Defining prefix mappings for a dataset

[GoogleCodeExporter]

Would it make sense to document a method for defining a set of prefix mappings 
for a dataset?

The particular use case would be in conjunction with SPARQL endpoint 
descriptions: If I want to use a voiD file to connect to a dataset via its 
SPARQL endpoint, then it would be really helpful if the dataset publisher could 
provide me with a default set of prefixes that I could use when writing queries 
against the dataset.

The VANN vocabulary would be an obvious starting point:
http://vocab.org/vann/

This is perhaps related to Issue 45.

Original issue reported on code.google.com by richard....@gmail.com on 9 Jun 2010 at 11:32

[GoogleCodeExporter]

Original comment by richard....@gmail.com on 9 Jun 2010 at 11:33

• Added labels: Type-Enhancement
• Removed labels: Type-Defect

[GoogleCodeExporter]

In today's teleconference, we decided to do this in 2.0, time permitting (low 
priority)

Original comment by richard....@gmail.com on 10 Jun 2010 at 12:57

• Added labels: Milestone-Release2.0, Priority-Low
• Removed labels: Priority-Medium

[GoogleCodeExporter]

I do the same kind of thing here: 
http://kwijibo.talis.com/voiD/sparql?uri=http%3A%2F%2Fapi.talis.com%2Fstores%2Fk
wijibo-dev2%2Fservices%2Fsparql

 the prefixes used are obviously not that meaningful ...

It would be nice to use the  usual namespace prefixes instead - but  it's not 
really the absence of anything in the voiD descriptions that stops me doing it  
- I could probably take a lot of them from prefix.cc or a local mapping. (This 
wouldn't satisfy the task of defining prefixes for instance data URI 
namespaces, but that's probably a less common requirement).

Actually a heuristic i've seen is to take the last path part before the final 
hash or slash of a uri and use that as the prefix (moving back to the next last 
path part until you find a string that is a valid prefix).

You could apply this heuristic to the example uris to get instance data 
namespace prefixes. 

If you look at 
http://nasa.dataincubator.org/mission/apollo-11/role/capsule-communicator/0.html
  it uses this approach and seems to work quite well.

Original comment by K.J.W.Al...@gmail.com on 10 Jun 2010 at 1:16

[GoogleCodeExporter]

As per today's call, we postpone this to voiD 3 because we are unsure about the 
best way of doing this and the requirements and use case is not 100% clear. 
Let's gather more experience.

Original comment by richard....@gmail.com on 29 Oct 2010 at 10:06

• Added labels: Milestone-Release3.0, Priority-Medium
• Removed labels: Milestone-Release2.0, Priority-Low

[GoogleCodeExporter]

Issue 49 has been merged into this issue.

Original comment by richard....@gmail.com on 2 Nov 2010 at 9:41