Endings

Customized a lot of stuff from Mariage to create a buildfile for the webapp, built it into the main build process, tested it, and began serious work on the search. The existing search results page is very odd in several ways, which I've outline in an email to EGW and JT; I believe it should be much simpler, and if they agree, I'll make it so in preference to reproducing the current weird setup.

Bringing it closer to the original appearance.

Spent most of the day completing the home page in the Graves site, using a flex layout instead of the original nested-table structure (there is still one table, but it is for a tabular block of content, I contend). What we have now seems nice and flexible, and all the form elements and labels now have captions and hints as expected. Next step is to build the eXist app.

First draft of intro coming in at 343 words.

Met with JT and worked on the outline for the Diagnostics submission. Team meeting with decision about book chapter.

I've started work on elaborating the currently-sparse HTML head content for the Graves HTML pages. The first thing I learned was that, in order to use DC and DCTERMS values in meta tags appropriately, I needed an HTML link element with an appropriate @rel; but that was invalid according to VNU. Upgrading VNU solved that, but the new VNU requires Java 8, which wasn't available on Jenkins; I had to add it to Jinks through a PPA. The new VNU also complains about the meta encoding tag in Mariage, so I've removed that. I'm currently running builds on three Endings projects and Landscapes to see whether the Java 8 and VNU upgrade has worked OK, or whether they need more tweaks.

In the meantime, I have done a bit of work on the Graves headers, and they're well on the way. I need to add substantial amounts of metadata in the headers to allow eXist and Solr to filter by document type. I've also contacted LW at the library to see if I can go and look at the MS to discover what happened to the missing page-images for 1938-05-11 and 1938-05-23, which are missing from our set, although transcribed.

Small-scale team meeting, followed by some individual discussions. One significant focus in the next month or so needs to be researching anything previously published on project diagnostics and adding it to Zotero.

JT and I have worked out a plan to upgrade several aspects of the XML encoding in advance of any rendering work; we've also figured out how we want to create static URLs to content. JT's now moving ahead with XML fixes. Other work (fixes for broken links etc.) is almost complete, thanks to TL and EGW.

met with the Library team to talk about Solr and how best to prepare documents for ingestion. Key points: my imagined usage scenario, where we host a site but they keep the index, and we use the Solr REST API to service our search pages, is very practical, and will probably constitute the first stage in an eventual migration to management by the Library. HCMC will build and test Solr instances to work on creating custom indexers that are aware of (e.g.) TEI or early modern French, as well as developing best practices for constructing HTML5 document headers for most effective ingestion (based on DC).

...for tomorrow, and in the process, clarifying some ideas about how we should proceed.