.
I guess my argument is that as humans language is our most fluent medium for communicating abstractions. In the general case I don't think it will be outperformed.
Sure, it might not be general purpose, but the "niche" is really quite large: there aren't too many websites written in Excel, but there aren't many written in C++ either :) I am really not a fan of Excel overuse, but it happens, so I think there must be something about it that makes its programming model attractive to people.
For one, constructs like loops that we have mostly internalized as natural, can look a bit intimidating to non-programmers (they are not found in regular language either though!). Even among fairly experienced coders, for example in statistical programming, people who come from general-purpose coding background sometimes find loops easier to read and write compared to "vectorized" code, and vice versa.
Thinking about it, the way say much Java code is written is not what I would call a linear language either, although that I suspect might be more a result of path dependence and optimizing for "large teams of replaceable programmers".
This is actually one of my core reasons for using Vim. I agree, but all of those "more powerful ways to interact with code" happen in my head, not in an IDE. And my head is far more flexible than the IDE. In pretty much every IDE I've used you start thinking according to the patterns of transformation that the IDE supports, which can be helpful for a beginner, but is ultimately very limiting. New languages start with no IDE support. This is mostly for practical reasons I concede, but it is usually also often the case that the new language is being written because it represents a new paradigm of thinking. It's not until the language matures and the common patterns of thought become known that they then become encoded as "helpers" in IDEs.
The other main objection to this, I think, is to point out that linear textual encoding of code is itself a bottleneck, and that a strict text editor like Vim will never let you move past that. Implicit in this assumption, however, is the belief that there is a better way to represent code than linear text. However, I would point out that throughout all of mankind's history, language (and text as it's persistent form) has been the highest form of communicated logical representation. We have had access to pictography and other forms of communication for millennia, but none has had the same ability to encode and transfer knowledge and logic like text. I think we are primarily adapted as humans to think in terms of language. My bet is that there isn't a more powerful abstraction than this given the structure of our brains. I'm happy to be proved wrong though.
Given both of the above, personally I find the most productive means of working is to work primarily at the conceptual/abstract level in my head, model the problem and solution in thought, then translate the solution into the most natural form for expressing thought: language (as text).
So most of the tools that the IDE gives me (especially regarding code transformation, etc.) are hard-coded, less powerful versions of the kind of mental processes I am already doing. I don't need those. What I do need is a way to encode my thoughts in text as quickly and efficiently as possible. I don't want to have to think about that at all. I want it to flow naturally from my mental model into prose (though not English prose, obviously). For that an IDE again gets in the way. It's slow and full of distractions. The Vim keybindings have become so ingrained that it's muscle-memory. And I've yet to find another paradigm that lets me manipulate text as effortlessly.