There's a block of metadata dated Sept 24 waiting to be used to create new markup sets for Helena to work on.
Claire has supplied new short titles for the image files we've processed in the last few days. Add those titles to the files.
The following new XML files are available with images for Helena to work on:
visse_sans_fin.xml
homme_de_mesnage.xml
femme_de_mesnage.xml
rasine.xml
homme_pousse_femme.xml
femmes_a_table.xml
alison_lubin.xml
dolens.xml
femme_desolee.xml
CC sent over nine more sets of metadata, so I started working through them. I've done six so far. Just for the record, this is the procedure (using PhotoShop):
- Archive the original image in a backup folder.
- Open in PhotoShop, and convert to RGB mode. Save (and keep saving) to the main
scans
folder. - Using the magic wand, select some background grey or speckle, then choose Select Similar, and delete. Play around with this step until you've got a relatively clean image, without losing detail.
- Clean up the edges of the image, along with any egregious surface marks in the picture.
- If there's a BN mark with a ruler, it can be moved in closer to the image sometimes, to keep the size down.
- Crop the image to remove excessive whitespace around the outside.
- Add CC's metadata to the header of the XML file, and then change all the size info to match the cropped image (there are three locations where dimensions appear).
- Resize the image using Bicubic Smoother to create one 800px wide. Save this image using Save For Web, with the
wv_
prefix, into thescans
folder. - Undo the resize to get back to the original image, then set the Select Style to "Fixed Aspect Ratio", and select an interesting square. Copy to clipboard.
- Create a new image, and paste from the clipboard.
- Resize this image to 200 x 200, then Save For Web to the
images
folder with thethumb_
prefix. - Copy the main image file from
scans
toxml/images
- Move the XML file from
scans
toxml/images
- Update the server directories so Helena can see the files.
Instructions on wildcard searching are now on the search page, in a small font (specified CSS for all three styles).
Made more changes to the TOC-friendly titles (<title type="trunc">
) on CC's instructions, and in the process noted a presentation problem: sorting was being done on the original title, not the TOC-friendly title, although the latter was being displayed, so the ordering was wrong. Fixing this isn't just a matter of switching the sort parameter because some items don't have a TOC-friendly title at all, so this is my current solution:
<xsl:sort select="concat(upper-case(./descendant::tei:title[@type='trunc']), upper-case(./descendant::tei:title[1]))" collation="http://saxon.sf.net/collation?class=TitleSortComparator"/>
which seems to work fine.
I integrated the new metadata sent by Claire into seven IMT files. Then I cropped and cleaned up the images for those files, and created Web View and thumbnail versions of them. They're now available in the server account, and Helena is working on them now. That should keep her going for a couple of weeks, I think.
Added Helena to the Mariage group, and made her a user on the blog. She finished doing the transcription of the text on the first two images, and those files are now on the server (and backed up). We did some basic markup intro, with <lg>
and <l>
tags; we'll do a bit more tomorrow. Greg set up Cilantro so that the SAN directory is mounted as drive M: when H logs on.