A "TreeMap" is a common method for visualising the data usage on a hard disk, the bounding rectangle reflects the overall size of the drive, and is subdivided into smaller sections based on the sizes of individual files and directories.
A "Sunburst" or "DaisyWheel" diagram is an alternative method
of representing the same thing, with the tree-like structure of the directories partitioning the disk.
A "Circle Packing" is another example of a visualisation that mixes size representations and hierarchical metaphors in a relatively well defined way.
Comparatively, I find the DaisyWheel delivers the most utility for disk-usage visualisation - but I think there's a better way even than that.
This idea is both a data-volume visualisation tool, but also with some extension could form an alternative interface into "Finder" or "Explorer" file and directory navigators.
The central drift draws on work by Maxime Causeret [see link below] where cell-like objects viewed from the top-down are caused to grow, as if on a petri-dish.
In this case, each cellular seed is determined by its position on the directory tree, its degree of "swell" determined by file/data content.
As the contents of the drive under review are read (or, if snapshots of the state of the drive are taken over time) the visualisation would spread and grow, quite organically, until such time as the limits of the petri-dish (or disk) are filled.
Making each object interactive, a scalpel-like mouse pointer could click and drag individual cells around, popping them through cell-membranes after a time-delayed pressure for relocation or deletion.
Similarly, regular or greppy expressions might be "washed" over the contents of a given petri-dish, staining or highlighting potential matches. These could then be copied, culled or otherwise examined for further analysis.
Setting parameters for the elasticness of cell-walls, and their propensity to fill the available space, and optional gentle pulsation amplitude and frequency might allow for some user-configuration.-- zen_tom,
Jan 09 2017
https://vimeo.com/196269431Max Cooper - Order from Chaos - Official Video by Maxime Causeret [zen_tom, Jan 09 2017]
http://www.datavizc...ogue.com/index.htmlDatavizcatalogue: A collection of different data-visualisation arrangements. [zen_tom, Jan 09 2017]
a video of blobs absorbing other blobs (agar.io) that reminds me of this idea
https://www.youtube...watch?v=Tts9NHL8D8I [beanangel, Jan 09 2017]
"Making each object interactive, a scalpel-like mouse pointer could click and drag individual cells around, popping them through cell-membranes after a time-delayed pressure for relocation or deletion." That is like if agar.io [link] were super cool!-- beanangel,
Jan 09 2017
Aha yes! Thanks [beanangel] I remember seeing someone
play that game a couple of years ago, but didn't
remember what it was called.
Part of the motivation behind this is a hunch that in time,
different applications will be identifiable more by their
storage profile, than by any kind of file-name recognition.
By visualising the storage-profile in this way, it ought to
become possible to perform a histological analysis across
a dataset and identify what its likely contents might be.-- zen_tom,
Jan 11 2017