gmane.text.tei.general

Re: can a corpus have a facsimile?

Sebastian Rahtz — 2013-05-23T16:29:23

On 23 May 2013, at 17:23, Martin Holmes <mholmes< at >UVIC.CA>
 wrote:


yes, definitely. use model.resourceLike. I don't think
there will be any knock-on effects. 


--
Sebastian Rahtz      
Director (Research) of Academic IT
University of Oxford IT Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

Re: can a corpus have a facsimile?

Martin Holmes — 2013-05-23T16:23:49

+1 from me too. It would presumably go from:

element teiCorpus
{
    att.global.attributes,
    att.global.linking.attributes,
    att.global.analytic.attributes,
    att.global.facs.attributes,
    att.global.change.attributes,
    attribute version { data.version }?,
    ( teiHeader, ( TEI | teiCorpus )+ )
}

to

element teiCorpus
{
    att.global.attributes,
    att.global.linking.attributes,
    att.global.analytic.attributes,
    att.global.facs.attributes,
    att.global.change.attributes,
    attribute version { data.version }?,
    ( teiHeader, facsimile, (TEI | teiCorpus )+ )
}

But should it also allow sourceDoc or fsdDecl, like this?

element teiCorpus
{
    att.global.attributes,
    att.global.linking.attributes,
    att.global.analytic.attributes,
    att.global.facs.attributes,
    att.global.change.attributes,
    attribute version { data.version }?,
    ( teiHeader, model.resourceLike, (TEI | teiCorpus )+ )
}

The same arguments apply as for facsimile, surely?

Cheers,
Martin

On 13-05-23 0

Re: can a corpus have a facsimile?

Gabriel Bodard — 2013-05-23T15:14:14

That sounds eminently sensible to me. We explicitly sanction the use of 
'facsimile' to encode images of objects and settings, not necessarily 
just the text-bearing part of them, so a facsimile of a corpus is 
absolutely as appropriate as a facsimile of a monument bearing several 
inscriptions, a folder containing many documents, or an anthology 
collecting multiple novels...

In short: +1

On 23/05/2013 10:46, Sebastian Rahtz wrote:

can a corpus have a facsimile?

Sebastian Rahtz — 2013-05-23T14:46:21

I find myself in a position where i have <teiCorpus> element and want to have a <facsimile>
child.   Why, you ask? because I have 2500 gravestones, which are "documents" in their
own right, so I have them  as a <TEI> each; but I also have a plan of the whole cemetery which
I represent as a <facsimile> with nice <surface>s and <zone>s. I want to wrap them all up together.

am I being silly? or should we allow <facsimile> as a child of <teiCorpus>?
--
Sebastian Rahtz      
Director (Research) of Academic IT
University of Oxford IT Services
13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431

Re: encoding choices for a particular document in ?

Andreas Triantafillidis — 2013-05-23T13:04:20

i was hoping you wouldn't send me down the ODD path just yet :) Do you suggest a specific starting point for a complete newbie like myself?




-------
andreas trianta(fillidis)
andreas< at >trianta.eu

resident SOB customer 
at thinking lpc

Re: encoding choices for a particular document in ?

Lou Burnard — 2013-05-23T12:58:42

If you want to document down to this level, which I hope you do, then 
the place for it is your project-specific ODD!

On 23/05/13 13:57, Andreas Triantafillidis wrote:

Re: encoding choices for a particular document in ?

Lou Burnard — 2013-05-23T12:57:23

<tagUsage> is for recording your *actual usage* of TEI elements.  It 
isn't the place for an explanatory essay saying why you chose to 
use<dictScrap> rather than <entryFree> for example.

Discussion  of your editorial principles, choices, praxis does seem to 
belong rather in <editorialDecl>


On 23/05/13 13:11, Laurent Romary wrote:

Re: encoding choices for a particular document in ?

Andreas Triantafillidis — 2013-05-23T12:57:06

hi laurent


it would seem so, but how would i go about describing more complex element+attribute cases (ie <div < at >type="task" < at >subtype"...">) or even moreso nested element and attribute templates?

thanks btw



-------
andreas trianta(fillidis)
andreas< at >trianta.eu

resident SOB customer 
at thinking lpc

Re: encoding choices for a particular document in ?

Laurent Romary — 2013-05-23T12:11:37

tagUsage is the perfect tool for this, isn't it?
Laurent

Le 23 mai 2013 à 14:07, Andreas Triantafillidis a écrit :


Laurent Romary
INRIA & HUB-IDSL
laurent.romary< at >inria.fr

encoding choices for a particular document in ?

Andreas Triantafillidis — 2013-05-23T12:07:40

hi

we would like to document, within each file, the choices we make regarding usage of TEI elements and attributes. Sometimes we even would like to list TEI alternatives for encoding the same thing, and why we chose one over the others

these are really meant for editors and developers so I suppose it would be better to store in the header (they are in no way part of body) 

i'm thinking <editorialDecl>

any thoughts, suggestions?

thanks :)



-------
andreas trianta(fillidis)
andreas< at >trianta.eu

resident SOB customer 
at thinking lpc

Re: placeName types

Tom Elliott — 2013-05-21T15:02:43

Thanks Stuart, will do.

Tom Elliott, Ph.D.
Associate Director for Digital Programs and Senior Research Scholar
Institute for the Study of the Ancient World (NYU)
http://isaw.nyu.edu/people/staff/tom-elliott



On May 20, 2013, at 4:05 PM, stuart yeates wrote:

Blackwell Companion to Digital Literary Studies

Susan Schreibman — 2013-05-21T13:15:12

The editors of The Companion to Digital Literary Studies, first published
by Blackwell in 2008, are pleased to announce that it has just been
released in paperback.

At $49.95, €36.00, and £29.99, it is now affordable for classroom use. Its
31 chapters -- ranging from ePhilology to e-Literature, from digital
poetry to cybertextuality -- are as provocative and pertinent now as when
the volume was first released. TEI is covered in many chapters.

For further details, and for instructors to request an evaluation copy,
please see
http://www.wiley.com/WileyCDA/WileyTitle/productCd-1118492277.html.

Susan Schreibman and Ray Siemens

ProDoc< at >DocEng 2013: Call for Submissions to the Doctoral Consortium

Tamir Hassan (DocEng13 — 2013-05-21T11:22:02

--------------------------------------------------------------------
                            ProDoc< at >DocEng
   Doctoral Consortium at the ACM Symposium on Document Engineering
            http://www.doceng2013.org/doctoral-consortium/

Call for Submissions

* Submission Deadline: June 5, 2013 (24:00 CEST)

* Notification of Acceptance: June 20, 2013

--------------------------------------------------------------------

For the first time, the ACM Symposium on Document Engineering will
feature a doctoral consortium, called ProDoc< at >DocEng.

PhD students present their dissertation project and will get
feedback from a panel of senior researchers as well as from the
general audience.

ProDoc< at >DocEng is intended to provide constructive criticism and help
PhD students in formulating their research question, deciding about
methods and approaches to use, and creating further ideas.  It is a
good place to learn about how to conduct a dissertation project and to
learn about leading edge research, the results of which

Re: placeName types

Tomaz Erjavec — 2013-05-21T08:01:15

Dear Thomas, all,
thanks for the suggestions - I personally am convinced now that we should indeed extract the places and have them, with geo, and, preferably, pointers to geonames and Wikipedia, stored separately. 
However, as I said, we are now only putting together a first prototype, where we hoped we could get away with being a bit sloppy - extracting and merging the placeNames will be a bit of work, some of the names are quite structured, in several languages, historical spellings, etc.
Btw, the current set of coordinates we have were extracted from Google - the reason was that geoplaces sometimes places coords by half a km off, which can be enough for the centre of a small town to be displayed in the middle of a forest... But then I didn't know that geoplaces can be corrected - with this, I think it would be well worth switching.
All the best,
Tomaž

-----Original Message-----
From: TEI (Text Encoding Initiative) public discussion list [mailto:TEI-L< at >LISTSERV.BROWN.EDU] On Behalf Of Thomas A. Carlson
S

Re: placeName types

stuart yeates — 2013-05-20T21:05:58


I would encourage you to file a bug, preferably with an exemplar which 
contrasts with the existing exemplars in the standard.

cheers
stuart

Re: placeName types

Lou Burnard — 2013-05-20T20:42:55

To be fair, the argument for saying that < at >type on xxxName could be 
providing information about the type of xxx rather than the type of name 
is not entirely stupid. We all say "London is a capital city", 
"Liverpool is a port", etc. without pausing for thought, because the 
mapping of name to named entity is transparent in natural language. And 
there are even situations where we use an xxxName precisely to indicate 
a < at >type of xxx -- as in "Cherbourg is the Liverpool of Northern France" 
(well, kind of)

However, none of that weakens the justice of your remark that we should 
try to clean up current ambiguities in the Guidelines, either by 
confronting them explicitly and saying where and in what circumstances 
it is convenient to attach a xxx-related typology to an xxxName, or by 
rooting out such heresies wherever they appear.

Lou

  On 20/05/13 21:32, Tom Elliott wrote:

Re: placeName types

Tom Elliott — 2013-05-20T20:32:04

Dear all:

My original post has prompted an interesting and important tangential discussion that I have no wish to undermine. In fact, I'd like to note here that the EpiDoc community considers it best practice to mark up place names in text and metadata with the form <placeName ref="URI"> where URI targets a linked data resource for the geographic feature in question (for Greek and Roman inscriptions, we encourage the use of "place" resources defined by http:/pleiades.stoa.org). This usage has been taught in EpiDoc training sessions for years, and will be more clearly reflected in revisions of the EpiDoc Guidelines now in preparation. 

But I'd like to come back to my original question, i.e., should the value of < at >type on <placeName> address the type of the name or the type of the referenced place (the latter being the usage enshrined in the TEI Guidelines at present)?. If I've read the responses in the thread aright, there are a couple of people who feel that this class of problem is long-standing in the TEI

Re: placeName types

Saašha Metsärantala — 2013-05-20T17:48:56

Hello!

One other problem occurs when a village or city moves: the administrative 
unit is (almost) the same, the inhabitants are (almost) the same and the 
name of the village (or city) is often the same, but the coordinates are 
not. As a recent example, you may want to check
http://www.wikidata.org/wiki/Q83232 where about 15,000 to 20,000 persons 
moved for the sake of mining. If you point your browser to
http://www.geonames.org/advanced-search.html?q=kiruna&featureClass=P&continentCode=EU 
and choose the first name there, you will find a map: The Swedish word 
"centrum" means "the center of the city" ... (or DOWNtown).

Sometimes, a whole village moves a long way (maybe to another 
administrative country) to avoid war or other catastrophes.

During the move (which may take weeks) we may also want to point to each 
of these places separately. How to encode (and / or update) such cases 
seems a bit unclear. Linking to semantic web resources may be useful, but 
requires an awareness of the (absence of) upd

Digital.Humanities< at >Oxford Summer School 2013

James Cummings — 2013-05-20T15:26:39

Only a few weeks left to book!

Places for this year's Digital.Humanities< at >Oxford Summer School 
are filling up already, so book your place soon!  Visit 
http://digital.humanities.ox.ac.uk/dhoxss/2013/ for more 
information. If you are awaiting the results of local funding and 
want to to see whether your chosen workshop is almost full, email 
courses< at >it.ox.ac.uk to find out!

====
The Digital.Humanities< at >Oxford Summer School (DHOxSS) is an annual 
event for anyone working in the Digital Humanities. This year's 
Summer School will be held on 8 - 12 July, at the University of 
Oxford. If you are a researcher, project manager, research 
assistant, or student of the Humanities, this is an opportunity 
for you to learn about the tools and methodology of digital 
humanities, and to make contact with others in your field. You 
will be introduced to topics spanning from creating, managing, 
analysing, modelling, visualizing, to publication of digital data 
for the Humanities. Visit 
http://digital.humanities.ox.ac.uk

Re: Poetess Archive Question

Paul F. Schaffner — 2013-05-20T15:10:56

I agree that this is a question of < at >rend or < at >rendition of
quotations and/or quoted (direct) speech. As usual, it is up
to you how much to preserve as literal text and how much to
encode as 'meta' information. Using literals can get very
messy, or impossible (especially if you are not encoding
line breaks); but encoding as < at >rend can also get a bit hairy,
since there are several competing and overlapping schemes
of punctuation as one moves from the 17th through the 18th
century, and several ways in which printers can fail to abide
by their scheme. Among the commonest in the 18th century is

But sometimes the opening mark is missing; and sometimes
there is a closing mark, and sometimes the marginal marks
do not quite align, and sometimes there are embedded
non-quoted phrases ("he said loudly") that may be variously
marked off, and sometimes there are quotes within quotes,
and sometimes there is a typeface toggle (italic/roman, etc.)
to contend with as well.

So an accurate record of the rendering, if that is wh

Re: Poetess Archive Question

James Cummings — 2013-05-20T14:58:45

 > "for your country." It was a

I would say under <quotation> is where I would expect to read it.

If one were concerned about exactly where the quotation marks 
fell, then these are really a form of rendition of the quotation 
as indicated by line breaks.  I'd have marked it as:

<q rend="linePrefacingMarks">Ask <lb/>not what your country can 
do <lb/>for you but what you can do <lb/>for your country.</q>

(or some other rend value) As far as I'm aware this is just a 
quite standard way of indicating a running blockquote so it part 
of the original rendition of the quote as a whole.

My tuppence,
-James