open-editions / corpus-joyce-ulysses-tei

James Joyce's novel Ulysses in TEI XML. Work-in-progress.
20 stars 17 forks source link

Adding <div> and @type within episodes #11

Open c-forster opened 7 years ago

c-forster commented 7 years ago

There are <div>s around the sections. Should they have a type?

My naive guess would be "yes," and something as straightforward as "div type='section'" might suffice. But before over-hastily concluding so, I wonder about the vocabulary for types in Ulysses overall, particularly around two questions:

yellwork commented 7 years ago

Thanks, Chris. My initial sense was that we’d maintain the eighteen distinct files (in part because I want to pull all this rich markup into the eighteen episode-file version of Ulysses: A Digital Critical and Synoptic Edition in development elsewhere). But I don’t see that conflicting with the double goal of making available all eighteen episodes in a single XML file and all eighteen episodes as a folder of eighteen episode-specific XML files with episode-specific <teiHeader>. In the former case, <div type="chapter [episode?]" n="1">, or some such encoding, would be perfect.

Here’s the situation (and some ideas) for the episodes:

c-forster commented 7 years ago

OK, so I think <div type='section'> makes sense here just to describe the sections; it will help avoid ambiguity with other types of div should, for instance, the episodes be combined together. The others you point to make some sense--though would you want each question/answer pair in "Ithaca" to be a section? Or even the 8 "sentences" of "Penelope." The div I can see... it's whether they're all of the same type.

‘Aeolus’: We have used

to mark the beginning of each new crosshead or headline (currently encoded as ) down to the end of the section of initial-style narrative that immediately follows it.

Of course; makes sense. (Though I could imagine other ways of handling "Aeolus."

‘Wandering Rocks’: Would we be able to use

(or something else) to thread the various interpolations back to their ‘parent’
, where appropriate? I like the idea of making explicit in the markup some of the internal connections and reiterations of the book. (In passing, I think we first put
into the markup to facilitate the use of .)

I think marking this up would be a good idea--but I think it would require a different strategy. The divs are not granular enough (you want to be able to the interpolated/reiterated text, not just the section from which it came, right? Would something like this work, using <spans> and refs?

<lb n="100201"/>raised his cap abruptly: <span xml:id="l100201">the young woman 
abruptly bent and with slow care
<lb n="100202"/>detached from her light skirt a clinging twig.</span>

And, later in the same file:

<p><lb n="100440"/><ref target="#l100201">The young woman with slow care detached from
 her light skirt a 
<lb n="100441"/>clinging twig.</ref>

There is, in "Wandering Rocks" a clear "original," and "reference" structure I think. It would be nice if were bidirectional (that the source pointed to where it was later interpolated), but off the top of my head I can't think of an obvious and semantic way of doing it. (Simply adding more ptrs/links--as folks do when they make HTML footnotes, with links back to the text seems wrong here.) I'd like to hear others thoughts on the best approach, but once there is consensus I'm happy to hop in and add the appropriate markup. (Note also, I've just created a rather weak id for the source span using the line number... I don't think that's a very good idea.)

‘Cyclops’: I like the idea of using

to separate the discrete parodies in the episode; would they all need to be situated within a larger
consisting of the Nameless One’s narration?

I don't think you'd need the larger containing; the outstanding question, for me, is how to describe those divs; are they sections too?

‘Sirens’: The overture?

Maybe... is it too weird to imagine the overture as a head, within the larger containing chapter?

‘Oxen of the Sun’: I’m a bit cautious here. How broad is the consensus on just where the breaks occur in Joyce’s self-described ‘nineparted episode without divisions’? I wouldn’t want us to force the matter. Sarah Davison has a great piece [link] on the issues raised here.

OK; makes sense. I was thinking of the Gifford breakdown, but looking over the link makes me a little cautious. (And yet, it would nice to have some markup to reflect the change over the episode).

In any case, I suspect many of these questions are separate issues, and the chief issue is, if it makes sense to provide for the divs in "Wandering Rocks," what type should it be? section seems reasonable.

yellwork commented 7 years ago

I think this is where we make the existing scholarship do the work for us. “section” is a critical standard for the units in “Wandering Rocks”; just as “sentence” is for “Penelope.” (What else? what others?) Given the more distant analysis on the novel that this encoding will facilitate, I think it’s important that we separate out the different types (or @type) of <div> we encounter in the corpus. So I would say, no, each question/answer pair in “Ithaca” is not a section. Neither are the eight “sentences” of “Penelope” or the parodies of “Cyclops.” (Quite what the question/answer pairs in “Ithaca” are, in terms of <div type="???">, I don’t know yet.

I’m agnostic on the current <div> encoding of “Aeolus.” Would love to hear ideas for an alternative. The problem with <div>, so far as I can see, is the fact that the initial style running through the whole episode is so clearly, as David Pumpkins’s skeletons would put it, “Its own thing!” The use of <div> chops into segments or units something that is not segmented.

An encoding of the original/reference, spur/reprise, signal/echo structure of “Wandering Rocks” could also be applied to the overture of “Sirens” and the rest of the episode (also preferably bidirectionally).

yellwork commented 7 years ago

I replaced the section <div> with <div type="section" n="*"> across the “Wandering Rocks” episode, but I feel like there’s enough interesting material left flagged in this issue not to merit closing it. So I’m retitling the issue and starting a separate issue for the “WR” interpolations.

workshub[bot] commented 2 years ago

@brianmuks started working on this issue via WorksHub.