I devoted about three days of this week to developing the new place-names map for the Iona project.  My major task was to make the resource ‘remember’ things, which has taken a long time to implement as basically everything I’d developed so far needed to be extended and reworked.  Now as you scroll around the map and zoom in and out the hash in the URL in the address bar updates.  Your selected ‘Display options’ also appear in the hash.  What this means is you can bookmark or share a specific view.  I can’t share the full URL of the map yet as it’s not publicly available, but for example the hash ‘#15.5/56.3490/-6.4676/code/tileNLS1/labelsOn’ provides a URL that loads the map zoomed in on the islands of Stac MhicMhurchaidh and Rèidh-Eilean, with the OS 1840-1880 base map selected, categorised by classification code and labels always on.  Any search or browse options you’ve entered or selected are also remembered in this way, for example the hash ‘#16/56.3305/-6.3967/language/tileSat/labelsOff/browse/nameCurrent|C’ gives a URL for a ‘browse’ showing current names beginning with ‘C’ on a satellite map categorised by element language focussed on the main settlement.

The same approach also works for the search facility, with search options separated by a double bar and the search term and its value separated by a single bar.  For example ‘#13.75/56.3287/-6.4139/date/tileSat/labelsOn/advancedSearch/eid|181||elementText|d%C3%B9n%20(G)||lang|G’ gives a URL for an advanced search showing results where the element language is Gaelic and the element in question is ‘dùn’, categorised by earliest recorded date.  With an advanced search you can now press on the ‘Refine your advanced search’ button and any advanced search options previously entered will appear in the search boxes.

You can also now bookmark and share a record when you open the ‘full record’, as the action of opening a record also adds information to the URL in the address bar.  So, for example ‘#14/56.3293/-6.4125/date/tileSat/labelsOn/advancedSearch/eid|181||elementText|d%C3%B9n%20(G)||lang|G/record/4262’ is the hash of the record for ‘Dùn nam Manach’.  Note that this also preserves whatever search or browse led to the place-name being opened too.

I also made some updates to the ‘full record’, the most important being that the text appearing as links can now be clicked on to perform a search for that information.  So for example if you press on the classification code in the record this will close the record and perform a search for the code.  Place-name element links are not yet operational, though, as these will link to the element glossary, which I still need to create.  I have however created a new menu item in the left-hand menu for the glossary and have figured out how it will work.  I’m intending to make it a modal pop-up like the record and advanced search.  I did consider adding it into the left-hand menu like the ‘browse’ but there’s just too much information for it to work there.

I also added in a new ‘Cite this record’ tab to the ‘full record’ which will display citation information for the specific record.  I still need to add this in.  Also new is a bar of icons underneath the left-hand menu options.  This contains buttons for citing the map view, viewing the data displayed on the map as a table and downloading the map data as a CSV file, but none of these are operational yet.

On Thursday I had a meeting with the Iona team to discuss the map.  They are pretty pleased with how it’s turning out, but they did notice a few bugs and things they would like done differently.  I made a sizeable ‘to do’ list and I will tackle this in the new year.

I spent most of the remainder of the week working on the Books and Borrowing project.  I updated the languages assigned to a list of book editions that had been given to me in a spreadsheet and added a few extra pages and page images to one of the registers.   I then returned to my ‘to do’ list for the project and worked through some of the outstanding items.  I moved the treemaps on the library and site-wide ‘facts’ pages to separate tabs and I went through the code to ensure that the data for all visualisations only uses borrowing records set to ‘in stats’.  This wasn’t done before so many of the visualisations and data summaries will have changed slightly.  I also removed the non-male/female ‘top ten’ lists in the library facts page, as requested.

I then moved on to creating a cache for the facts page data, which took about a day to implement.  I firstly generated static data for each library and stored this as JSON in the database.  This is then used for the library facts page rather than processing the data each time.  However, the site-wide facts page lets the user select any combination of libraries (or select all libraries) and the ‘top ten’ lists therefore have to dynamically reflect the chosen libraries.  This meant updating the API to pull in the ‘facts’ JSON files for each selected library and then analyse them in order to generate new ‘top tens’ for the chosen libraries.  For example, working out the top ten genres for all selected libraries meant going through the individual top ten genre lists for each library, working out the total number of borrowings for each genre and then reordering things after this merging of data was complete.

Despite still requiring this processing the new method of using the cached data is considerably faster than querying and generating the data afresh each time the user requests it.  Previously displaying the site-wide ‘facts’ page for all libraries was taking up to a minute to complete whereas now it takes just a few seconds.  I also made a start on updating the site text that Katie had sent me earlier in the week.  A large number of tweaks and changes are required and this is likely to take quite a long time, but I hope to have it finished next week.

Towards the start of the week I also spent some time in discussions about what should become of the Pilot Scots Thesaurus website.  The project ended in 2015, the PI moved on from the University several years ago and the domain name will expire in April 2024.  We eventually decided that the site will be archived, with the data added to the Enlighten research data repository and the main pages of the site archived and made available via the University’s web archive partner.

Towards the end of the week I did some further work for the Anglo-Norman Dictionary, including replacing the semi-colon with a diamond in the entry summary box (e.g. see https://anglo-norman.net/entry/colur_1) and discussing whether labels should also appear in the box (we decided against it).  I also had a discussion with the editor Geert about adding new texts to the Textbase and thought a little about the implications of this, given that the texts are not marked up as TEI XML and the Textbase was developed around TEI XML texts.  I’ll probably do some further investigation in the new year once Geert sends me on some sample files to work with.