HCMC Blogs

1847 is posted and fine-tuned for now, barring any niggling errors that were missed during the final edit phase. Throughout the 1847 process a number of tasks were accomplished, including the update of "when" to the dates attributes for all files to date, a second final pass over the 1846 documents, and a couple of necessary entries to the workflow document. I also added a few biography entries and updated the places file.

Next, I will move on to parse the remainder of the 1848 images. Luckily, there do not appear to be too many left, especially after processing the 1848 images that lead up to the final 1847 scans. I should be able to complete the 1848 batch by tomorrow afternoon.

Hi Ellen, I am going to try to answer your questions as methodically as possible:

1. The Lexigos site (the cool site Claire was referring to, I believe), is at this address: http://www.lexilogos.com.
2. The Encyclopédie Universalis address is as follows: http://www.universalis.fr/.
3. The Laval bibliography might also be online, but I was actually talking about a two-page handout Claire made for us (I will photocopy one for you this afternoon and then on our coffee date Friday I can give it to you :)).
4. The Petit Robert I was talking about is actually not online either (to my knowledge, though it might be, but it's not what I'm referring to for my references). It's Claire's Petit Robert : Dictionnaire illustré des noms propres. Dictionnaires le Robert : Paris, septembre 1994. (Sorry for the sloppy citation method but I'm almost out of time!) If you want to come and meet me in the HCMC, I can show it to you (or I can also bring it up to my office -- I'm the ony one using it.)

Cheers everyone! Lauren

I've written the basic code for the collected references page on the site (XQuery, XSLT, CSS and JS). It shouldn't yet be linked from the menu, but the basics are there. I need to add some code for "find references to this in the text", but that will come later. In the meantime, LCC is working on cleaning up the existing references, and I'll be turning my attention to bibliographical data.

We're now making a habit out of marking up all dates wherever they occur in the documents, and this caused a problem with the XSLT processing dates, which assumed that they were block-level elements (as they are in the openers and closers of despatches and minutes). I'm now processing them only when they occur outside <p> and <ab> tags, to avoid rendering them as <div> elements in the output when they're actually inline. At some point I may refine this to detect when they're inline, and also perhaps add some extra processing to produce a title attribute which could display the full date in a tooltip (since many dates are partial, and thus ambiguous if you're scanning or searching a document rather than reading it sequentially).

Set up the file for the final Vol 18 text, completed the headers, marked up the first para, turned the biblio into comments and marked up the first entry (to make the file valid), and then went through the word-processor file in OpenOffice, doing a set of search-and-replace operations that are designed to reduce the markup workload later by introducing any tags that can be done in this way. This was especially useful in this text, because it mentions dozens of short-story titles that can be tagged based on their quote marks (although many quotes were reversed).

So going off the Laval bibliographic citation page that Claire gave us, my citations of the Petit Robert are formatted like this:

<bibl><title level = "a">Plessis-les-Tours</title>, <title level="m">Le Petit Robert : Dictionnaire illustré des noms propres</title> (septembre 1994), Paris, Dictionnaires le Robert, p. 1700.</bibl>

I have also begun using the quote and cit tags as suggested by MH to LSPW:

<p><cit><quote>quoted material!</quote></cit></p>

Updated the 1846 and 1847 files in the db. KSW is now doing a last check through 1846 again, but basically we're done up to the end of 1847 (except with some incomplete reference info). Also we took the time to clean up some old copies of files on home1t/coldesp so that we don't have confusing multiple copies of files on the server. The setup which is supposed to enforce the group as coldesp on the server doesn't seem to be working, so all of ksw's uploads have been going up as owner ksw, group ksw, making them unwritable by anyone else. We had to fix all that at the command line through SSH.

This free Windows anti-virus app left beta yesterday. Tested it on my Windows 7 VM -- works fine. Now I'm installing it into my workaday XP VM (the image of my old computer), which I still use for a lot of key tasks. Fingers crossed...

Did a set of site updates for Greek vocabulary (Chapter 2, and the review of Ch 2 in Chapter 3). In the process, I re-learned a lesson that I've had to learn before, and haven't documented, so here it is:

The Greek site is really designed to be built en masse using Masher batch files; there's a batch file for each of the main sections (Grammar, Vocab, etc.). Key info for building each exercise is in the Batch file, rather than the individual Masher files in each of the chapter folders. Therefore, if you want to build an individual unit or an exercise from a unit, this is what you need to do:

• Load the Masher file for that unit/chapter.
• Load the Masher Batch file for that section (e.g. vocab). This loads extra info into the Masher interface, required when the exercise is being built.
• Build the unit or chapter.

When you're done, check the exercises look right on the server. Symptoms of borked settings include a green background in drag-drop cards in matching exercises, or a missing background image (replaced by plain green).