Tagging system-wide is a real paradigm-change

Tue Oct 13, 2009 10:23 pm

Think it through. Tagging is a whole second layer to the old folder-pardigm.

How you work today: you open openoffice-text, write a text, save the file -> a requester opens to select to folder to save the file to, give the file a name and you save it.

And think of it: by doing this, you in reality attach several tags to the file: being the names of the subfolders it resides in: e.g. documents/computer/programming/kde/plasma and the filename itself. You do this by clicking through the folder-tree, that presents subkategories, resp. cathegorical folder names to you.

If you would *really* use the full power of tagging, you can forget folders!

You would do as described, however, instead of opening a requester for placing the file in a subfolder of folders, you would be asked to give some tags to the file. I imagine a requester containing a tag cloud of used tags that change color when clicked and a text field to enter new tags. You do not need a filename either, because full-text indexing, tags and especially journaling ('the letter I wrote on sunday') should be much faster to find you the file than any folder/filename-technology.

So we see: tagging needs a complete change in working with files. Otherwise it is an additional task nobody understands: if I have to save a file (a task a lot of people do not understand and find annoying) and now they should give the file a lot of tags (even more annoying).

We need to do the whole step and that begins with changing file storage paradigm and with the way applications save files and ask us for filenames (the file requester is history). If we want tagging be effective and strong, we need to change really a lot!

incredion Registered Member Posts 87 Karma 0 OS	Tagging system-wide is a real paradigm-change Tue Oct 13, 2009 10:23 pm Think it through. Tagging is a whole second layer to the old folder-pardigm. How you work today: you open openoffice-text, write a text, save the file -> a requester opens to select to folder to save the file to, give the file a name and you save it. And think of it: by doing this, you in reality attach several tags to the file: being the names of the subfolders it resides in: e.g. documents/computer/programming/kde/plasma and the filename itself. You do this by clicking through the folder-tree, that presents subkategories, resp. cathegorical folder names to you. If you would really use the full power of tagging, you can forget folders! You would do as described, however, instead of opening a requester for placing the file in a subfolder of folders, you would be asked to give some tags to the file. I imagine a requester containing a tag cloud of used tags that change color when clicked and a text field to enter new tags. You do not need a filename either, because full-text indexing, tags and especially journaling ('the letter I wrote on sunday') should be much faster to find you the file than any folder/filename-technology. So we see: tagging needs a complete change in working with files. Otherwise it is an additional task nobody understands: if I have to save a file (a task a lot of people do not understand and find annoying) and now they should give the file a lot of tags (even more annoying). We need to do the whole step and that begins with changing file storage paradigm and with the way applications save files and ask us for filenames (the file requester is history). If we want tagging be effective and strong, we need to change really a lot!
TheBlackCat Registered Member Posts 2945 Karma 8 OS	Tagging system-wide is a real paradigm-change Tue Oct 13, 2009 10:44 pm This is already being implemented in KDE 4.4. Man is the lowest-cost, 150-pound, nonlinear, all-purpose computer system which can be mass-produced by unskilled labor. -NASA in 1965
incredion Registered Member Posts 87 Karma 0 OS	Tagging system-wide is a real paradigm-change Tue Oct 13, 2009 11:03 pm you mean KDE 4.4 has no file-save requesters and a db-like filesystem?