Scott Rosenberg's 2006 post on Outliners then and now is still interesting. Just like Rosenberg, I have never understood why (single-pane) outliners never caught on "as a primary writing environment." After all, they do "allow[.] you to dump huge amounts of information into the outline efficiently, move big pieces around easily, and swoop quickly from a top-level overview to the finer details."
Nor have I ever understoo why outliners are considered harmful. The claim that "when you use a tool that encourages you to think in terms of hierarchies, everything looks like a hierarchy" is clearly an exaggeration. A tool that encourages you to think in terms of hierarchies may indeed encourage you to look at everything in terms of hierarchy, but it does not force you to do so. It's ultimately up to the user of outliners to determine whether to look at the world hierarchically or non-hierarchically. Furthermore, an outliner is a tool to structure the information one wants to get across to others in the most plausible way, and claims become arguments in virtue of their structure. This has nothing to do with viewing the world exclusively in hierarchical ways. As Rosenberg points out in Outliners, Trees, and Meshes , "what matters [in outlining] is that outlines give ... easy handles to move chunks of loosely structured information around, and they let [one] quickly zoom from a low-altitude view to a high-altitude overview and back." This is why he still uses Ecco. And this is why I also find outlining useful in writing and for presentation, but not in note-taking, where a meshes or a networked or linked structure seems more useful—at least at first. In my experience (two-pane) outliners, which basically present a fixed view of hierarchically arranged categories, create more problems than they solve within the context of note-taking. But this is more a practical issue than one that has to do with the nature of reality.
Perhaps it is true that "the world is much much messier than [hierarchical thinking suggests]. Almost everything is actually a mesh not a hierarchy." Perhaps it is even true that "when hierarchies do exist in the data, it's very likely that you will find 2 or more inherent hierarchies that are orthogonal and in most real world situations it's more like 10. Which leads you to all sorts of mess with regional divisions versus category divisions and the same item being placed in lots of different places in the parallel hierarchies you're forced into using."
Non-hierarchical "tagging" or networks, in all likelihood, are better at capturing some aspects of reality, but none of this means that "outliners are harmful. The trick is to use both outlines and networks, or so it seems to me. Each of them has an appropriate use, and the trick is to know when to use what. The (presumed) fact that "the world is actually mess(h)y and not structured into elegant trees," does not only not imply that our attempt to understand the world must slavishly copy this structure, but may actually be taken to suggest that elegant trees give a perspective on the world that goes beyond the naive grasp of the surface structure of phenomena.