This is a major new release of the P5 Guidelines, which
introduces some significant new material as well as implementing
an unusually large set of other changes and error
corrections. Since the last release in March 2011, members of
the TEI community have proposed over 60 feature requests and
reported about the same number of errors; dealing with these has
kept the TEI Council busy, and most of these tickets have now
been closed. Discussion of each modification and each action
taken is available on the relevant Sourceforge ticket, which is
also referenced in the change log, as usual.
We mention here just a few of the more notable changes carried out
for this release:
- Several new elements have been introduced to facilitate
a more "document-focussed" (as opposed to "text-focussed") way
of working. A new <document> element can now be used to
organize digital images of a source document in much the same
way as the existing <facsimile>, but with the added
ability to embed detailed source transcriptions of the source
within the same structure. A number of other elements have
been added or modified to support the detailed encoding of
transcriptions of individual documents, and a mechanism has
been defined for the encoder to represent the evolution of a
text through various writing stages. The existing chapter on
Physical Transcription has been exhaustively
revised to cater for these revisions which will, it is hoped,
go some way to improve the support offered by the TEI scheme
for documentary and genetic editing.
- The licensing conditions applicable to TEI products were changed
(moving to a dual license model of CC-BY and BSD2 for most TEI
outputs), and made more evident; the source code tree was
tidied up, with all unused items being moved out; some old annoyances
— notably the meaningless distinction between "mwa" and "opt" as
properties of attributes — were
removed
- A new <notatedMusic> element was added to support
representation of musical notation within texts in a way compatible
with the existing Music XML standard.
- A number of minor changes were made to make it possible to
represent dates, and normalised dates, using non-Julian calendar
systems; a new attribute class was defined to facilitate provision of
sort keys.
- Several modifications were made to enhance the representation of
more carefully structured or richer bibliographic entries, notably the
ability to nest one <bibl> within another, and the ability to
specify a work's language.
- All examples were revised to drastically reduce the amount of
unnecessarily divergent and encoding practices (such as the values used for
the @rend attribute, or for language identifier codes)
- The content models of some elements were relaxed as part of
ongoing work to align encoding by the Text Creation Partnership to TEI
P5, to permit (for example) stage directions within verse, or verse
within headings.
- The content model of the <subst> element was revised in
line with what we believe to be good practice for this element (the
only legal contents of <subst> are now <del>,
<add>). At the same time, a new element <substJoin>
element has been added to deal with more complex situations. It is
regretted that this may result in some existing documents becoming
invalid under the new schema.
- A new <spGrp> (group of speeches) element has been added
to represent floating div-like sequences of speeches which function as a
single unit, for example "numbers" within the script of a musical or opera.
As far as possible, and in line with stated
policy, none of the changes made at this release should
invalidate existing documents. Rather than simply removing features or practices
which are now considered erroneous, care has been taken to indicate
that such features are ‘deprecated’ and may be
removed at a subsequent release. Work is ongoing on how exactly to
formalise this policy.