Over at EINVAL, Wojciech Siewierski considers multiple cursors (in Emacs) and whether it’s suboptimal. His post is entitled Multiple cursors considered… suboptimal, which seems rather provocative but the post is actually very even handed and concludes, as you’d expect, that whether or not multiple cursors makes sense depends on the task at hand.
I like multiple cursors and am still in awe every time I watch Magnor Sveen’s astounding demo, especially the last bit where he calculates the total time spent watching his videos. That said, I always find it a bit hard to use and, like Siewierski, usually fall back to keyboard macros. That’s probably because there’s not much to remember when using keyboard macros, whereas multiple cursors has a lot of commands and several edge cases. Doubtless, I’d be better at using them if I made a point of invoking them more often.
One of the most compelling use cases of multiple cursors for me is changing every instance of a word, phrase, or even regular expression. It’s perfect for that but I find that I generally use iedit, again because there’s not much to remember.
Still, there are some cases where you just can’t beat multiple cursors. The examples in Sveen’s video provide several cases of that. I do use and like it and should probably make a point of doing so more often so that I internalize its operation.
Chris Wellons makes a compelling case that multiple cursors is not very efficient in Emacs because it doesn’t play well with the gap buffer mechanism that is at the heart of Emacs editing. I agree with that but don’t care because, at least on the machines I use, there’s no detectable delay. Wellons has other complaints about multiple cursors that you can read about at the above link if you’re interested in the controversy but they mostly concern the edge cases where multiple cursors doesn’t work very well.