Week Beginning 15th April 2024

The Books and Borrowing project has its official launch next week, and I spent some time this week preparing for it.  This included fixing a speed issue with the site-wide fact page (https://borrowing.stir.ac.uk/facts/) that was taking far too long to load its default view, something that has somehow got worse since I originally created the feature.  After undertaking some investigation into what was causing the slow loading time it turned out that the main sticking point was the loading of data for the two ‘Borrowings through time’ visualisations.  This was taking quite a long time to calculate for all libraries.  I therefore created cached files for the data used for all libraries and have set the API to call on these rather than querying the database whenever specific libraries are not requested.  This has greatly speeded up the loading of the page, which (for me at least) is now practically instantaneous, while before it was taking up to 30 seconds.

I also went through all of my blog posts to extract text relating to the project going back to create a single document.  It’s 130 pages long and contains more than 56,000 words, covering all of the work I did for the project over the course of four years.  I then spent some time preparing a talk I’ll give at the launch about the development of the resource.  I haven’t finished working on this yet and will continue with it next week.  I also met with Matt to discuss the British Association for Romantic Studies journal (https://www.bars.ac.uk/review/index.php/barsreview) that needs and overhaul.  I’m going to work on this for Matt and hopefully get an updated version in place before the end of May.

Also this week I continued to work for the Speech Star project, adding another batch of videos to the website.  This took a fair amount of time to implement as the videos needed to get added to a variety of different pages.

I then returned to working on the Speak For Yersel follow-on project, focussing on Wales this time.  I needed to generate a top-level ‘regions’ geojson file that amalgamated the area polygons into 22 larger regions.  I did this in QGIS, manually selecting and merging the areas.  With the regions file in place I could then set up the Wales survey using my survey setup script.  This all went pretty smoothly, although there were some inconsistencies between the area geojson file and the settlements spreadsheet that I needed to fix before I could get the survey to setup successfully.

I’d also received feedback about the Republic of Ireland and Northern Ireland surveys, including many changes that needed to be made to the survey questions.  I decided that the easiest way to handle this would be to delete the surveys (which were still only test versions with no real data) and start again with updated question / answer spreadsheets.  There were some other updates that had been suggested that would apply to all surveys and I updated all three sites to incorporate these.  I then spent a bit of time investigating QGIS and whether it could be used to create simpler versions of the region polygons in order to generate a logo for each of our new sites that would be analogous to the Scottish survey.  After discussing this with Jennifer and Mary we agreed that Mary would take this forward, so we’ll hopefully see the outcome next week.