[digiKam-users] Sharing Database/Collections

[Remco Viëtor]

On jeudi 19 septembre 2019 11:32:43 CEST Sveinn í Felli wrote:
> Þann 19.9.2019 08:49, skrifaði Martin Burnicki:
> > Andrey Goreev wrote:
> >> I do not think digiKam can do that...
> > 
> > Shouldn't this work if the collection is in a cloud folder like
> > Nextcloud (or Dropbox, FWIW)?
> > 
> > If you make changes to photo or an album (i.e. just the file or photo)
> > on one machine then the changes are synchronized to all machines that
> > share these folders, and if you start DK next time it will detect the
> > changes and update the local database accordingly.
> > 
> > But of course this is only going to work if you write tags and other
> > metadata to the image file or a sidecar file.
> > 
> > If you have configured DK to *not* write metadata (tags etc.) to the
> > files but keep them only in the DB you'd have to synchronize the
> > database in a consistent way, which can be much harder.
> 
> IMHO it could be instructive to set up Nextcloud on a NAS over a LAN,
> and try out how Digikam reacts to such synchronization - both databases
> and image files. Probably I will test this kind of a setup, but not
> until late October when time permits.
> 
> Nextcloud/OwnCloud has fine-grained access-controls for users/groups and
> sharing, and has pretty advanced mechanisms for resolving conflicts in
> case of simultaneous edits.
> 
> I presume that if using a synchronized database one would have to share
> _all_ the files in each local collection _and_ the paths would have to
> be the same on all devices, right?
> 
> Anything else obvious?

OP seems to want to be able to use a shared collection, and  modify it, 
*without* network access, e.g. while travelling. That will mean he will have 
to keep a copy of his collections on his local hard drive, and sync after each 
trip.

If he's the only user of those collections, then the easiest would be to copy 
collections and database(s) back and forth. That seems to be what he does now.

Having Digikam detect changes can be very slow, especially over a network (too 
many files to read).

Remco.