Help with overlapping structures in critical texts

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

Help with overlapping structures in critical texts

Ian Scott

Hi folks,

 

I have a question about how to mark up overlapping structures. I'm trying to migrate some critical texts from my project's legacy markup to TEI using the parallel segmentation approach. In TEI markup documents are structured as nested <div> elements with the lowest level containing a series of <app> elements and their constituent <rdg> elements:

 

<div>

                <div>

                                <app></app>

                                <app></app>

                                <app></app>

                </div>

</div>

 

Now, in our texts the <app> elements mark up the smallest level of meaningful textual variation. But there are sometimes places where we want to signal that a series of <app> elements are all part of one larger variation pattern. In our legacy markup we did this by way of a "group" attribute. Various <app> elements were related by having a shared "group" number:

 

<div>

                <div>

                                <app xml:id="1" group="#7></app>

                                <app xml:id="2" group="#5 #7"></app>

                </div>

                <div>

                                <app xml:id="3" group="#5 #7"></app>

                                <app xml:id="4" group="#5"></app>

                                <app xml:id="5"></app>

                </div>

</div>

 

Here <app> elements 2, 3  and 4 are identified as belonging to one larger-scale textual variation pattern, group number 5. The elements 1, 2 and 3 are also part of another overlapping variation structure, group number 7. Notice that both “groups” cross the boundary between two lowest-level <div> elements. So I can’t mark this up by nesting <app> elements within some larger element.

 

My question is how to mark this up in TEI. I’ve considered the following possibilities:

 

  • @corresp
    • I could give each of the related <app> elements a “corresp” attribute with a pointer, say, to the first <app> in the structure. This has the advantage of allowing multiple overlapping structures, but it doesn’t seem to fit the semantics of @corresp 
  • @next and @prev
    • Semantically it would make more sense to link the members of each larger variation structure using “next” and “prev” attributes. The disadvantage here is that the variation structure doesn’t have any direct identifier. I can’t point to it directly, but have to infer its presence from the presence of the “next” and “prev” attributes. I also have to follow the trail of “next” and “prev” to find its extent whenever I want to refer to it.
  • milestone
    • I could use a <milestone> element at the beginning and end of each larger variation structure (at the same structural level as <app>). As with @corresp, this feels a bit like an abuse of the semantics of <milestone>, partly because it doesn’t seem intended to mark the end of a structure. I also need to use <milestone> for marking up alternative structuring schemes, so I would rather not use it in two such different ways.
  • link
    • Somewhere in the document header I could place a <linkGrp> with a <link> child element for each of these larger variation patterns. I could just list the ids of the constituent <app> elements in the @target of the <link>. Logically, this seems to be the least confusing option. But it removes the information entirely from the actual text markup. I don’t like the idea of having some text-critical information encoded in the text body and other (essentially equivalent) information encoded in a separate, central place.

 

So none of the options I’ve found so far seems really satisfactory. Am I missing things in the TEI schema that would work? Is there one of the four possibilities above that you think would be best?

 

Thanks,

 

Ian

 

 

 

 

 

Ian W. Scott, PhD

Associate Professor of New Testament

Tyndale Seminary, Toronto, Canada

www.ianwscott.com

 

Paul’s Way of Knowing: Story, Experience and the Spirit (Baker Academic [Mohr Siebeck], 2006)

The Online Critical Pseudepigrapha (SBL, 2006-; pseudepigrapha.org)

 

 

Reply | Threaded
Open this post in threaded view
|

Re: Help with overlapping structures in critical texts

Philipp Vanscheidt-2
Dear Ian,

is it possible to explain the semantics of "groups" or "larger variation
patterns"  a littlebit further or give an example? Maybe @ana would be
an option but I suppose that you want something like @varSeq for <rdg>s.
Wouldn't it make more sense to link the readings themselves to such a
variation pattern instead of the apparatus entries?

Best,
Philipp


Am 2018-06-13 17:08, schrieb Ian Scott:

> Hi folks,
>
> I have a question about how to mark up overlapping structures. I'm
> trying to migrate some critical texts from my project's legacy markup
> to TEI using the parallel segmentation approach. In TEI markup
> documents are structured as nested <div> elements with the lowest
> level containing a series of <app> elements and their constituent
> <rdg> elements:
>
> <div>
>                 <div>
>                                 <app></app>
>                                 <app></app>
>                                 <app></app>
>                 </div>
> </div>
>
> Now, in our texts the <app> elements mark up the smallest level of
> meaningful textual variation. But there are sometimes places where we
> want to signal that a series of <app> elements are all part of one
> larger variation pattern. In our legacy markup we did this by way of a
> "group" attribute. Various <app> elements were related by having a
> shared "group" number:
>
> <div>
>                 <div>
>                                 <app xml:id="1" group="#7></app>
>                                 <app xml:id="2" group="#5 #7"></app>
>                 </div>
>                 <div>
>                                 <app xml:id="3" group="#5 #7"></app>
>                                 <app xml:id="4" group="#5"></app>
>                                 <app xml:id="5"></app>
>                 </div>
> </div>
>
> Here <app> elements 2, 3  and 4 are identified as belonging to one
> larger-scale textual variation pattern, group number 5. The elements
> 1, 2 and 3 are also part of another overlapping variation structure,
> group number 7. Notice that both "groups" cross the boundary between
> two lowest-level <div> elements. So I can't mark this up by nesting
> <app> elements within some larger element.
>
> My question is how to mark this up in TEI. I've considered the
> following possibilities:
>
>
>   *   @corresp
>      *   I could give each of the related <app> elements a "corresp"
> attribute with a pointer, say, to the first <app> in the structure.
> This has the advantage of allowing multiple overlapping structures,
> but it doesn't seem to fit the semantics of @corresp
>   *   @next and @prev
>      *   Semantically it would make more sense to link the members of
> each larger variation structure using "next" and "prev" attributes.
> The disadvantage here is that the variation structure doesn't have any
> direct identifier. I can't point to it directly, but have to infer its
> presence from the presence of the "next" and "prev" attributes. I also
> have to follow the trail of "next" and "prev" to find its extent
> whenever I want to refer to it.
>   *   milestone
>      *   I could use a <milestone> element at the beginning and end of
> each larger variation structure (at the same structural level as
> <app>). As with @corresp, this feels a bit like an abuse of the
> semantics of <milestone>, partly because it doesn't seem intended to
> mark the end of a structure. I also need to use <milestone> for
> marking up alternative structuring schemes, so I would rather not use
> it in two such different ways.
>   *   link
>      *   Somewhere in the document header I could place a <linkGrp>
> with a <link> child element for each of these larger variation
> patterns. I could just list the ids of the constituent <app> elements
> in the @target of the <link>. Logically, this seems to be the least
> confusing option. But it removes the information entirely from the
> actual text markup. I don't like the idea of having some text-critical
> information encoded in the text body and other (essentially
> equivalent) information encoded in a separate, central place.
>
> So none of the options I've found so far seems really satisfactory. Am
> I missing things in the TEI schema that would work? Is there one of
> the four possibilities above that you think would be best?
>
> Thanks,
>
> Ian
>
>
>
>
>
> Ian W. Scott, PhD
> Associate Professor of New Testament
> Tyndale Seminary, Toronto, Canada
> www.ianwscott.com<http://www.ianwscott.com>
>
> Paul's Way of Knowing: Story, Experience and the Spirit (Baker
> Academic [Mohr Siebeck], 2006)
> The Online Critical Pseudepigrapha (SBL, 2006-;
> pseudepigrapha.org<https://pseudepigrapha.org/>)

--
Technische Universität Darmstadt, Institut für Sprach- und
Literaturwissenschaft
Dolivostraße 15, 64293 Darmstadt
Tel.: 06151/1657405
[hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Help with overlapping structures in critical texts

Gerrit Brüning
Dear Ian and Philipp,

As Philipp suggests, my first guess would be that @ana is the appropriate solution here.
If I get you right, "textual variation pattern" comes close to some kind of classification.

I am not sure about the meaning of @varSeq: Until now I thought that @varSeq only gives local information about the chronological order, so that a "2" does not necessarily correspond to other "2"s elsewhere in the document, does it?

Best,

Gerrit

---
Dr. Gerrit Brüning
Freies Deutsches Hochstift | Historisch-kritische Edition von Goethes Faust | beta.faustedition.net
Goethe-Universität Frankfurt am Main | Institut für deutsche Literatur und ihre Didaktik | IG-Hochhaus 1.155

> -----Original Message-----
> From: TEI (Text Encoding Initiative) public discussion list [mailto:TEI-
> [hidden email]] On Behalf Of Philipp Vanscheidt
> Sent: Wednesday, June 13, 2018 5:22 PM
> To: [hidden email]
> Subject: Re: Help with overlapping structures in critical texts
>
> Dear Ian,
>
> is it possible to explain the semantics of "groups" or "larger variation patterns"
> a littlebit further or give an example? Maybe @ana would be an option but I
> suppose that you want something like @varSeq for <rdg>s.
> Wouldn't it make more sense to link the readings themselves to such a
> variation pattern instead of the apparatus entries?
>
> Best,
> Philipp
>
>
> Am 2018-06-13 17:08, schrieb Ian Scott:
> > Hi folks,
> >
> > I have a question about how to mark up overlapping structures. I'm
> > trying to migrate some critical texts from my project's legacy markup
> > to TEI using the parallel segmentation approach. In TEI markup
> > documents are structured as nested <div> elements with the lowest
> > level containing a series of <app> elements and their constituent
> > <rdg> elements:
> >
> > <div>
> >                 <div>
> >                                 <app></app>
> >                                 <app></app>
> >                                 <app></app>
> >                 </div>
> > </div>
> >
> > Now, in our texts the <app> elements mark up the smallest level of
> > meaningful textual variation. But there are sometimes places where we
> > want to signal that a series of <app> elements are all part of one
> > larger variation pattern. In our legacy markup we did this by way of a
> > "group" attribute. Various <app> elements were related by having a
> > shared "group" number:
> >
> > <div>
> >                 <div>
> >                                 <app xml:id="1" group="#7></app>
> >                                 <app xml:id="2" group="#5 #7"></app>
> >                 </div>
> >                 <div>
> >                                 <app xml:id="3" group="#5 #7"></app>
> >                                 <app xml:id="4" group="#5"></app>
> >                                 <app xml:id="5"></app>
> >                 </div>
> > </div>
> >
> > Here <app> elements 2, 3  and 4 are identified as belonging to one
> > larger-scale textual variation pattern, group number 5. The elements
> > 1, 2 and 3 are also part of another overlapping variation structure,
> > group number 7. Notice that both "groups" cross the boundary between
> > two lowest-level <div> elements. So I can't mark this up by nesting
> > <app> elements within some larger element.
> >
> > My question is how to mark this up in TEI. I've considered the
> > following possibilities:
> >
> >
> >   *   @corresp
> >      *   I could give each of the related <app> elements a "corresp"
> > attribute with a pointer, say, to the first <app> in the structure.
> > This has the advantage of allowing multiple overlapping structures,
> > but it doesn't seem to fit the semantics of @corresp
> >   *   @next and @prev
> >      *   Semantically it would make more sense to link the members of
> > each larger variation structure using "next" and "prev" attributes.
> > The disadvantage here is that the variation structure doesn't have any
> > direct identifier. I can't point to it directly, but have to infer its
> > presence from the presence of the "next" and "prev" attributes. I also
> > have to follow the trail of "next" and "prev" to find its extent
> > whenever I want to refer to it.
> >   *   milestone
> >      *   I could use a <milestone> element at the beginning and end of
> > each larger variation structure (at the same structural level as
> > <app>). As with @corresp, this feels a bit like an abuse of the
> > semantics of <milestone>, partly because it doesn't seem intended to
> > mark the end of a structure. I also need to use <milestone> for
> > marking up alternative structuring schemes, so I would rather not use
> > it in two such different ways.
> >   *   link
> >      *   Somewhere in the document header I could place a <linkGrp>
> > with a <link> child element for each of these larger variation
> > patterns. I could just list the ids of the constituent <app> elements
> > in the @target of the <link>. Logically, this seems to be the least
> > confusing option. But it removes the information entirely from the
> > actual text markup. I don't like the idea of having some text-critical
> > information encoded in the text body and other (essentially
> > equivalent) information encoded in a separate, central place.
> >
> > So none of the options I've found so far seems really satisfactory. Am
> > I missing things in the TEI schema that would work? Is there one of
> > the four possibilities above that you think would be best?
> >
> > Thanks,
> >
> > Ian
> >
> >
> >
> >
> >
> > Ian W. Scott, PhD
> > Associate Professor of New Testament
> > Tyndale Seminary, Toronto, Canada
> > www.ianwscott.com<http://www.ianwscott.com>
> >
> > Paul's Way of Knowing: Story, Experience and the Spirit (Baker
> > Academic [Mohr Siebeck], 2006) The Online Critical Pseudepigrapha
> > (SBL, 2006-;
> > pseudepigrapha.org<https://pseudepigrapha.org/>)
>
> --
> Technische Universität Darmstadt, Institut für Sprach- und
> Literaturwissenschaft Dolivostraße 15, 64293 Darmstadt
> Tel.: 06151/1657405
> [hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Help with overlapping structures in critical texts

Philipp Vanscheidt-2
Dear Gerrit,

I suppose that your interpretations of @varSeq accords with mine. I
first thought that this would match with one possible meaning of "larger
variation patterns" but that seems to be a misunderstanding.

Best,
Philipp

Am 2018-06-14 09:54, schrieb Gerrit Brüning:

> Dear Ian and Philipp,
>
> As Philipp suggests, my first guess would be that @ana is the
> appropriate solution here.
> If I get you right, "textual variation pattern" comes close to some
> kind of classification.
>
> I am not sure about the meaning of @varSeq: Until now I thought that
> @varSeq only gives local information about the chronological order, so
> that a "2" does not necessarily correspond to other "2"s elsewhere in
> the document, does it?
>
> Best,
>
> Gerrit
>
> ---
> Dr. Gerrit Brüning
> Freies Deutsches Hochstift | Historisch-kritische Edition von Goethes
> Faust | beta.faustedition.net
> Goethe-Universität Frankfurt am Main | Institut für deutsche Literatur
> und ihre Didaktik | IG-Hochhaus 1.155
>
>> -----Original Message-----
>> From: TEI (Text Encoding Initiative) public discussion list
>> [mailto:TEI-
>> [hidden email]] On Behalf Of Philipp Vanscheidt
>> Sent: Wednesday, June 13, 2018 5:22 PM
>> To: [hidden email]
>> Subject: Re: Help with overlapping structures in critical texts
>>
>> Dear Ian,
>>
>> is it possible to explain the semantics of "groups" or "larger
>> variation patterns"
>> a littlebit further or give an example? Maybe @ana would be an option
>> but I
>> suppose that you want something like @varSeq for <rdg>s.
>> Wouldn't it make more sense to link the readings themselves to such a
>> variation pattern instead of the apparatus entries?
>>
>> Best,
>> Philipp
>>
>>
>> Am 2018-06-13 17:08, schrieb Ian Scott:
>> > Hi folks,
>> >
>> > I have a question about how to mark up overlapping structures. I'm
>> > trying to migrate some critical texts from my project's legacy markup
>> > to TEI using the parallel segmentation approach. In TEI markup
>> > documents are structured as nested <div> elements with the lowest
>> > level containing a series of <app> elements and their constituent
>> > <rdg> elements:
>> >
>> > <div>
>> >                 <div>
>> >                                 <app></app>
>> >                                 <app></app>
>> >                                 <app></app>
>> >                 </div>
>> > </div>
>> >
>> > Now, in our texts the <app> elements mark up the smallest level of
>> > meaningful textual variation. But there are sometimes places where we
>> > want to signal that a series of <app> elements are all part of one
>> > larger variation pattern. In our legacy markup we did this by way of a
>> > "group" attribute. Various <app> elements were related by having a
>> > shared "group" number:
>> >
>> > <div>
>> >                 <div>
>> >                                 <app xml:id="1" group="#7></app>
>> >                                 <app xml:id="2" group="#5 #7"></app>
>> >                 </div>
>> >                 <div>
>> >                                 <app xml:id="3" group="#5 #7"></app>
>> >                                 <app xml:id="4" group="#5"></app>
>> >                                 <app xml:id="5"></app>
>> >                 </div>
>> > </div>
>> >
>> > Here <app> elements 2, 3  and 4 are identified as belonging to one
>> > larger-scale textual variation pattern, group number 5. The elements
>> > 1, 2 and 3 are also part of another overlapping variation structure,
>> > group number 7. Notice that both "groups" cross the boundary between
>> > two lowest-level <div> elements. So I can't mark this up by nesting
>> > <app> elements within some larger element.
>> >
>> > My question is how to mark this up in TEI. I've considered the
>> > following possibilities:
>> >
>> >
>> >   *   @corresp
>> >      *   I could give each of the related <app> elements a "corresp"
>> > attribute with a pointer, say, to the first <app> in the structure.
>> > This has the advantage of allowing multiple overlapping structures,
>> > but it doesn't seem to fit the semantics of @corresp
>> >   *   @next and @prev
>> >      *   Semantically it would make more sense to link the members of
>> > each larger variation structure using "next" and "prev" attributes.
>> > The disadvantage here is that the variation structure doesn't have any
>> > direct identifier. I can't point to it directly, but have to infer its
>> > presence from the presence of the "next" and "prev" attributes. I also
>> > have to follow the trail of "next" and "prev" to find its extent
>> > whenever I want to refer to it.
>> >   *   milestone
>> >      *   I could use a <milestone> element at the beginning and end of
>> > each larger variation structure (at the same structural level as
>> > <app>). As with @corresp, this feels a bit like an abuse of the
>> > semantics of <milestone>, partly because it doesn't seem intended to
>> > mark the end of a structure. I also need to use <milestone> for
>> > marking up alternative structuring schemes, so I would rather not use
>> > it in two such different ways.
>> >   *   link
>> >      *   Somewhere in the document header I could place a <linkGrp>
>> > with a <link> child element for each of these larger variation
>> > patterns. I could just list the ids of the constituent <app> elements
>> > in the @target of the <link>. Logically, this seems to be the least
>> > confusing option. But it removes the information entirely from the
>> > actual text markup. I don't like the idea of having some text-critical
>> > information encoded in the text body and other (essentially
>> > equivalent) information encoded in a separate, central place.
>> >
>> > So none of the options I've found so far seems really satisfactory. Am
>> > I missing things in the TEI schema that would work? Is there one of
>> > the four possibilities above that you think would be best?
>> >
>> > Thanks,
>> >
>> > Ian
>> >
>> >
>> >
>> >
>> >
>> > Ian W. Scott, PhD
>> > Associate Professor of New Testament
>> > Tyndale Seminary, Toronto, Canada
>> > www.ianwscott.com<http://www.ianwscott.com>
>> >
>> > Paul's Way of Knowing: Story, Experience and the Spirit (Baker
>> > Academic [Mohr Siebeck], 2006) The Online Critical Pseudepigrapha
>> > (SBL, 2006-;
>> > pseudepigrapha.org<https://pseudepigrapha.org/>)
>>
>> --
>> Technische Universität Darmstadt, Institut für Sprach- und
>> Literaturwissenschaft Dolivostraße 15, 64293 Darmstadt
>> Tel.: 06151/1657405
>> [hidden email]

--
Technische Universität Darmstadt, Institut für Sprach- und
Literaturwissenschaft
Dolivostraße 15, 64293 Darmstadt
Tel.: 06151/1657405
[hidden email]
Reply | Threaded
Open this post in threaded view
|

Re: Help with overlapping structures in critical texts

Elena Spadini
Thanks for this interesting question!

Wouldn't it make more sense to link the readings themselves to such a 
variation pattern instead of the apparatus entries?
I was thinking the same, and make use of <rdgGrp>, but this would not avoid overlaps ..

Elena

Il giorno gio 14 giu 2018 alle ore 10:13 Philipp Vanscheidt <[hidden email]> ha scritto:
Dear Gerrit,

I suppose that your interpretations of @varSeq accords with mine. I
first thought that this would match with one possible meaning of "larger
variation patterns" but that seems to be a misunderstanding.

Best,
Philipp

Am 2018-06-14 09:54, schrieb Gerrit Brüning:
> Dear Ian and Philipp,
>
> As Philipp suggests, my first guess would be that @ana is the
> appropriate solution here.
> If I get you right, "textual variation pattern" comes close to some
> kind of classification.
>
> I am not sure about the meaning of @varSeq: Until now I thought that
> @varSeq only gives local information about the chronological order, so
> that a "2" does not necessarily correspond to other "2"s elsewhere in
> the document, does it?
>
> Best,
>
> Gerrit
>
> ---
> Dr. Gerrit Brüning
> Freies Deutsches Hochstift | Historisch-kritische Edition von Goethes
> Faust | beta.faustedition.net
> Goethe-Universität Frankfurt am Main | Institut für deutsche Literatur
> und ihre Didaktik | IG-Hochhaus 1.155
>
>> -----Original Message-----
>> From: TEI (Text Encoding Initiative) public discussion list
>> [mailto:[hidden email]
>> [hidden email]] On Behalf Of Philipp Vanscheidt
>> Sent: Wednesday, June 13, 2018 5:22 PM
>> To: [hidden email]
>> Subject: Re: Help with overlapping structures in critical texts
>>
>> Dear Ian,
>>
>> is it possible to explain the semantics of "groups" or "larger
>> variation patterns"
>> a littlebit further or give an example? Maybe @ana would be an option
>> but I
>> suppose that you want something like @varSeq for <rdg>s.
>> Wouldn't it make more sense to link the readings themselves to such a
>> variation pattern instead of the apparatus entries?
>>
>> Best,
>> Philipp
>>
>>
>> Am 2018-06-13 17:08, schrieb Ian Scott:
>> > Hi folks,
>> >
>> > I have a question about how to mark up overlapping structures. I'm
>> > trying to migrate some critical texts from my project's legacy markup
>> > to TEI using the parallel segmentation approach. In TEI markup
>> > documents are structured as nested <div> elements with the lowest
>> > level containing a series of <app> elements and their constituent
>> > <rdg> elements:
>> >
>> > <div>
>> >                 <div>
>> >                                 <app></app>
>> >                                 <app></app>
>> >                                 <app></app>
>> >                 </div>
>> > </div>
>> >
>> > Now, in our texts the <app> elements mark up the smallest level of
>> > meaningful textual variation. But there are sometimes places where we
>> > want to signal that a series of <app> elements are all part of one
>> > larger variation pattern. In our legacy markup we did this by way of a
>> > "group" attribute. Various <app> elements were related by having a
>> > shared "group" number:
>> >
>> > <div>
>> >                 <div>
>> >                                 <app xml:id="1" group="#7></app>
>> >                                 <app xml:id="2" group="#5 #7"></app>
>> >                 </div>
>> >                 <div>
>> >                                 <app xml:id="3" group="#5 #7"></app>
>> >                                 <app xml:id="4" group="#5"></app>
>> >                                 <app xml:id="5"></app>
>> >                 </div>
>> > </div>
>> >
>> > Here <app> elements 2, 3  and 4 are identified as belonging to one
>> > larger-scale textual variation pattern, group number 5. The elements
>> > 1, 2 and 3 are also part of another overlapping variation structure,
>> > group number 7. Notice that both "groups" cross the boundary between
>> > two lowest-level <div> elements. So I can't mark this up by nesting
>> > <app> elements within some larger element.
>> >
>> > My question is how to mark this up in TEI. I've considered the
>> > following possibilities:
>> >
>> >
>> >   *   @corresp
>> >      *   I could give each of the related <app> elements a "corresp"
>> > attribute with a pointer, say, to the first <app> in the structure.
>> > This has the advantage of allowing multiple overlapping structures,
>> > but it doesn't seem to fit the semantics of @corresp
>> >   *   @next and @prev
>> >      *   Semantically it would make more sense to link the members of
>> > each larger variation structure using "next" and "prev" attributes.
>> > The disadvantage here is that the variation structure doesn't have any
>> > direct identifier. I can't point to it directly, but have to infer its
>> > presence from the presence of the "next" and "prev" attributes. I also
>> > have to follow the trail of "next" and "prev" to find its extent
>> > whenever I want to refer to it.
>> >   *   milestone
>> >      *   I could use a <milestone> element at the beginning and end of
>> > each larger variation structure (at the same structural level as
>> > <app>). As with @corresp, this feels a bit like an abuse of the
>> > semantics of <milestone>, partly because it doesn't seem intended to
>> > mark the end of a structure. I also need to use <milestone> for
>> > marking up alternative structuring schemes, so I would rather not use
>> > it in two such different ways.
>> >   *   link
>> >      *   Somewhere in the document header I could place a <linkGrp>
>> > with a <link> child element for each of these larger variation
>> > patterns. I could just list the ids of the constituent <app> elements
>> > in the @target of the <link>. Logically, this seems to be the least
>> > confusing option. But it removes the information entirely from the
>> > actual text markup. I don't like the idea of having some text-critical
>> > information encoded in the text body and other (essentially
>> > equivalent) information encoded in a separate, central place.
>> >
>> > So none of the options I've found so far seems really satisfactory. Am
>> > I missing things in the TEI schema that would work? Is there one of
>> > the four possibilities above that you think would be best?
>> >
>> > Thanks,
>> >
>> > Ian
>> >
>> >
>> >
>> >
>> >
>> > Ian W. Scott, PhD
>> > Associate Professor of New Testament
>> > Tyndale Seminary, Toronto, Canada
>> > www.ianwscott.com<http://www.ianwscott.com>
>> >
>> > Paul's Way of Knowing: Story, Experience and the Spirit (Baker
>> > Academic [Mohr Siebeck], 2006) The Online Critical Pseudepigrapha
>> > (SBL, 2006-;
>> > pseudepigrapha.org<https://pseudepigrapha.org/>)
>>
>> --
>> Technische Universität Darmstadt, Institut für Sprach- und
>> Literaturwissenschaft Dolivostraße 15, 64293 Darmstadt
>> Tel.: 06151/1657405
>> [hidden email]

--
Technische Universität Darmstadt, Institut für Sprach- und
Literaturwissenschaft
Dolivostraße 15, 64293 Darmstadt
Tel.: 06151/1657405
[hidden email]


--
Reply | Threaded
Open this post in threaded view
|

Re: Help with overlapping structures in critical texts

Ian Scott

Hi everyone,

 

Thanks for your thoughts so far. Yesterday I sent Philipp an example of the markup I’m trying to convert, but forgot to send it to the list as well. Here it is:

 

        <div number="0">
          <unit id="1" group="1" parallel="">
            <reading option="0" mss="P V Brock ">Διαθήκη </reading>
            <reading option="1" mss="S ">Διαταξης </reading>
            <reading option="2" mss="Kraft ">Διάταξις</reading>
            <reading option="3" mss="(Slavonic) "></reading>
          </unit>
          <unit id="2" group="1" parallel="">
            <reading option="0" mss="P Brock (Slavonic) "/>
            <reading option="1" mss="S V Kraft ">τοῦ </reading>
          </unit>
          <unit id="3" group="1" parallel="">
            <reading option="0" mss="P S Kraft Brock (Slavonic) "/>
            <reading option="1" mss="V ">ἀμέμπτου καὶ πολυάθλου καὶ μακαρίου </reading>
          </unit>
          <unit id="4" group="1" parallel="">
            <reading option="0" mss="P S V Kraft Brock ">Ἰώβ</reading>
            <reading option="1" mss="(Slavonic) "></reading>
          </unit>
        </div>

 

The <unit> elements here correspond to <app> elements in TEI. In these four units we mark up omissions and substitutions among the Greek manuscripts of the document. But the whole stretch of text is omitted by the Slavonic tradition. I want to encode the more granular variations among the Greek mss. But I also want to encode the fact that there’s a larger-scale difference here between the Greek and Slavonic traditions. The “group” attribute is the way we’ve been showing that in our legacy markup.

 

In case you’re wondering, we try to mark up textual variants in a neutral way, without assuming a base text or making judgements about the reasons for particular variations. So the basic division of <unit> elements is pretty fine-grained, with a new unit beginning whenever the textual relationships among the witnesses shift.

 

Thanks again for your help,

 

Ian

 

 

_____________________________________

Ian W. Scott, PhD
Associate Professor of New Testament
Tyndale Seminary, Toronto, Canada
www.ianwscott.com


Paul's Way of Knowing: Story Experience and the Spirit (Baker Academic [Mohr Siebeck], 2006)
The Online Critical Pseudepigrapha (SBL, 2006-; pseudepigrapha.org)

 

From: [hidden email]
Sent: June 14, 2018 6:10 AM
To: [hidden email]
Subject: Re: Help with overlapping structures in critical texts

 

Thanks for this interesting question!

Wouldn't it make more sense to link the readings themselves to such a 
variation pattern instead of the apparatus entries?
I was thinking the same, and make use of <rdgGrp>, but this would not avoid overlaps ..

Elena

Il giorno gio 14 giu 2018 alle ore 10:13 Philipp Vanscheidt <[hidden email]> ha scritto:
Dear Gerrit,

I suppose that your interpretations of @varSeq accords with mine. I
first thought that this would match with one possible meaning of "larger
variation patterns" but that seems to be a misunderstanding.

Best,
Philipp

Am 2018-06-14 09:54, schrieb Gerrit Brüning:
> Dear Ian and Philipp,
>
> As Philipp suggests, my first guess would be that @ana is the
> appropriate solution here.
> If I get you right, "textual variation pattern" comes close to some
> kind of classification.
>
> I am not sure about the meaning of @varSeq: Until now I thought that
> @varSeq only gives local information about the chronological order, so
> that a "2" does not necessarily correspond to other "2"s elsewhere in
> the document, does it?
>
> Best,
>
> Gerrit
>
> ---
> Dr. Gerrit Brüning
> Freies Deutsches Hochstift | Historisch-kritische Edition von Goethes
> Faust | beta.faustedition.net
> Goethe-Universität Frankfurt am Main | Institut für deutsche Literatur
> und ihre Didaktik | IG-Hochhaus 1.155
>
>> -----Original Message-----
>> From: TEI (Text Encoding Initiative) public discussion list
>> [mailto:[hidden email]
>> [hidden email]] On Behalf Of Philipp Vanscheidt
>> Sent: Wednesday, June 13, 2018 5:22 PM
>> To: [hidden email]
>> Subject: Re: Help with overlapping structures in critical texts
>>
>> Dear Ian,
>>
>> is it possible to explain the semantics of "groups" or "larger
>> variation patterns"
>> a littlebit further or give an example? Maybe @ana would be an option
>> but I
>> suppose that you want something like @varSeq for <rdg>s.
>> Wouldn't it make more sense to link the readings themselves to such a
>> variation pattern instead of the apparatus entries?
>>
>> Best,
>> Philipp
>>
>>
>> Am 2018-06-13 17:08, schrieb Ian Scott:
>> > Hi folks,
>> >
>> > I have a question about how to mark up overlapping structures. I'm
>> > trying to migrate some critical texts from my project's legacy markup
>> > to TEI using the parallel segmentation approach. In TEI markup
>> > documents are structured as nested <div> elements with the lowest
>> > level containing a series of <app> elements and their constituent
>> > <rdg> elements:
>> >
>> > <div>
>> >                 <div>
>> >                                 <app></app>
>> >                                 <app></app>
>> >                                 <app></app>
>> >                 </div>
>> > </div>
>> >
>> > Now, in our texts the <app> elements mark up the smallest level of
>> > meaningful textual variation. But there are sometimes places where we
>> > want to signal that a series of <app> elements are all part of one
>> > larger variation pattern. In our legacy markup we did this by way of a
>> > "group" attribute. Various <app> elements were related by having a
>> > shared "group" number:
>> >
>> > <div>
>> >                 <div>
>> >                                 <app xml:id="1" group="#7></app>
>> >                                 <app xml:id="2" group="#5 #7"></app>
>> >                 </div>
>> >                 <div>
>> >                                 <app xml:id="3" group="#5 #7"></app>
>> >                                 <app xml:id="4" group="#5"></app>
>> >                                 <app xml:id="5"></app>
>> >                 </div>
>> > </div>
>> >
>> > Here <app> elements 2, 3  and 4 are identified as belonging to one
>> > larger-scale textual variation pattern, group number 5. The elements
>> > 1, 2 and 3 are also part of another overlapping variation structure,
>> > group number 7. Notice that both "groups" cross the boundary between
>> > two lowest-level <div> elements. So I can't mark this up by nesting
>> > <app> elements within some larger element.
>> >
>> > My question is how to mark this up in TEI. I've considered the
>> > following possibilities:
>> >
>> >
>> >   *   @corresp
>> >      *   I could give each of the related <app> elements a "corresp"
>> > attribute with a pointer, say, to the first <app> in the structure.
>> > This has the advantage of allowing multiple overlapping structures,
>> > but it doesn't seem to fit the semantics of @corresp
>> >   *   @next and @prev
>> >      *   Semantically it would make more sense to link the members of
>> > each larger variation structure using "next" and "prev" attributes.
>> > The disadvantage here is that the variation structure doesn't have any
>> > direct identifier. I can't point to it directly, but have to infer its
>> > presence from the presence of the "next" and "prev" attributes. I also
>> > have to follow the trail of "next" and "prev" to find its extent
>> > whenever I want to refer to it.
>> >   *   milestone
>> >      *   I could use a <milestone> element at the beginning and end of
>> > each larger variation structure (at the same structural level as
>> > <app>). As with @corresp, this feels a bit like an abuse of the
>> > semantics of <milestone>, partly because it doesn't seem intended to
>> > mark the end of a structure. I also need to use <milestone> for
>> > marking up alternative structuring schemes, so I would rather not use
>> > it in two such different ways.
>> >   *   link
>> >      *   Somewhere in the document header I could place a <linkGrp>
>> > with a <link> child element for each of these larger variation
>> > patterns. I could just list the ids of the constituent <app> elements
>> > in the @target of the <link>. Logically, this seems to be the least
>> > confusing option. But it removes the information entirely from the
>> > actual text markup. I don't like the idea of having some text-critical
>> > information encoded in the text body and other (essentially
>> > equivalent) information encoded in a separate, central place.
>> >
>> > So none of the options I've found so far seems really satisfactory. Am
>> > I missing things in the TEI schema that would work? Is there one of
>> > the four possibilities above that you think would be best?
>> >
>> > Thanks,
>> >
>> > Ian
>> >
>> >
>> >
>> >
>> >
>> > Ian W. Scott, PhD
>> > Associate Professor of New Testament
>> > Tyndale Seminary, Toronto, Canada
>> > www.ianwscott.com<http://www.ianwscott.com>
>> >
>> > Paul's Way of Knowing: Story, Experience and the Spirit (Baker
>> > Academic [Mohr Siebeck], 2006) The Online Critical Pseudepigrapha
>> > (SBL, 2006-;
>> > pseudepigrapha.org<https://pseudepigrapha.org/>)
>>
>> --
>> Technische Universität Darmstadt, Institut für Sprach- und
>> Literaturwissenschaft Dolivostraße 15, 64293 Darmstadt
>> Tel.: 06151/1657405
>> [hidden email]

--
Technische Universität Darmstadt, Institut für Sprach- und
Literaturwissenschaft
Dolivostraße 15, 64293 Darmstadt
Tel.: 06151/1657405
[hidden email]


--
Reply | Threaded
Open this post in threaded view
|

Re: Help with overlapping structures in critical texts

Ian Scott
In reply to this post by Ian Scott

Thanks Patrick.

 

<seg> does seem to fit the semantics well. I still wonder, though, how to handle overlaps. We would have <seg> entities that crossed <div> boundaries and we would potentially have multiple <seg> entities that overlapped. Do you see a way of solving that problem?

 

Moving our document structure from nested <div>s entirely to milestones would solve the problem of overlapping with <div> boundaries. Is that “done” these days in TEI circles? It seems like the spec expects the primary organizational system to be nested <div>s. And that still wouldn’t solve the problem of overlapping <seg> elements.

 

Hope your week is going really well too.

 

Ian

 

_____________________________________

Ian W. Scott, PhD
Associate Professor of New Testament
Tyndale Seminary, Toronto, Canada
www.ianwscott.com


Paul's Way of Knowing: Story Experience and the Spirit (Baker Academic [Mohr Siebeck], 2006)
The Online Critical Pseudepigrapha (SBL, 2006-; pseudepigrapha.org)

 

From: [hidden email]
Sent: June 14, 2018 10:12 AM
To: [hidden email]; [hidden email]
Subject: Re: Help with overlapping structures in critical texts

 

Ian,

I started to suggest the use of <seg>, with your present group attribute values being held in @type.

Unfortunately, @type on <seg> has datatype teidata.enumerated, which excludes the use of white space, thus denying the potential of @type to carry multiple values.

I suspect the single XML name requirement came about to enable pointing at @type, but given the power of XQuery, it is a legacy of an earlier technology.

Personally I would extend TEI to include an attribute on <seg> with datetype xsd:NMTOKENS, which would capture your legacy data, keeps the critical analysis where seen in the text, and, encourages the use of XQuery with the resulting text.

Hope you are having a great week!

Patrick

PS: Aside to the list: Is it by design that links to the appendices put the reader at the head of each appendix? I have that behavior in both FireFox and Adobe Reader on Ubuntu. Thanks!

 

On 06/13/2018 11:08 AM, Ian Scott wrote:

Hi folks,

 

I have a question about how to mark up overlapping structures. I'm trying to migrate some critical texts from my project's legacy markup to TEI using the parallel segmentation approach. In TEI markup documents are structured as nested <div> elements with the lowest level containing a series of <app> elements and their constituent <rdg> elements:

 

<div>

                <div>

                                <app></app>

                                <app></app>

                                <app></app>

                </div>

</div>

 

Now, in our texts the <app> elements mark up the smallest level of meaningful textual variation. But there are sometimes places where we want to signal that a series of <app> elements are all part of one larger variation pattern. In our legacy markup we did this by way of a "group" attribute. Various <app> elements were related by having a shared "group" number:

 

<div>

                <div>

                                <app xml:id="1" group="#7></app>

                                <app xml:id="2" group="#5 #7"></app>

                </div>

                <div>

                                <app xml:id="3" group="#5 #7"></app>

                                <app xml:id="4" group="#5"></app>

                                <app xml:id="5"></app>

                </div>

</div>

 

Here <app> elements 2, 3  and 4 are identified as belonging to one larger-scale textual variation pattern, group number 5. The elements 1, 2 and 3 are also part of another overlapping variation structure, group number 7. Notice that both “groups” cross the boundary between two lowest-level <div> elements. So I can’t mark this up by nesting <app> elements within some larger element.

 

My question is how to mark this up in TEI. I’ve considered the following possibilities:

 

-          @corresp

o   I could give each of the related <app> elements a “corresp” attribute with a pointer, say, to the first <app> in the structure. This has the advantage of allowing multiple overlapping structures, but it doesn’t seem to fit the semantics of @corresp 

-          @next and @prev

o   Semantically it would make more sense to link the members of each larger variation structure using “next” and “prev” attributes. The disadvantage here is that the variation structure doesn’t have any direct identifier. I can’t point to it directly, but have to infer its presence from the presence of the “next” and “prev” attributes. I also have to follow the trail of “next” and “prev” to find its extent whenever I want to refer to it.

-          milestone

o   I could use a <milestone> element at the beginning and end of each larger variation structure (at the same structural level as <app>). As with @corresp, this feels a bit like an abuse of the semantics of <milestone>, partly because it doesn’t seem intended to mark the end of a structure. I also need to use <milestone> for marking up alternative structuring schemes, so I would rather not use it in two such different ways.

-          link

o   Somewhere in the document header I could place a <linkGrp> with a <link> child element for each of these larger variation patterns. I could just list the ids of the constituent <app> elements in the @target of the <link>. Logically, this seems to be the least confusing option. But it removes the information entirely from the actual text markup. I don’t like the idea of having some text-critical information encoded in the text body and other (essentially equivalent) information encoded in a separate, central place.

 

So none of the options I’ve found so far seems really satisfactory. Am I missing things in the TEI schema that would work? Is there one of the four possibilities above that you think would be best?

 

Thanks,

 

Ian

 

 

 

 

 

Ian W. Scott, PhD

Associate Professor of New Testament

Tyndale Seminary, Toronto, Canada

www.ianwscott.com

 

Paul’s Way of Knowing: Story, Experience and the Spirit (Baker Academic [Mohr Siebeck], 2006)

The Online Critical Pseudepigrapha (SBL, 2006-; pseudepigrapha.org)

 

 



-- 
Patrick Durusau
[hidden email]
Technical Advisory Board, OASIS (TAB)
Editor, OpenDocument Format TC (OASIS), Project Editor ISO/IEC 26300
Co-Editor, ISO/IEC 13250-1, 13250-5 (Topic Maps)
 
Another Word For It (blog): http://tm.durusau.net
Homepage: http://www.durusau.net
Twitter: patrickDurusau