new release of TEI P5

classic Classic list List threaded Threaded
6 messages Options
Reply | Threaded
Open this post in threaded view
|

new release of TEI P5

Sebastian Rahtz-2
We are pleased to announce that a new packaged release (0.1.9) of TEI P5
is now
available.  The associated projects of Roma and XSL stylesheets are also
available in new
versions to be in sync.

You can get TEI P5 in several ways:

   * from the TEI SourceForge site at http://tei.sf.net in the File
Releases area.
    The  release consists of 5 .zip files for the different areas of P5,
or as a single
     .tar.gz archive ready to install in a recommended layout
   * using CVS on Sourceforge
   * Debian Linux users can add
        deb http://tei.oucs.ox.ac.uk/teideb binary/
     to their /etc/apt/sources.list file and get the packages
       tei-emacs tei-p4-doc tei-p4-lite tei-p4-schema
       tei-p4-xsl tei-p5-database tei-p5-doc tei-p5-exemplars tei-p5-schema
       tei-p5-source tei-p5-test tei-p5-xsl tei-roma
     for the full technicolor effect
     !!!! Debian users are _strongly_ advised to first delete all
existing TEI-related packages!!!
  * RPM users can try the RPM packages at http://tei.oucs.ox.ac.uk/teideb/,
    though these have not been tested

in addition:

  * the test TEI Emacs release for Windows  noted earlier today has up
to date support files
  * the next release of the oXygen XML editor will, if all goes well,
have an updated TEI framework
     (watch out for their release announcements)

We would be delighted in people using other Windows or Mac setups for
their TEI XML
work could write down the notes on how (if at all!) they get TEI P5
installed and working,
and send them in for inclusion on the web site.

Notable changes since the last packaged  release (0.1.2) are listed
briefly below
(please refer to the ChangeLog for the full list)

* several modifications made to the test suite and the Makefile used
  in production and validation of P5 drafts
* Roma processor revised substantially to generate more reliable schemas
* content models of <list> and <biblStruct> simplified
* content model of editorialDecl changed to use new "policy" class
* new <biblItem> element added
* new <binaryObject> element added
* new MS (manuscript description) chapter substantially revised
* new "examplars" directory containing templates for a range of P5
  documents
* <index> element changed to supply values as content
* removed <indexSpan> element and added new "spanning" class
* content model for <del> changed to match that of <add>
* erroneous definition for phrase.seq corrected
* content model for <dateLine> changed for consistency with other
  divtop elements
* <interp> and <interpGrp> changed to supply  values as content

Priorities for the next release include

- removal of remaining "text" attributes
- overhaul of the class system
- major revision of ST (the chapter which describes the module system)
- further revision of SA (the chapter describing linking and alignment)

We welcome all comments and suggestions from beta-testers! Please use
the Sourceforge site both to monitor progress and to propose
enhancements.

A draft tutorial on "Getting started with P5" is available at
http://www.tei-c.org/Drafts/edw88.xml

LB 11 Jul 05

--
Sebastian Rahtz
Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

OSS Watch: JISC Open Source Advisory Service
http://www.oss-watch.ac.uk

Reply | Threaded
Open this post in threaded view
|

Re: new release of TEI P5

ron.vandenbranden
Administrator
Hi,

I see the release notes for package 0.1.9 state that the new MS Description
chapter has been substantially revised. Since I am integrating the MS
description module in a P4 DTD, I'd like to know what is meant exactly:
1) only the documentation changed?
2) both the documentation and the module changed?

A quick scan suggests that 1) is the case. However, this leads to
inconsistencies between documentation and source:

E.g.: the chapter text documents all elements of <msDescription> that occur
also in the reference part; yet, when the chapter text quotes (?) the formal
specification of <msDescription>, it states the following (§ 13.2 The
manuscript description element):

"msDescription.content = empty"

Another one is the documentation of <origPlace> as specialisation of the
<name> element (§ 13.3.1 Origination):
"As a specialisation of the <name> element, the <origPlace> element has a
reg attribute which can be used to supply a regularized form of the place
name. Alternatively, the key attribute can be used to provide an identifying
code or key for the name."

Yet, there is no trace of this in the reference part and the source itself,
where <origPlace> is only declared as member of tei.phrase.

Here I stopped reading, confusion took over.

Can anyone tell me what's wrong or out of sync or misunderstood?

Ron Van den Branden

Reply | Threaded
Open this post in threaded view
|

Re: new release of TEI P5

Peter Heslin
In reply to this post by Sebastian Rahtz-2
Sebastian Rahtz <[hidden email]> writes:

> We are pleased to announce that a new packaged release (0.1.9) of TEI
> P5 is now available.

I have updated version 1.3 of Easymacs to include these files, and have
added the new TEI document templates.

http://www.dur.ac.uk/p.j.heslin/Software/Emacs/Easymacs/index.php

On a somewhat related note, I have taken my Emacs package
tei-html-docs.el, which allows you to view the TEI guidelines for the
current document element with a single keystroke, and divided it into
two new packages, tei-html-p5-docs.el and tei-html-p4-docs.el for P5 and
P4 respectively.  This was prompted by a change in the HTML format of
the P5 guidelines, but it makes sense anyway to keep these separate.

http://www.dur.ac.uk/p.j.heslin/Software/Emacs/index.php

Peter

Reply | Threaded
Open this post in threaded view
|

Re: new release of TEI P5

Sebastian Rahtz-2
In reply to this post by Sebastian Rahtz-2
Peter Heslin wrote:

>
>On a somewhat related note, I have taken my Emacs package
>tei-html-docs.el, which allows you to view the TEI guidelines for the
>current document element with a single keystroke, and divided it into
>two new packages, tei-html-p5-docs.el and tei-html-p4-docs.el for P5 and
>P4 respectively.  This was prompted by a change in the HTML format of
>the P5 guidelines
>
>
yes, I am glad you mentioned that. for those who have
not noticed, the P5 HTML Guidelines have files which
are named after the name of the element/class they document,
rather than being named after the ID of the specification for that
element. For historical reasons, those IDs were uppercased
and sometimes truncated forms of the element/class macro,
which is an endless source of confusion

Sebastian

Reply | Threaded
Open this post in threaded view
|

Re: new release of TEI P5

Sebastian Rahtz-2
In reply to this post by Sebastian Rahtz-2
Ron Van Den Branden wrote:

>it states the following (§ 13.2 The
>manuscript description element):
>
>"msDescription.content = empty"
>
this is an error which I will trace tonight. for the truth, see

http://www.tei-c.org.uk/Query/tag.xq?name=msDescription


Sebastian

>
>

Reply | Threaded
Open this post in threaded view
|

Re: new release of TEI P5

Sebastian Rahtz-2
In reply to this post by Sebastian Rahtz-2
Ron Van Den Branden wrote:

>E.g.: the chapter text documents all elements of <msDescription> that occur
>also in the reference part; yet, when the chapter text quotes (?) the formal
>specification of <msDescription>, it states the following (§ 13.2 The
>manuscript description element):
>
>"msDescription.content = empty"
>
>
Congratulations, you have discovered a weird and unnatural bug in the
stylesheet
which generates the RelaxNG definition in the body of the guidelines.
It is a genuine mysery to me what went wrong, and I have worked around it.

This does not affect the schemas or DTDs.

>Another one is the documentation of <origPlace> as specialisation of the
><name> element (§ 13.3.1 Origination):
>"As a specialisation of the <name> element, the <origPlace> element has a
>reg attribute which can be used to supply a regularized form of the place
>name. Alternatively, the key attribute can be used to provide an identifying
>code or key for the name."
>
>Yet, there is no trace of this in the reference part and the source itself,
>where <origPlace> is only declared as member of tei.phrase.
>
>
>
I defer to the chapter authors to discuss that one.

I had hoped to get a new release out tonight fixing the bug
noted above; but in the process I foolishly upgraded my XSLT
processor (libxslt1.1) and found that it completely, utterly,
broke the TEI processing system. The recovery from this
has duly ruined my evening..... (and I hope to goodness
I _have_recovered, am running all tests now).

--
Sebastian Rahtz
Information Manager, Oxford University Computing Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

OSS Watch: JISC Open Source Advisory Service
http://www.oss-watch.ac.uk