Home

Acknowledgements

Documentation

The corpus

People

Publications

Sponsors

References

Links

Appendices

NOTE:DON'T FORGET TO INCLUDE THE SOCIAL DATA TAGS AT THE START OF EACH INTERVIEW

In the sequence of entity references &tls001; to &pvc918; that immediately follows the global header, each reference denotes a single constituent interview of the corpus. Each such interview is itself a TEI-conformant XML document, and all have same structure:

 where:

• Each <TEI.2> element contains a single interview which is uniquely identified by an "id" attribute whose value is one of the entity names in the range specified in the DOCTYPE declaration.

• <teiHeader> contains information specific to the interview. This header has the same structure as the global one, but it is much simpler, since most characteristics of the individual informant cases are described globally. In fact, it contains only one of the four main elements of the global header, <profileDesc>, and <profileDesc> contains two subnodes: <creation> and <particDesc>. The first of these identifies the source of the material in the file --TLS or PVC-- using the full names of these projects, that is, 'Tyneside Linguistic Survey' and 'Phonological Change and Variation project'. The <particDesc> element lists the participants in the interview using the <person> tag with unique "id" attribute, together with the associated social data using the tags and attributes in Guidelines 23.2.2.  Note that  the social information available to NECTE from both TLS and PVC was not uniformly complete; where information is missing the relevant tag is left empty. Note also that there is no social data for interviewers.

• The <text> element contains the text of the interview. This element is quite complex: accounts of its internal structure and associated XML tags are available via the following links:

-- The overall structure of the <text> element

-- Alignment

-- Speech tags

-- Part-of-speech tags

-- General editorial tags