I spent quite a bit of time this week completing the redevelopment of the ‘Readings in Early English’ resource. This was the first of the old STELLA desktop applications that I converted into an app and web resource more than a decade ago and the interface was in need of some modernisation. Last week I’d put together a first attempt at a new interface and this week I refined it and migrated all of the content across. This took quote some time as I needed to reformat much of the content for the 28 texts. Here are a couple of ‘before’ and ‘after’ shots. Firstly the homepage:
And one of the texts:
The ‘notes’ for each text were previously hidden in a pop-up but I’ve mode these permanently visible as the notes can include a lot of useful information that could otherwise be missed. I haven’t gone live with the new interface yet, but will do so this week. I also still need to decide whether to pursue the publication of app versions or whether we’ll just stick with web-based resources now. I’m in two minds about this. It takes a lot of time and effort to publish apps on the app stores and I’d rather not have to jump through all of these hoops on an annual basis in order to ensure the apps remain visible. But having the resources in app form would make them more visible to a broader audience, so it may be worth the effort. I’ll need to think about this some more.
Also this week I spent some time working for the Dictionaries of the Scots Language. I’d been sent some video recordings of presentations made at a recent team meeting, discussing new projects that are underway. I watched these and gave some feedback on how these projects may require the front-end to be updated. I also participated in a Teams call with the DSL people on Thursday, at which we discussed further upcoming projects and ongoing tasks such as the new date searches. I’d also been sent a few further updates to make to the new date searches so implemented these. This included adding in the total extent of attestation as text to the left of the sparklines in the search results and updating the way the ‘filter by date’ works when single years are entered, bringing this into line with how the ‘search by date’ boxes work. I also updated the algorithm that generates the sparkline text to tweak the way entries with no end date are displayed. This took a bit of time to update and test, and after I was satisfied with the output I then needed to regenerate the Solr index data as the sparkline text is stored in these indexes. As I can’t update the online Solr instance myself I submitted a Helpdesk job to get things sorted, but as of yet this has not been acted upon. I’m sure it will all get sorted next week.
I devoted some time this week to preparing for my Performance and Development Review, as it’s that time of year again. This involved going through all of my blog posts for the past year and picking out my main accomplishments then writing the necessary sections of the PDR documentation.
Also this week I continued to work on the old STELLA apps. I’d contacted the app people in IT Services on Monday and someone got back to me later in the week, although I’m still not certain whether we will be able to relaunch the resources as iOS and Android apps. We would have to commit to releasing updated versions of the apps at least once a year, and as these resources are reference works for subjects that don’t exactly change much (Old English, English grammar) releasing regular updates isn’t going to be particularly suitable. I’ll need to have a think about this, as it might be better if we just release updated web-based resources instead.
I focussed for some of this week on creating a new interface for the oldest of the STELLA apps: Readings in Early English. I created this more than a decade ago and an update is long overdue (currently the old version can be accessed here: https://stella.glasgow.ac.uk/readings/). I created a potential new interface for the resource this week, although I still have to add in all of the content and try out some alternatives so there’s still lots to be done.
Other than these things I had an email conversation with Thomas Clancy about another place-name project that wanted some advice on developing online resources. I also fixed a bug in the new map interface for place-names. When viewing a restricted set of place-names (e.g. as the result of a search or browse) and then pressing on a cross-reference link in a place-name pop-up, if the place-name labels are set to ‘always on’ and the cross-referenced place-name is not in the restricted set of names currently visible then the full dataset will be loaded into the map. When this happens the ‘always on’ labels should be reset of ‘off’ as having a map with thousands of ‘always on’ labels slows the interface to a crawl and is impossible to navigate. However, although the labels option in the menu is reset to ‘hover only’ the labels were continuing to be displayed. This isn’t a major bug as clearly it only triggers in very specific circumstances, but I wanted to sort it and thankfully I was able to, updating the code on both the Iona and Ayr websites (the maps are not yet publicly available on either of these resources). I also updated the Ayr resource to add in a ‘Loading’ spinner when the data first loads in. The initial load can take several seconds, during which it looks like the resource is just an empty map, so having a ‘Loading’ spinner on a greyed out map makes it clearer that something is actually happening. The initial load takes so long because of the parish boundary GeoJSON data rather than the place-name markers themselves. The parish boundary data is pretty large so takes some time to pull in.
Also this week I gave some advice to the researcher in charge of the Robert Fergusson website, had a chat with Geert of the AND about future AND work, watched some videos from a recent team meeting that Rhona of the DSL sent to me and resent the staff credentials for the new Speak For Yersel resources to Mary.
Also this week the old Italian Emblems website (https://www.italianemblems.arts.gla.ac.uk/) was finally reinstated after some recent reworking of it.
Back at the start of the year I spent a bit of time creating a new version of ‘An Anthology of 16th and Early 17th Century Scots Poetry’, a website that was originally created 25 years ago featuring 155 poems. The site was originally a plain HTML 4 site, with the poem text and glosses laid out in table rows and columns. The site is still actively used for teaching, and in redeveloping the site I decided to migrate the poems to TEI XML, separating out the data from the presentation and ensuring the data could then be reused more easily in future. It took rather a long time to migrate the poems and create the site, but I’d completed the work by March. However, the person in charge of the site, Theo Van Heijnsbergen, wanted to make some updates to the content before I replaced the old site so I had to wait until he got back to me. After a bit of prompting Theo emailed me with the updates this week, and I was finally able to launch the new site, which can be accessed here: https://stella.glasgow.ac.uk/anthol/. The site gives access to the XML source files for all of the poems and I’ve ensured that all pages of the old site redirect to the corresponding new page. It’s great to finally have the new site publicly available at last.
Also this week the researcher I’d given some advice to last week about displaying map data got back to me with an update and a couple of further questions, and I spent a bit of time thinking of possible solutions and replying to him. The editor of the Anglo-Norman Dictionary also got in touch this week, asking me to generate a list of all main entries (as opposed to cross-references) beginning with letters A to K. I ran the necessary query and exported a spreadsheet for him featuring the lemma, homonym and ‘slug’ fields. There are 14,736 entries beginning with these letters.
I spent the rest of the week returning to app development. Many years ago I’d created a series of iOS and Android apps primarily based on old STELLA teaching resources, but I hadn’t been able to update these since 2019 as the Mac I was using became too old to receive updates, and you can’t publish iOS apps without a Mac. Also, up until this point I was managing the University’s iOS and Android accounts but control of these was then taken over by IT Services and I no longer had access to them. As the apps hadn’t been updated in so long they ended up being taken down from the app stores. Last year I took delivery of a new Mac Mini which could be used for app development, but since then I haven’t had the time to get started. This week I had a bit of time so took the plunge.
To develop the apps I use a tool called Apache Cordova (https://cordova.apache.org/) which allows you to develop apps using HTML, CSS and JavaScript, which Cordova then ‘wraps’ in the native code for whichever platforms you’re targeting. It’s a very handy tool, but setting up a computer for app development using Apache Cordova is not straightforward. It would be great if you could just press a button and automatically have a working platform, but there are an awful lot of dependencies, software installations, system configurations and updates that all need to be handled manually. To make matters worse the documentation is often not up to date and doesn’t necessarily reflect what actually needs to happen. I’m not going to go into any detail, but suffice to say that it took several days of installing, tweaking, configurating, upgrading and in some cases downgrading, processing and building before the platform was working and I had the default test app working in the iOS and Android emulators, my iPad and my Android phone.
At this point I was able to try to get one of the old apps working, and I decided to begin with the last app I’d created, as I figured that it wouldn’t need much updating. This app is ARIES – Assisted Revision in English Style, which is available as a standard website here: https://stella.glasgow.ac.uk/aries/. I made a few tweaks to the code, but on the whole I think it holds up pretty well. I managed to wrap the app and deploy it to my iOS and Android test devices, and all is looking pretty good. The next step will be to contact IT Services to see how I should go about publishing the updated version on the app stores, assuming they’ll let me do so. I’ll see about this next week.
I was on holiday for the previous two weeks and returned to a full week this week. I spent quite a bit of the week catching up with emails that had arrived whilst I was away and dealing with the issues contained in them. One such issue was sorting out the Italian Emblems website, which had been taken down during the College of Arts server crisis that we tackled more than a year ago. I wasn’t involved in the creation of the site, which was created almost 20 years ago, and hadn’t realised it was still offline until I received an email from someone in Special Collections. I worked with Luca this week to ensure the site’s code would work on a new server and had several email conversations to arrange for someone to be the named owner of the site, as all people involved in its creation are no longer associated with the university. We are now in a position where we can get the resource online again and are just waiting for someone in IT Services to sort out the domain, which will hopefully happen soon.
Also this week I sorted an issue with the Books and Borrowing content management system. A researcher is continuing to add data to the resource, even though the project has now officially ended, and was encountering an issue when attempting to associate existing book holding records with new borrowing records. It turns out the issue was book names that included double-quotes in their titles. These were resulting in malformed JSON files being sent to the book holding autocomplete field, meaning the list of possible books was not getting populated. We’d agreed back in the day that book titles should not include double-quotes, but ideally I should have added a check for this to the CMS. However, now that the project is officially over I decided the simplest fix would be to replace the offending double-quotes with curly double-quotes in the database and warn the researcher about using double-quotes in future.
I implemented a few other minor fixes for other projects this week, including updating a banner image on a project website and updating the Ayr Place-names map interface to ensure that when the map was reset the view also reset (previously the view remained focussed on the region the user was previously viewing). I also replied to an email from fellow developer David Wilson regarding linked open data and how we might host this and had an email conversation with Deven Parker about her 18th century playbills project.
Before I went on my holiday a researcher who is creating a place-names project about Carrick in south Ayrshire. He is using an existing system for displaying map data called with GeoJSON Dashboard (https://github.com/fulcrumapp/geojson-dashboard) and wanted some advice on developing an advanced search. I spent some time this week looking into the tool, figuring out how it worked and giving some advice on how it could be updated.
Finally this week I devoted some time to a few outstanding issues relating to the new advanced search on the test version of the Dictionaries of the Scots Language. Although I completed work on this new version, with its sparklines and more comprehensive quotation search options many months ago we’ve still not gone live with it as the team are still testing things out. Before I went away they got back to me with some suggested updates and I implemented them this week. This included changing the way error messages are displayed when no search term is entered and updating how the system processes one single date being entered rather than a date range. Previously one single date was treated as a search for a specific year, but the team instead wanted this to be treated as a range, either from the given date to the end of the dictionary period when the ‘from’ box alone was populated or from the start of the dictionary period to the given date when the ‘to’ box was populated. I therefore updated the code to reflect this. A couple of further suggested updates were proposed, but I had some misgivings about implementing them and therefore replied to the team to voice these and ask for clarification. I’ve not heard back regarding these items yet so will just need to see what they think.