Week Beginning 26th October 2020

This was something of an odd week as I tested positive for Covid.  I’m not entirely sure how I managed to get it, but I’d noticed on Friday last week that I’d lost my sense of taste and thought it would be sensible to get tested and the result came back positive.  I’d been feeling a bit under the weather last week and this continued throughout this week too, but thankfully the virus never affected my chest or throat and I managed to more or less work all week.  However, with our household in full-on in isolation our son was off school all week, and will be all next week, which did impact on the work I could do.

My biggest task of the week was to complete the work in preparation for the launch of the second edition of the Historical Thesaurus.  This included fixing the full-size timelines to ensure that words that have been updated to have post-1945 end dates display properly.  As we had changed the way these were stored to record the actual end date rather than ‘9999’ the end points of the dates on the timeline were stopping short and not having a pointy end to signify ‘current’.  New words that only had post-1999 dates were also not displaying properly.  Thankfully I managed to get these issues sorted.  I also updated the search terms to fix some of the unusual characters that had not migrated over properly but had been replaced by question marks.  I then updated the advanced search options to provide two checkboxes to allow a user to limit their search to new word or words that have been updated (or both), which is quite handy, as it means you can fine out all of the new words in a particular decade, for example all of the new words that have a first date some time in the 1980s:




I also tweaked the text that appears beside the links to the OED and added the Thematic Heading codes to the drop-down section of the main category.  We also had to do some last-minute renumbering of categories, which affected several hundred categories and subcategories in ’01.02’ and manually moved a couple of other categories to new locations, and after that we were all set for the launch.  The new second edition is now fully available, as you can see from the above link.

Other than I worked on a few other projects this week.  I helped to migrate a WordPress site for Bryony Randall’s Imprints of New Modernist Editing project, which is now available here: https://imprintsarteditingmodernism.glasgow.ac.uk/ and responded to a query about software purchased from Lisa Kelly in TFTS.

I spent the rest of the week continuing with the redevelopment of the Anglo-Norman Dictionary website.  I updated my script that extracts citations and their dates, which I’d started to work on last week.  I figured out why my script was not extracting all citations (it was only picking out the citations form the first sense and subsense in each entry rather than all senses) and managed to get all citations out.  With dates extracted for each entry I was then able to store the earliest date for each entry and update the ‘browse’ facility to display this date alongside the headword.

With this in place I moved on to looking at the advanced search options. I created the tab-based interface for the various advanced search options and implemented searches for headwords and citations.  The headword search works in a very similar way to the quick search – you can enter a term and use wildcards or double quotes for an exact search.  You can also combine this with a date search.  This allows you to limit your results to only those entries that have a citation in the year or range of years you specify.  I would imagine entering a range of years would be more useful than a single year.  You can also omit the headword and just specify a citation year to find all entries with a citation in the year or range, e.g. all entries with a citation in 1210.

The citation search is also in place and this works rather differently.  As mentioned in the overview document, this works in a similar (but not identical) way to the old ‘concordance search of citations’.  You can search for a word or a part of a word using the same wildcards as for the headword and limiting your search to particular citation dates.  When you submit the search this then loads an intermediary page that lists all of the word forms in citations that your search matches, plus a count of the number of citations each form is in.  From this page you can then select a specific form and view the results.  So, for example, a search for words beginning with ‘tre’ with a citation date between 1200 and 1250 lists 331 forms in citations will list all of the ‘tre’ words and you can then choose a specific form, e.g. ‘tref’ to see the results. The citation results include all of the citations for an entry that include the word, with the word highlighted in yellow.  I still need to think about how this might work better, as currently there is no quick way to get back to the intermediary list of forms.  But progress is being made.