3 ms·
It's a source of ongoing perplexity to me that so many Linux file managers don't even offer a tree view, never mind some of the more advanced possibilities. Con
by anigbrowl 7y ago
It's a source of ongoing perplexity to me that so many Linux file managers don't even offer a tree view, never mind some of the more advanced possibilities. Consider Robertson's work on Cone Trees from Xerox Parc in....1991.
https://research.tableau.com/sites/default/files/p189-robertson.pdf https://research.tableau.com/sites/default/files/p189-robert...
and variants thereof at https://infovis-wiki.net/wiki/Cone_Trees https://infovis-wiki.net/wiki/Cone_Trees
Oddly, there has been some work done on source code trees such as Gource, but for some reason nobody has yet thought of applying this to file systems or browser navigation paths.
https://gource.io/ https://gource.io/
- j88439h84 7y agoThere's https://dystroy.org/broot/ https://dystroy.org/broot/ Are those fancy tree displays actually useful for anything or is just pretty?
- ShamelessC 7y agoYeah I think it's obvious that graphical trees are prettier than they are effective user interfaces. Broot is a good example of just how much effort goes into making nested structures easily navigable. Trees may be good for presentation but ultimately lists with indentation for nesting are for more navigable. This translates to GUI concepts as well. Both Windows and MacOS show a few flat lists relative to the current directory rather than showing all nestings at once like a tree does.
- anigbrowl 7y agoThat's for console use. Trees are useful for looking at the whole thing at once, much as it's easier to appreciate a real tree that way than by maintaining a list of the leaves and branching distances. I'm a a researcher and my source material is ~50gb spread across ~50k hand-picked files with pretty arbitrary structure, plus another 4-5 times that in mirrors of other collections. Trees let me see things that I would overlook by reading lists all the time. Right now I half-ass use some bioinformatics tools originally intended for cladistics and proteome research to visualize the collection. Some days I go looking for specific information or catalog what I have, other days I stumble across a cache of valuable data and might select hundreds of individual files.
- j88439h84 7y agoI'm having trouble imagining this. Would you be able to show a screenshot?
- ORioN63 7y agoNot OP, but I'm guessing it's something similar to how I watched fast moving logs to look for anomalies: I reduced text size enough, to a point it wasn't really text I was looking, but patterns of line content and length. This was enough to see that things were running proper.
- anigbrowl 7y agoI use a piece of software called Dendroscope to kick out circular cladograms, as documented here: https://software-ab.informatik.uni-tuebingen.de/download/dendroscope/manual.pdf https://software-ab.informatik.uni-tuebingen.de/download/den... It works OK, but it means directory listings first have to be kicked into a Newick tree format: https://en.wikipedia.org/wiki/Newick_format https://en.wikipedia.org/wiki/Newick_format I use a variety of other graph-based tools too, such as Maltego, Gephi, and Cytoscape, but to be honest I do so in a very ad-hoc way for specific tranches of data ( afile collection, or a relevant social media stream) and keep most of it my head. Otherwise I will get hung up on tooling rather than the underlying material.
- hnuser123456 7y agoWindows only, but https://github.com/mrange/FolderSize https://github.com/mrange/FolderSize
- PetahNZ 7y agohttps://www.jam-software.com/treesize_free https://www.jam-software.com/treesize_free
- anigbrowl 7y agoWinDirStat is probably the best treemap style product. It's faster that doing directory listings on the command line, amusingly enough.
- hnuser123456 7y agoI much prefer foldersize to windirstat
- jmiskovic 7y agoWith file system the resulting graph would be too big and too complicated to be useful. You would need heavy filtering to make any sense of it. In best case it would be a decent for determining where most of your files are, but those are probably hidden system directories and .git metadata. Most tree visualization research is focused on static rendering of whole tree. For file browsing and many other applications it's better to have just section of tree with interactive navigation around trees. This makes visualization an easier with more predictable layout, and the user is not overwhelmed with data. This approach is basically replicating model of our attention/focus and how we make sense of complex structures. The design of file browsers indeed seems to be stuck in 80s. Online storage services are copying this design as well. Is anyone aware of better design, in form of paper or prototype?