Came across a couple of instances in the latest volume where hyphenation would have been desirable, but hadn't occurred, even though hyphenation is turned on in XEP. One was a long German word ("Systematisierungsmöglichkeit"); I was able to trigger hyphenation in this by inserting the Unicode soft hyphen like this: "Systematisierungs­möglichkeit" The oddest cases are those in which there is an existing hyphen in the word (e.g. "Skole-Kammeratern"). XEP fails to break the word at the hyphen even if it would help the layout. If you then insert a soft hyphen, you end up with two hyphens. The solution in this case was to insert a zero-width space ("Skole-​Kammeratern"), which triggers hyphenation but doesn't affect the layout in any other way.
Started working through JT's complete corrected proof copy, and got many pages into it. This resulted in some back-and-forth about the use of ellipses, and the punctuation and spacing around them; we are now determined to be following the three-dot method from Chicago, and where an ellipsis comes between the end of a sentence and a following sentence or fragment, the period at the end of the first sentence is discarded (but other meaningful punctuation such as commas, colons, semicolons, q and excl. marks are retained). Ellipses are preceded and followed by spaces. Many of our old document don't conform to this, but we're leaving those alone, and only correcting the upcoming documents. Many cases of all types needed fixing.
Another decision is to standardize on the use of a comma to separate a University and its campus or city name (University of Wisconsin, Madison).
Many more corrections to go...
Did the teiHeader, introductory bits (titles, bios, abstract), first paragraph, and first biblio entry, then added all the other biblio info in comment form.
HT sent more fixes from the proofing PDF, and so I made those. The changes to NVD's review sent by the author last week have precipitated a screw-up in the page-break setup, because they've pushed subsequent reviews forward, causing a widow problem. Fixing this was really difficult, and I've ended up having to use keeps and breaks in profusion, which has resulted in some short pages (where content is pushed to the next page to avoid widows). This is really the best we can do; the situation arises because of the large number of reviews (10), and the length of headers (typically up to four lines) and footers (two or more lines) for each review, which mean that the chances of widows/orphans arising is very high. With a smaller number of reviews, there's a better chance that messing around with the vertical spacing between them can solve the problem, but it's not really possible in this case.
Made a few more changes based on author's request to one of the reviews, then uploaded it and rebuilt the PDF. Sent the PDF out. I think we're close now; there are a couple of outstanding issues for JT to rule on with regard to the Houe article, but as soon as they're squared away we're ready to go.
A lot of detail changes to several articles. Still working on this; some are problematic (author's preference for order of biblio components is different from our norm, for instance).
Abstract translations arrived for the five remaining articles (one already had a translation). Entered them and generated the PDF -- thankfully, they all fit on the page OK. Took a quick scan through, and it looks as though the reviews are all starting and ending in reasonable places, so I think we're good to go. Sent the final PDF to the editors. Then I gingerly uploaded the five files into eXist, with all my fingers and toes crossed, and everything went fine. It appears as though the only current problem with eXist is the "unable to store generated fragment" problem that's scuppering the Mariage search page.
What remains to be done:
- When the PDF is deemed final, extract the TOC and massage it into the cover in Illustrator.
- Generate all the colour separations etc. for the cover, following steps already outlined on the blog.
- Add the page numbers into all the volume 18 documents, now that they're fixed.
- Liaise with the editors with regard to the timing of the release of the contents of vol 18 on the website -- when they're to be released, remove the
@rend="proof_only"attribute, and re-upload them.
Started work on the cover for volume 18, which needs the usual changes to date and volume number, but also requires adding HT to the cover as book review editor. Since this was all done in Windows, and the master volume is being built in Windows because that's where I have my working desktop XEP, I'm working in Illustrator CS2 on XP in my VM. I started following my own instructions elsewhere in the blog, changing the text where required, and remembering that there are knockouts for the years at the bottom, so you have to change the year twice, once in the black top layer and once in the white layer below it (send the top layer back one in the z-order to get at the white layer, then send the white layer below it when you're done).
I have a lot of work to do once the TOC and page numbers are finalized (this will depend on the final proofing, which can't be done until the translated abstracts are in, so I have to wait a bit). When everything else is ready, I'll follow these steps to produce the final cover. A more detailed description of the process is here. In the meantime, I wanted to do a test print, so I tried to export from Illustrator as a PDF; Illustrator froze every time I tried it. After reboots didn't help, I did some Googling and found that this is a symptom of a corrupted preferences file, so I deleted c:\Documents and Settings\[username]\Application Data\Adobe\Adobe Illustrator CS2 Settings\AIPrefs. That did the trick, and I was able to save a PDF. Then after reconfiguring my printer in XP (the IP has changed since I last printed from the VM), I was able to do a quick test print on 8x11, which looks OK.
Making tweaks and edits to lay out volume 18 better (specifically reducing spacing between book reviews, to avoid widows and orphans, and adding the book reviewer to the cover); several fixes to bios and keywords. Still in discussions about the date of the volume, and the possibility of adding formatting (italicization) to the keyword index.
Finished marking up the last article for volume 18. When I came to upload it, I got the same duplicate node error we had last week; a restart of Tomcat solved it, but we need to watch out for this. If it happens frequently, we probably have corrupted indexes and we'll need to follow the instructions here to clear the index files.
Author corrections to two volumes came in, and I did those.
Also did the first build of the Volume 18 PDF. I'm doing this in my Windows VM, simply because that's already got XEP installed in it, and I'm not sure (after five years) whether it'll be practical or possible to obtain the Linux version of that particular XEP, which we purchased so long ago, and get it authorized. If ScanCan stays with us after this year, then we'll make a 5-year plan which involves a newer desktop version of XEP for Linux.