Week Beginning 6th July 2015

I continued working on the new website for the Thesaurus of Old English (TOE) this week, which took up a couple of days in total. Whilst working on the front end I noticed that the structure of TOE is different to that of the Historical Thesaurus of English (HTE) in an important way: There are never any categories with the same number but a different part of speech. With HTE the user can jump ‘sideways’ in the hierarchy from one part of speech to another and then browse up and down the hierarchy for that part of speech, but with TOE there is no ‘sideways’ – for example if there is an adjective category that could be seen as related to a noun category at the same level these categories are given different numbers. This difference meant that plugging the TOE data into the functions I’d created for the HTE website just didn’t work very well as there were just too many holes in the hierarchy when part of speech was taken into consideration.

The solution to the problem was to update the code to ignore part of speech. I checked that there were indeed no main categories with the same number but a different part of speech (a little script I wrote confirmed this to be the case) and then updated all of the functions that generated the hierarchy, the subcategories and other search and browse features to ignore part of speech, but instead to place the part of speech beside the category heading wherever category headings appear (e.g. in the ‘browse down’ section or the list of subcategories). This approach seems to have worked out rather well and the thesaurus hierarchy is now considerably more traversable.

I managed to complete a first version of the new website for TOE, with all required functionality in place. This includes both quick and advanced searches, category selection, the view category page and some placeholder ancillary pages. At Fraser’s request I also added in the facility to search with vowel length marks. This required creating another column in the ‘lexeme search words’ table with a stricter collation setting that ensures a search involving a length mark (e.g. ‘sǣd’) only finds words that feature the length mark (e.g. ‘sæd’ would not be found). I added an option to the advanced search field allowing the user to say whether they cared about length marks or not. The default is not, but I’m sure a certain kind of individual will be very keen on searching with length marks. If this option is selected the ‘special characters’ buttons expand to include all of the vowels with length marks, thus enabling the user to construct the required form. It will be useful for people who want to find out (for example) all of the words in the thesaurus that end in ‘*ēn’ (41) as opposed to all of those words that end in ‘*en’ disregarding length marks (1546).

I think we’re well on track to have the new TOE launched before the ISAS conference at the beginning of next month, which is great.

I would also like to have the new ‘Essentials of Old English’ app and website available before the conference as well, but I haven’t had a chance to act on Christian’s extensive feedback since I got back from my holiday a number of weeks ago. I’m happy to say that I managed to find a little bit of time to work on the updates this week, and have so far managed to implement most of the required updates. The only thing left to update is the reordering of the ‘plus’ exercises, which I’m afraid is going to cause me some serious problems due to the way the JavaScript that processes the exercises works. I’ll try and get this sorted next week though. I also need to ‘wrap’ the code into an app, test it on iOS and Android devices, create store records for the apps, generate screenshots and icons and all those other tedious tasks associated with actually getting an app published. I’m hoping to get all of this done next week but I’ll just need to see if anything else crops up that eats into my time. I did finally manage to set up a Google Play account for the University this week, thanks to James Matthew and his purchasing credit card. I should now be able to use this to create and publish Android versions of all of the STELLA apps. It will also be useful for the Mapping Metaphor app too.

I spent some further time this week on Gavin Miller’s Science Fiction and the Medial Humanities project, continuing to work on the WordPress plugin I’ve created for creating and managing a bibliography. Last week I created the form that will be used to upload bibliographical items while this week I created the code that handles the actual upload and editing of the items, and also the tabular view of items in the WordPress admin interface. It seems to be working pretty well, although facilities to handle error checking for custom data types seem to be a bit limited. I found a way to do this on this Stackoverflow page: http://stackoverflow.com/questions/13216633/field-validation-and-displaying-error-in-wordpress-custom-post-type , although I had to update it to store actual error messages relating to individual fields as a session variable in order to display these messages after the form has submitted. It’s still not ideal as the pesky error messages are being displayed as soon as I open an item to edit, even before I’ve submitted any updates. I might have to resort to JavaScript error messages, which might not be so bad as it’s only project staff who will be uploading the data.

I continued working on the Scots Thesaurus project this week as well. I met with Susan and Magda on Tuesday to talk them through using the WordPress plugin I’d created for managing thesaurus categories and lexemes. Before this meeting I ran through the plugin a few times myself and noted a number of things that needed updating or improving so I spent some time sorting those things out. The meeting itself went well and I think both Susan and Magda are now familiar enough with the interface to use it. I created a ‘to do’ list containing outstanding technical tasks for the project and I’ll need to work through all of these. For example, a big thing to add will be facilities to enable staff to upload lexemes to a cateogory through the WordPress interface via a spreadsheet. This will really help to populate the thesaurus.

So, that’s four major areas of work that I covered this week. I also met with Craig Lamont again to try once more to get a historical map working within the University’s T4 system. A helpful T4 expert in the University had updated the maximum number of characters a section could contain, which allowed use to upload all of the necessary JavaScript information to power the map (e.g. the leaflet.js library). After that we had a working map (of sorts) in the preview pane! However, none of the markers were appearing and the formatting was completely off. However, with a further bit of tweaking I managed to fix the formatting and get the markers to appear, so all is looking good!

I also spent a little time contributing to a Leverhulme bid application for Carole Hough and did a tiny amount of DSL work as well. I’m still no further with the Hansard visualisations though. Arts Support are going to supply me with a test server on which I should be able to install Bookworm, but I’m not sure when this is going to happen yet. I’ll chase this up on Monday.