Well, I think this sounds quite good, in fact it sounds so good that it is probably not true. And reading to the page, I stumble over the sentence<p>> An IDE like Medium, not Vim<p>Thing is, vim is not complicated because of some misplaced elitism, vim is complicated because it helps with complicated problems. (I am a zealot for the church of Emacs, when I say something "nice" about vim, it is because honesty or pointy objects force me.)<p>Unfortunately it goes downhill from there, it is nice that there is a way to visualize memory in five lines of code, but what about the memory of the following three processes (and the one thread in red and the others in some other color?) In my experience, there is a trade off between the power of a language and the impressiveness of the examples. (A better example would probably be the twitter api, does that mean I need to beg the core developers to talk to the next "slack for dogs" api?)<p>Then there is the "zoom" feature, I have not the slightest idea what that is supposed to mean. (Actually I have, I just don't think that this side of a strong AI coworker it is possible to hide information in any useful way. (Hiding information is just the inverse of zoom.)<p>And to top it of, there is a link to the demo and the first thing I notice is, scrolling is broken. (Arch, Firefox, NoScript with in this case all JS sources allowed) Well, the first thing a IDE should allow me to do is to display text in the most reliable way possible, even if that means more than one page. (Plus it is in a browser, which may or may not be only for demonstration purposes, but at the very least the browser uses the right mouse button to display browsery options not IDE options.)<p>After having fired up chromium, there the IDE works and it actually works kind of well. To keep with the overall negative tone of the comment, almost as good as Jupyter. The first thing I notice, is that one can switch code blocks on and off, but the output depends non linearly on the set of switched on code blocks. For this one would likely need some kind of toggles for sets of code blocks if it should be usable at all. (So to switch on extra test cases, or to test only one part of the program, etc.)<p>In conclusion, it is a nice project and I wish the devs all the best. Hopefully one of these days someone manages to get literate programming to work. But I think that there are quite a few things which make me doupt that this is, what makes literate programming work in the end.