The big problem with files is that they are linear. Up to computers, the standard interface for people to work is the sheet of paper, which is a two dimensional surface. In fact, it was even better because you can organize papers into books or collections, which gives a 3rd dimension. Files, being linear, make it difficult to parse and analyze, which is always the problem we need to solve with computers. This problem is much easier to solve when we can work on 3 dimensions and organize data spatially (this is the guiding principle of charts and tables).
I disagree. The linearity is super helpful for scanning and finding quickly, by time, by name, etc. Id argue that any app that makes the interface "spatial first" will have to add these linear features on top of what they're doing to find things fast, at a minimum for full-text search.
I do think spatial organization can help for human consumption of not easily orderable datasets >1000 items, but I think that's a niche case for everything ive interacted with so far.
>I do think spatial organization can help for human consumption of not easily orderable datasets >1000 items, but I think that's a niche case for everything ive interacted with so far.
Well, when the data gets above 1000 items, you have to start filtering, finding, or revisualizing it to be useful. You can no longer keep the whole dataset comfortably in your head, or even on screen.
For revisualizing, it might be comfortable to plot it on a 2D map or to simply graph an interesting scalar.
For revisualizing in a 3D spatial setting... I haven't seen any actual good 3D spatial layout of data. Disregarding "experiences" (like art galleries, which lay out data in a way to promote interest in a topic and fractal-like amounts of subtopics, but is human curated, like Saganworks[1]), the only one I guess is like the high level "portal" to categories of data. Like "Documents", "Downloads", etc but think 500+ categories of more granular entity types. Everything is non-comparable apples and oranges (except for perhaps shared common data like timestamp) and there's no way to quickly scan it (5 or so seconds to find alphabetically perhaps), so spatial might help to quickly locate your entity type with muscle memory. I haven't seen another use case, but would love to explore what others are doing in this space!
I meant specifically what organization did you use for "not easily orderable" data. Or was i wrong in assuming you meant simply lack of (natural number like) order and didn't mean total lack of clear classifiability?
Ohh, the only one I have found was the one I mention in the last paragraph. Large number of categories. Everything else I've found has at least one good way to order it
I meant organization scheme not any particular organization if unclear. That app looks like some kind of 3d world/metaverse to me I will study it more later.
I'm confused. Do you mean what particular organization algorithm would I use to organize unorganizable data? Not just the type of visualization it would use (ie, 2d list vs 3d spatial)?
I was just clarifying and asking if you understood the word "organization" to mean some particular company or organization like that sagan thing or in my intended sense of organization scheme/visualization technique in general?
