thank you helix for all these informations but i think now it takes us
a bit to far away from the subject.
I think it is good to know that URN's may be supported in 4.0.
I can import items with outside generated URN's
Storage is separated from DSpace.
But the needed URN generation is not implemented yet.
And yes, i think it should be, as you say, implemented properly.
I'm not a fan of patching an application to get a core feature working.
Again, thank you for that amazing support!
Zitat von helix84 <helix84@...>:
> On Tue, Oct 30, 2012 at 3:18 PM, <marco.weiss@...> wrote:
>> how we will submit items is not clear today because i evaluating different
>> applications to get a basis of information for discussion which system are
>> possible the best matching one.
> I see, so you're not using DSpace yet.
>> After that i have to work out a process and it depends on the
>> interconnection between, where is the storage, how can i access the storage,
>> do we need an integration of an image server ... and if i can answer all
>> these question i can tell you what we will use or both.
> Storage is simple, at least from the DSpace point of view, because
> it's just a directory (or a set of directories) on the local
> filesystem. How you mount them is another problem out of scope of
> But you're probably talking about deciding whether to store large
> files in DSpace or link to them from DSpace metadata.
> In any case, that doesn't have much bearing on deciding between import
> vs. submit.
>> Ok i extract that if i use, for example the batch import, i can generate the
>> URN before and put it in DSpace during the import?
> Right, that would be my suggestion if you didn't use the submission
> process, like me. It wouldn't require any changes in DSpace.
>> Yes i think it will be useful for more institutions in Germany as well in
>> other countries.
> I thought I'd advise you how to change XMLUI or JSPUI to accomodate
> the identifiers during submission, but if it would be a widely used
> change, it should be implemented properly in the dspace-api underlying
> all the interfaces. There's actually a service for identifiers, but
> only handles are currently first-class citizens and more developments
> in this area are expected. Unfortunately, I'm not very familiar with
> this part of the codebase. It would be best to contact @mire and ask
> about their unfinished work in this area (External Identifier
> Compulsory reading: DSpace Mailing List Etiquette