Issue: Features for NIF URIs
From NLP2RDF-Wiki
This page is one of the Issues, which are currently under discussion
| Timline, NIF 2.0 and todos in this wiki (help is appreciated, registration is ) e |
|---|
| All is welcome, see here how to Get Involved, see here for a post about the roadmap |
Collection of use cases, requirements and Category:Issues <- current phase, your best chance to contribute
|
* Bug tracker for software:
|
Contents |
Introduction
We are currently collecting possible features. Note that this discussion is independent from the syntax of identifiers, which is discussed here: Issue:_Syntax_for_NIF_URIs. It is easy to see, that '#offset_717_729' and '#char=717,12' can have equivalent meaning.
Examples
Some examples to make the discussion more concrete.
LinkedData HTML
First occurrence of the string Semantic Web on http://www.w3.org/DesignIssues/LinkedData.html . Note that the document:
- has a total of 26610 characters
- is an information resource
- the string Semantic Web occurs for the first time at position 717, has 12 characters and ends at 729
- we use the prefix ld: for 'http://www.w3.org/DesignIssues/LinkedData.html#'
Collection of possible features Please add any you can think of
- desc=offset or desc=hash
- used in NIF 1.0 to tell how the fragment identifier should be parsed
- beginIndex = 717
- endIndex = 729
- offset = 717,729
- length = 12
- length of the string
- encoding = utf-8
- text = Semantic%200Web
- hash_with_context = 4_12_711ffecc1815feff00f9314eeb6eaa12_Semantic%20Web
- produced by md5 ("The (Semantic Web) isn");
- hash_of_text = 103b850abb89c034ccc2a2d2e6756fe3
- produced by md5 ("Semanic Web");
- contextLength = 4
- leftContext = 'The '
- rightContext = ' isn'
- regex = Semantic\wWeb
- selects all occurences that matches the regex
Examples
Options
Links
Previous discussions
Current NIF 1.0 URIs
- Recent stability benchmark and survey of alternative schemes: EKAW 2012 paper
- 1.0 Spec, URI Scheme section
- Technical Report
Literature
Other (W3C and IETF)
Category:
