Say I don't wont to run the file index, at least not in it's current slow sweep form, which is done everytime you login/restart .

Why doesn't semantic data get gathered on a usage basis, say I open pdf-1, why doesn't it's semantic data not get added to the nepomuk database? Now, it's only added to the recent used documents, but the pdf's real semantic data, for say the author's name, is now not added.

Maybe I'm wrong and something else is the matter, but I've tested this with a pdf which had a proper author defined, but it apparently is not included to the database.

It doesn't have to look for everything on start. You can disable that - http://userbase.kde.org/Nepomuk/FileInd ... artup_Scan

If there is some pdf file whose author is not being detected, then please file a bug and make sure your upload the pdf file. If we can reproduce it, then we can fix it.

Also, Nepomuk doesn't index file when you open them. If you want them to be indexed, please put them in an indexed directory. It will then index them whenever they are modified which is a better approach.

moved to Semantic Deskop forum

vHanda wrote:Also, Nepomuk doesn't index file when you open them. If you want them to be indexed, please put them in an indexed directory. It will then index them whenever they are modified which is a better approach.

Modified also means when they are copied/moved there for a first time? Basically what I'd like to know is, besides a full filesystem index, what are triggers for individual files to be indexed?