Mariage

Met with CC, and fixed a typo in sur_un_jaloux; then planned the menu system for adding articles etc. We will have a root menu header of "Paratexte", with the following submenus: Articles, Références, Bibliographies. Under Articles will be a TOC pointing to the articles, and similarly with Bibliographies. The Références item will eventually show us the full list of all the references from all the documents, but how that will be displayed is still not decided.

Met with LSPW, LCC and CC. Tasks for me, in order of priority:

• Answer LSPW's questions re Sonnet 1609.
• Get together with all three markup folks to review the two (or more)? versions of references.xml that exist, and put a system in place for avoiding duplication etc. This will happen next Tuesday at 10.30.
• Upload the latest version of the Varin and Cholière to the db, and check what it looks like; tweak the appearance/layout settings in the header, and work with LCC and EM on any necessary adjustments to the XML.
• Add XSLT handling for the socratic dialogue tagging in the Complaintes.
• When the Candale is complete and proofed, upload it and start working on the appearance/layout stuff.
• Write the handling for the items in the references.xml file -- this is partly done, I think, but will need more work.

Also reviewed draft grant application and tweaked my SSHRC attachment; once we're happy with it, I'll add it to my SSHRC profile and then I'll have completed the process of "joining" the project on the SSHRC site.

I've made the basic changes to the site suggested in the last post from CC, including changing the Bienvenue! bit to Accueil, which I presume is what was meant (rather than just removing the heading entirely from that page). I also wonder whether we should add to Équipe the names of those who worked on the project prior to the summer, but have now left?

Also uploaded LSPW's most recent version of the Sonnet 1609. Haven't had a chance to get to anything else yet -- still wading through 532 emails that had stacked up while I was away.

All the Mariage crew know pretty well what they're doing, so they shouldn't need much help, but just in case, here's a bit of background on the ongoing work:

• Documents currently undergoing editing are in the "mariage" account on TAPoR, here:
  /mariage/ccarlin/xml/unfinished/
  
• LSPQ and LCC are working largely on the server directly; EM is working locally, and using Fugu to upload her changes to the server.
• The main schema is mariage.rng, but LCC is using a more complex schema for one of her documents because she needs the drama module; that will get rolled into the main schema eventually.
• LSPW is used to editing a file, and then having me upload it into the database to see the appearance on the server and make sure it's right. While I'm away, though, she has a lot of other stuff she can be doing that doesn't need uploads to the server, so if you don't feel like figuring out the eXist client upload process, there's no need to do that. I can do uploads when I get back from Galliano.

LCC has marked up the Complaintes text as part of the DHSI workshop, treating it as a drama, which means I had to build a new schema to incorporate the drama module for her. This hasn't been migrated to the main project yet. She's also been encouraged to use the @ref attribute on regular tags instead of our normal practice of using <ref> with @target, and I can see the advantages of this when there is already a tag enclosing the text you're referring from, so I'll have to write handler code for that at some point. Put it on the list...

Spent some time setting up WinSCP on her laptop too, so she can work at home to get some extra hours in.

LSPW is reaching the end of the 1609 Sonnet. Uploaded the latest edits.

One more article came in, this time for the film volume -- that's on the back burner for the moment -- and more corrections for the NVD review. Did the corrections, and also uploaded corrections to the Sayer text I'd made this morning.

Uploaded LSPW's changes to 1609 Sonnet, and answered a question regarding catchwords.

The 1609 Sonnet has little marginal headings that we've designated as arguments, tagged with the <argument> tag, and these were correctly marked with @rend attributes containing float: left or right, but up to now I hadn't added any XSLT handlers to make them appear correctly on the site. I've now done that, and in the process we've added negative margin settings to the left-floating arguments to take them out of the way of the regular text. The right-floating ones are fine without any special handling. This is a mildly interesting issue -- should the @rend attributes actually need to make such a distinction, or is this just a product of the XHTML document instance? This is almost too complicated to figure out, but I'm tending more and more to the opinion that CSS is more than adequate as a layout and description language, so the more simply we can pass it through to the XHTML output, the better. Ultimately we might be able to use pure XML+CSS, but of course then we wouldn't get any annotations or interactivity.

Cleaned up some of the CSS for the reference links, to left-align the text instead of justifying it, and also worked through the references XML document itself, removing pointless placeholders and inserting a standard text for references which are not complete yet.

Implemented the second option for reference linking, and got it all working. Briefly:

• When the XQuery retrieves a document from the db, it checks for <ref type="reference"> tags. If it finds any, it creates a <back> element containing a <div type="references"> element.
• For each ref link it finds in the document, it retrieves the matching reference <div> from the references subcollection and puts it into the <div> above. This means the XML document now has a matching explanatory text for each unique reference in its back matter.
• The XSLT then detects the presence of these, and turns each <ref> tag into a link which calls some JavaScript.
• The JavaScript pops up the reference information on the right of the screen.

The display of reference info looks very similar to that of note info, but it's not; the notes are displayed through pure CSS using :target, but the references cannot be set to position: fixed because they're frequently very long and need to scroll, so they're set to position: absolute and just placed next to the calling reference link.