search

laurenmcguigan / DecameronProject

4 stars 7 forks source link

Decameron Project Tasks! #7

Open ebeshero opened 8 years ago

ebeshero commented 8 years ago
RJP43 commented 8 years ago

Research Ideas proposed in class 2/24:
10 days - 10 stories a day
What marks a separation between days of the entire narrative?
What marks a separation between stories of a single day?
What levels of structure matter: stories, paragraphs, sentences?
Where is there overlap?
How are characters associated with places within a structural unit?
What stories do characters appear in?
What stories do places appear in?
What stories do these characters and places overlap?
What stories do characters overlap?
What stories do places overlap?

ebeshero commented 8 years ago

I am going to take a look at the Project Gutenberg file and see if it's formatted differently in a way that helps to signal the nested narratives better than this Obdurodon/Brown file. @ebeshero

ebeshero commented 8 years ago

Decameron team! What is happening with the project? @jlm323 @mmm202 @laurenmcguigan @mmm202

mmm202 commented 8 years ago

@jlm323 @mjb232 @laurenmcguigan Have we picked what each of us can do?

jlm323 commented 8 years ago

We haven't decided who is doing what yet but I think a good place to start is doing the project midterm. Once we've started looking at the Decameron file individually and marking it up, I say we get together and discuss what we've found and start assigning tasks. I think we can draw from the markup each of us does and go from there.

laurenmcguigan commented 8 years ago

Sorry I haven't been able to do much for on this yet, I've been working all week. But I am off tomorrow, so I can get started on the schematron if you guys want me to. Should we all be doing individual ones or should we make one together? perhaps we can each come up with at least one rule and post it here? And I can also try to extract names and places. Can I do that using find and replace? Or should I use XSLT to extract them and turn them into html?

mmm202 commented 8 years ago

I think we need to at least start our markup before Monday and write the schematron. What ever way everyone wants to do it though is fine with me.

RJP43 commented 8 years ago

Hey everyone @mmm202 @mjb232 @jlm323 @laurenmcguigan ,

I want to offer some advice on a workflow for the project, but first I wanted to make sure I understand where the project sits as of right now. Could someone link here what version of the text is being used and if it is the text that is already marked up by the Pitt main students what tags will you be keeping versus getting rid of? Did you collectively decide to work within the TEI and if so do you need help finding tags for what you wish to markup? I ask all of these questions because in order to knock out some schema rules it would be best to have a list of what is being kept, what isn't, and what is being added so then we can come up with the appropriate rules and an XSLT to do an identity transformation (XSLT EXERCISE 1) that will remove the unwanted tags. If you are going with a blank text (like the Project Gutenberg text) then we need to come up with a tag set and knock the schema out first thing so a bunch of us can jump on marking it up. If someone wants to take the lead on this and just answer therse few questions listed out more clearly here I think it would be a good start to getting everyone on the same page....

  1. What text are you using?
  2. Are there tags already in the text that need removed?
  3. What do you plan on marking up to later be queried?
  4. Are you using the TEI? If so do you need help choosing logical tags for what you wish to markup? If not have you come up with element and attribute names to represent what you wish to markup?
RJP43 commented 8 years ago

I agree with @mmm202 getting your markup and a schematron in place is most important and will help answer the rest of the midterm questions. Once you have a clear plan on what is being marked and how it is being marked you will know what it is you want to pull. The questions go hand in hand.

RJP43 commented 8 years ago

@laurenmcguigan by extracting names and places do you mean that those are the two parts of the Pitt Main code you decided on keeping and if so then I would suggest writing an XSLT identity transformation to turn their XML into a little less marked up (and confusing) XML for your use. Then can add markup to grab the layers of the narrative in your new XML using Regex. Great job thinking of ways to use code you know to get the files prepped for additional markup. Once you guys have marked up files we can use XQuery and XSLT to extract the certain aspects of the markup that you want to create visualizations from or create HTML pages with.

jlm323 commented 8 years ago

I think using the Oakland team's Decameron file would be good for this, since I read through the first day and I think we can work with the markup they've already done. However, I think we may need to change the elements slightly to be more accommodating. Looking at their file, do we need to keep the milestone self closing tags? And is that only for TEI? I think it would be good for us to use TEI because it seems more standard/has longevity.

I read the First Day and took some notes on it:

RJP43 commented 8 years ago

@jlm323

In an attempt to decode their code here are some links we might find useful in trying to figure out the markup:

Here is the source site that provided the basic markup they based their revised markup on from Brown.

This is the final project site of the Pitt team, which might be worth reading through their methods. And this is their related GitHub.

My only big concern with using their markup is trying to figure out how the elements line up to what you want to investigate; however, you seem to be having a good handle in understanding most of what they are using structurally to hold the different narratives. I'm not sure I entirely understand the number endings on the divs or the milestones, but if you do then great! Luckily they used TEI so we can refer to the TEI guidelines to try to get some clues as to what they meant with the q elements. This is what I found that may be useful regarding quoted material.

jlm323 commented 8 years ago

@RJP43 The divs seem to be divided like this:

<div1>
    First Day
    <div2>
        Novella 1
        <div3>Author/narrator's commentary</div3>
        <div3>Character's commentary</div3>
    </div2>
    <div2>
        Novella 2
        <div3>Author/narrator's commentary</div3>
        <div3>Character's commentary</div3>
    </div2>
</div1>

I think if we wanted to start with totally fresh markup, it would be fine. We just need to decide on what tags to use as soon as possible. Also, I don't think the milestones are completely necessary because they just seem to be denoting a change in the text. Edit: I think if we chose to use the Oakland markup to look at nested narratives, we may need to create a different element or an additional element so we can actually see how it is nested. Just a thought.

laurenmcguigan commented 8 years ago

So are we keeping the div tags and extracting the quote tags? And what other tags can we keep? Maybe we should just make a list first of the tags we are keeping and the ones we are extracting. The we can write a schema rule that says "don't include 'old tag'" to ensure everything is extracted.

laurenmcguigan commented 8 years ago

Also, it doesn't look to me like names are tagged within the text. Is that something we have to add? I see that they are all listed at the beginning of the file, but not inside the text. Is that right or am I looking at it wrong?

jlm323 commented 8 years ago

@laurenmcguigan Yes, the names and places aren't tagged but are listed at the very beginning with some informational attributes alongside them. We may have to start tagging them. I think we should remove the milestone tags and keep the paragraph tags since they seem to be the original breaks in text. What tags do you think we should keep or remove?

I noticed <pb n="22"/> tags that continue to count up. I'm guessing those are page numbers although I'm not sure.

laurenmcguigan commented 8 years ago

@jlm323 I thought so! And according to XPath, there are 594 different names listed, so we might want to get started on that. But, do we want to tag ALL of the names? Type //person into Xpath and see if you think we should keep all of them. It doesn't seem like they are all super important, but I'm not sure because I haven't read a lot of it. Maybe we could tag just the main characters?

jlm323 commented 8 years ago

@laurenmcguigan I think just tagging the main characters is a good place to start! The main ladies are: Pampinea, Fiametta, Filomena, Emilia, Lauretta, Neifile, & Elisa. And the main men are: Pamfilo, Filostrato, Dioneo.

laurenmcguigan commented 8 years ago

@jlm323 sounds good! Also, I believe pb is page break, so I also think we should keep those. I agree that we don't need the milestone tags. I'm really not quite sure what they are used for. We would also need to add the place names within the text. Is there anything else we might want to add besides people and places?

jlm323 commented 8 years ago

@laurenmcguigan I think that's it for now. I'm going to keep reading through it and see if I can find anything else that might need tagged or changed.

RJP43 commented 8 years ago

@jlm323 oh okay yea that structure makes sense then. So <div1> always holds a day, <div2> always holds a novella, and <div3> always holds the commentary of both narrator/author and other characters. Questions to ask and things to consider when preparing to create an XSLT for identity transform and schematron:

Are the div#s ever out of that sequence of first <div1>, then <div2>, then <div3> and how is the prologue handled that may be different?

Are we sure the <milestone> doesn't mark anything of importance regarding the structure of the frame narrative (I'm not entirely sure what you (@jlm323) mean by " denoting a change in the text")?

In TEI you will want to determine whether you want to use <placeName> versus <place> as well as <person> versus <persName>. TEI documentation on all this can be found here. You might also try just searching each of these element names in the TEI Guidelines since that interface can be somewhat easier to navigate quickly.

Might move the personography file outside of the main XML and into its own file with 2 simple XSLT identity transforms one that sets up the new personography file (and placeography... they can be in the same file... like the Mitford Site index and @ebeshero or I can help you with setting up a basic structure for that) and the new XML with only the certain elements you wish to keep included.

@laurenmcguigan I think it is wise to make sure you keep paragraph <p> and page breaks <pb n="#"/> but you might want to write a rule saying when paragraphs appear (always inside a div of certain kind?) and that the page break numbers are in sequence.

Let's figure out what the different attributes are on <q> and how the significance of each one might be useful to us when we go to query the information. We might also want to consider what we could add to the <q> elements.... is there a way right now that we can identify through xpath when a main character versus minor character is speaking, or who is having the most conversation with who else? I don't know if either of these are relevant but when you start considering things to add this might be something and then the research question could take a twist on gender or character types.

@jlm323 You might start assigning people to specific tasks ... like @laurenmcguigan can work on the schematron maybe creating rules for some of the things we have listed above and more as they come up. @laurenmcguigan perhaps start an issue with only suggestions for schematron rules if that is what you guys decide is your task. You could assign someone else on the task of separating the two documents and removing milestone markers by doing the two XSLT transforms (whoever is doing that feel free to check out the Nelson site index to see how information for different things could be organized). And you might get someone else started on going through a specific day or number of days to mark people and places (only after you make the decision on which TEI tags to use of course.) Starting a code book sometimes is helpful to organize the whole team on decisions made about the way it has been decided to mark something (again if you need a references of what that might look like the Nelson Codebook and the Mitford Codebook would be two options to gain the general idea.... but basically just state the elements and what attributes correspond and what the elements and attributes are for and this could be done in a word doc or whatever ... I think a GitHub wiki easiest since you can literally rep. code because wikis are written in markdown just like issue comments). Having a codebook helps when you do a lot of the methodology write up for the project for the main site too.

ebeshero commented 8 years ago

@jlm323 @RJP43 The use of div1, div2, div3, etc is really not favored in TEI community (even though the Guidelines sort of support them). It's an old TEI method and the problem with it was if the divs were not actually nested correctly, the numbers would be meaningless--and worse, it's often unclear what the numbers mean to people visiting your project and viewing your code (as you found here). So here's something you could change right away. Make those <div> elements be meaningful by converting them into something like: <div type="day"> <div type="story"> etc.

Here's some reading from the TEI Guidelines Chapter 4: Default Text Structure, with lots of examples: http://www.tei-c.org/release/doc/tei-p5-doc/en/html/DS.html#DSDIV3

Now, here's a really serious question: I wonder if milestone elements were used in the text to signal some issue with overlapping hierarchy, or the narrative levels being shifted out of order? So, for example, what if the frame narrative interrupts inside an internal story? I don't know because I've not reviewed the Oakland team's code that closely...But <milestone> is certainly not the ideal element to use because there's a TEI element designed specifically for that problem. It's called <floatingText>, and when you see an interruption in the nested layering of stories, you should consider using this: http://www.tei-c.org/release/doc/tei-p5-doc/en/html/DS.html#DSGRPF examples: http://www.tei-c.org/release/doc/tei-p5-doc/en/html/DS.html#DSFLT Hope that helps!

ebeshero commented 8 years ago

@jlm323 @laurenmcguigan @mjb232 @mmm202 @RJP43 Wow! I see that Bocaccio's Decameron is a featured example on the page about the use of the <floatingText> element in the TEI Guidelines. So if it is NOT being used in the Oakland team's code, perhaps our team had best introduce it! http://www.tei-c.org/release/doc/tei-p5-doc/en/html/DS.html#DSFLT

The point of <floatingText> is to handle shifts in the levels of a narrative like the Decameron or the Arabian Nights. These texts present us with a "frame story" on the outside that holds the whole thing together (so the story of the the people gathered together in quarantine to sit out the plague, who entertain each other by sharing stories). The stories within those stories could be said to be nested inside, except that sometimes, someone in the frame level steps in and interrupts. (An interruption could be something shallow like, "Okay, I need a glass of water." or "There's someone knocking on the door--let me go get it..." but more likely it's built into the structure of the storytelling to make things interesting and remind us that what you're reading is delightfully complex, that there are layers in it. So the frame narration might seriously involve someone in the frame level reacting to a story being told or responding in some way to it.

When you represent Decameron online it would be useful to be able to tease out where the frame narration intersects with the internal stories. And for your collection of information about people and places, you'd want to be able to distinguish that too--what places are mentioned "out" in the frame, and what places are inside the internal stories? Just some ideas...

jlm323 commented 8 years ago

Here's an excerpt of one of the stories including some milestones:

<milestone id="p01020022"/>Which matters,
 <pb n="37"/>with many others which are not to be mentioned, our modest and
 sober-minded Jew found by no means to his liking, so that, his
 curiosity being fully satisfied, he was minded to return to Paris; which
 accordingly he did. <milestone id="p01020023"/>There, on his arrival, he was met by Jehannot;
 and the two made great cheer together. Jehannot expected Abraham's
 conversion least of all things, and allowed him some days of rest before
 he asked what he thought of the Holy Father and the cardinals and
 the other courtiers. <milestone id="p01020024"/>To which the Jew forthwith replied:

It seems to just note a shift in the story, not necessarily an interruption in the frame narrative.

If I'm understanding correctly, we would use floating text whenever there is a break in the frame narrative. For instance, 

<floatingText>
When Neifile had brought her story to a close amid the
      commendations
      of all the company, Filomena, at the queen's behest, thus
      began:
</floatingText>

So far, the stories are told uninterrupted besides the return to the main narrative between the stories. Also, I really prefer the <div type="day"> and <div type="story"> tags over the numbered div tags.

As for tasks, @laurenmcguigan has the Schematron. @mjb232 if you could move the personography and placeography into it's own file. And @mmm202 if you could start marking up the people and places for the First Day once we decide on tags for them. I created a GitHub wiki for our codebook.

mjb232 commented 8 years ago

@jlm323 Ill get started on the Personography and Placeography right away.

mjb232 commented 8 years ago

Also, something we might want to think about is whether or not we want to actually mark up the entire text. At this point in the game what we're doing is fine, since all of our mark-up is more or less simple and applies to every story. However, if we want to do deep analysis and comparisons, using the entire text would be impossible for the time we have, so we need to select specific stories. Now, figuring out which stories to use depends on what we are actually trying to accomplish, so we need to start considering what our purpose here is.

One idea is that we could take a few of the main story tellers and compare their stories side-by-side. We could look for reoccurring themes, motifs, imagery, places, characters etc. We could also just do a comparison of the stories contained within a day. Each day has a certain theme and so there are bound to be some consistencies between each of the stories.

Of course we could work with the entire text by simply comparing how many instances of specific things within each story: such as characters, places, and themes. This can be done using RegEx to mark everything up, and then XPath to get our numbers. We would then need to figure out a way to present all this information on the website.

This is just some food for thought while working, and any other ideas are welcome! We just need to find our goal here so that we can start working to achieve said goal.

jlm323 commented 8 years ago

Good thinking! I think we should consider focusing on a few characters and compare their stories. I'd like to warn you that the character Dioneo seems to be a special case. At the end of the First Day, he asks if he can be the last story teller from then on and he asks if he can deviate from the theme of the day. So his stories do not necessarily fit in with the theme of the other stories on a given day.

mjb232 commented 8 years ago

That's interesting actually. I'd like to take a look at Dioneo's stories and see how they match-up to one another, and to the other stories.

mjb232 commented 8 years ago

@ebeshero @RJP43 @jlm323 Also, I'd hate to say it but I'm not sure how to approach the Personography and Placeography. How should I format it, and should I be modifying the tags and attributes? Does anyone know of any material I could read or examples I could use to base ours after? Sorry, but I never really got a firm handle on the prosopography. Tips?

ebeshero commented 8 years ago

@mjb232 No worries--prosopography sounds difficult, but you just want to read something about it and see some examples to get a sense of it. Here's something I wrote up for the Digital Mitford project team: Digital Mitford Explanation of Prosopography

And here's a nice simple working example from our Amadis project--it's kind of similar to the sort of thing you probably want to be doing with Decameron, since these are kind of similar texts: Amadis Prosop File on GitHub

Actually, you may want to download that file from here, the raw view and do some XPath on it or look at it in Outline view so you have a sense of this. It's basically a file full of different kinds of lists. Every entry on each list has a distinct xml:id on it.

@jlm323 @laurenmcguigan I'm not sure you want to actually remove every <person> tag just because you don't think you'll be interested in any but the main characters. But you may want to find a way to locate the characters you care about and flag them somehow--with an attribute. One simple way will be when you have a Prosopography list ready, the thing that Matt will work on. That list will contain distinct @xml:id identifiers for the people and places you're working with, and you'll want to point @ref attributes (with hashtags # + id) on elements in the Decameron text so they refer to your xml:ids in the Prosopography.

laurenmcguigan commented 8 years ago

@ebeshero The personography makes sense, but didn't the oakland team already make one at the beginning or is that something different? Also, while we are working on that we might want to also include in the tag whether they are a main character or a minor character.

laurenmcguigan commented 8 years ago

Also, I think using floating text is a good idea instead of the milestone tags, but I assume they would be placed differently than the milestone tags, we can't just replace them. So, I don't think it would be practical to add floating text to the whole document, but instead just a few stories like Matt suggested. Although do we really need to add floating text tags for the purpose of our project? I suppose if we are saying who is interupting that could be useful. But I think our focus for the project is on the characters. So I guess my question is, is floating text tags necessary? And how would they fit into the purpose of our project?

mjb232 commented 8 years ago

See, what they have at the beginning is structured differently. Usually it looks something like this: <person xml:id="st"> <persName> <surname>Triplette</surname> <forename>Stacey</forename> <roleName>Editor</roleName> </persName> but what the XML has is this: <person id="aristippo1008" religion="pagan" sex="m" role="philosopher" origin="Atene">Aristippo (X.8)</person> The difference lies in the fact that every description of the person is used as an attribute in the Decameron file, while the examples provided us the xml:id attribute and describe the person using elements. So @ebeshero, does this mean that whats provided at the beginning of the document isn't actually a Prosopography? Or is it another form of Prosopography?

ebeshero commented 8 years ago

@laurenmcguigan I think you all have taken a closer look at the Oakland files more recently than I have, so I'll defer to you on this. Yes, from what you're saying it looks like the <milestone/> elements aren't the same as <floatingText> at all.

The reason you might want to "map" the structure of this narrative is because it's got stories nested inside stories. And that means that some of your characters only appear inside the frame, while others are embedded inside particular internal stories. If you have the narrative levels marked in a particular day, you could then retrieve information about which characters turn up in particular levels.

As for the personography, definitely have a look at the Oakland team's files, because you may just want to start with theirs and modify it or build on it. I've not taken a close look at their prosopography...so I'm not sure exactly what they've done..How does it look to you guys? (I'm happy to look at it with you, too!)

ebeshero commented 8 years ago

@mjb232 Ohhh...I see what they did. Yes, it's an alternative way to code some prosopography information, and alas what they did doesn't look like it's really TEI conformant to me... It's not exactly wrong, but it's not quite how the TEI does things. It looks like they are confining themselves to little tiny strings of text you're allowed in an attribute value so you can never record very much. I think you all can build something better here!

mjb232 commented 8 years ago

Yeah, I don't like it. Alright, so I could do a very basic Prosopography, just involving the main characters. I can designate them as male/female, storyteller, and since I liked @laurenmcguigan idea I'll also designate them as the main characters. Sound good?

ebeshero commented 8 years ago

@mjb232 I suspect the Oakland team wasn't reading up on the use of @xml:id as you all have. The @id is an attribute we usually use in HTML, and you can use it in XML too, but it doesn't have the same punch as @xml:id. Let me see if I can explain the difference.

In HTML we can use an @id on an element, like you just did recently to make a table of contents at the top of a file and have internal links down to each little section (each Emily Dickinson poem). You're only allowed to use an @id once inside a file, but you could reuse it in another file in your project--it's only a matter of using it just once per unit file. You can refer to those @id attributes using <a href="#id"> and build internal links to let your site visitors bop down to each @id you've set in your text.

In XML, you're supposed to use an @xml:id only once in your entire project. So if you decided to break up the Decameron text into multiple files (one per each day) and you defined a list of names with @xml:id attributes, you're only really supposed to represent each @xml:id just once anywhere, and you can write a Schematron rule to make sure that every @ref (or other pointer attribute with a #) is always pointing to an @xml:id that you've defined somewhere. So, across 8 XML files in a project you could have hundreds of @ref attributes "pointing at" just one @xml:id defined in your central prosopography file. When you use those @ref attributes, the XML isn't really linked like with clickable links in HTML, but you can transform it into HTML with links, or do other things to pull those pieces together--(like make little javascript mouseovers, so when we roll over a character's name, we get a little blurb of information about them). You've got a lot of options for what you do with it in web presentation mode (for transforming into HTML), but the point is that you're writing code that pulls related information together--or data binding.

The Bocaccio team put their @id attributes in the same file, I think--up in their header. I usually have a separate file that just contains the list of names and @xml:id attributes, and then write Schematron to help make sure the @ref attributes in the body of the texts we're coding always have hashtags and always point to my central prosopography file, kind of my rolladex or contact list for the project. I'm happy to bootstrap my Schematron rule for hooking a Prosopography into your main files, but for now, we just want to think about how best to handle your list of names...

ebeshero commented 8 years ago

Okay, I just modified that several times! Just refresh the page... ;-)

ebeshero commented 8 years ago

@mjb232 After the long answer...yes, I think your plan sounds good!

mjb232 commented 8 years ago

Dope. Can you think of anything else I should use to label them?

ebeshero commented 8 years ago

That's kind of up to you and the team, really. One question is whether the characters go by alternative names or nicknames in the text--so often a good prosop. list will help by holding the different names and somehow setting the official one apart...

ebeshero commented 8 years ago

...But you're probably not ready to note alternative names yet--that usually comes with reading and marking the text. For now, what you make might be pretty simple, and you can always keep building and adding detail to it as you go.

mjb232 commented 8 years ago

Sounds good. I just finished a very simple personography. Here's a sample: <person xml:id="nea"> <persName> <surname>Pampinea</surname> <forename></forename> <roleName>Story teller</roleName> </persName> <sex>Female</sex> </person> It's in TEI (which is sooooooo easy to work with).

ebeshero commented 8 years ago

Looks good...but question: Is Pampinea a surname (surname = a last name)? (Do the characters have last names that you can see?) If not, surnames and forenames are optional, and you could just list her as <persName>Pampinea</persName>. TEI is perfectly happy with that. Also, you'll want to make sure you've opened and saved this as a TEI ALL file in oXygen.

So, here's something neat about TEI personographies--neat as in tidy: The order of your names can matter, so you know where to find the most important or useful names when you hunt for them with XPath:

<person xml:id="ebb">
      <persName>
            <surname>Beshero-Bondar</surname>
            <forename>Elisa</forename>
            <forename>Eileen</forename>
     </persName>
     <persName type="alt'>
             <surname>Beshero</surname>
             <forename>Elisa<forename>
             <forename>Eileen<forename>
      </persName>
     <persName type="nickname">Dr. B</persName>
</person>
. . .
<note resp="#mjb232">Dr. B was this crazy professor at <placeName ref="#upg">Pitt-Greensburg</placeName>...yada yada yada...and she knew a bunch of people, including <persName ref="#mjb232>Matt Burch</persName>, <persName ref="#rjp43">Becca Parker</persName>, etc etc etc...</note>

TEI doesn't have a "middle name" element, so you just use a bunch of <forename> elements and list them in order from first name through middle names. You could add the other stuff in the entry too. But see how a <note> element could come in handy to help document relationships in your lists?

laurenmcguigan commented 8 years ago

So this is all I've got for the schematron so far and I don't know where to go from here `<?xml version="1.0" encoding="UTF-8"?> <schema xmlns:sch="http://purl.oclc.org/dsdl/schematron" queryBinding="xslt2" xmlns:sqf="http://www.schematron-quickfix.com/validator/process" xmlns="http://purl.oclc.org/dsdl/schematron">

<pattern>
    <rule context="tei:text//tei:pb[following::tei:pb]">
        <assert
            test=""> the page breaks must be in chronological order</assert>
    </rule>
    </pattern>

`

How do I say that they have to be in chronological order? I thought I could use substring, but the page breaks aren't in specific places.

ebeshero commented 8 years ago

@laurenmcguigan Do you mean numerical order? Or really, chronological order (by date)? confused

laurenmcguigan commented 8 years ago

Yes, I meant numerical! @ebeshero

jlm323 commented 8 years ago

I'm going to be reading and marking up Day Two once we have the ids squared away, but I wanted to ask about the <q> elements: would it be beneficial to add attributes that said whether the speech is coming from a main character or a minor character? Like this:

@who values who="mainChar" for showing a main character is speaking who="minorChar" for showing a minor character is speaking

Or is that confusing because typically the main characters will be conversing at the frame narrative level in the <floatingText> element? Also, I'm wondering how we can markup the stories within the stories like I mentioned before. Would using a <div> element with an attribute value like "nestedNarrative" work there or would it be better to include some sort of attribute value on the <q> elements like the Oakland team did with <q type="novella">?

ebeshero commented 8 years ago

@jlm323 Hi Jessica! We're having a confab in my office with Becca and Dr. Triplette, and here's what we thinking:

mmm202 commented 8 years ago

@jlm323 I'm just wondering if I am still marking up day one with the places and people for the project midterm?