The Medical Humanities Network ‘soft launched’ on Friday this week so I had quite a bit of last minute tweaking and adding of features to manage before this happened. This included updating the structure to allow people and collections to be associated with each other, fixing a number of bugs, ensuring ‘deleted’ content could no longer be accessed through the site (it was previously available for test purposes), adding a new ‘contact’ section and adding a feature that ensures people agree that they have the rights to upload images. It’s all looking pretty good and as far as I’m aware it’s going to be officially launched in a week’s time.
I also received the final pieces of information I required this week to allow paid apps to be published through the Apple App Store and the Google Play Store. This is something that has been dragging on for a while and it is really good to get it out of the way. It’s actually something that’s required by people outside of the Critical Studies and getting it sorted took quite a bit of effort so it’s especially good to get it sorted.
I also took ownership of the Pennant project’s technical stuff this week. This is a temporary arrangement until a new developer is found, but in the meantime I noticed some problems with the project’s WordPress installation. There is some sort of issue that is stopping WordPress connecting to external servers. It’s being blocked somehow and as this is affecting things such as the Akismet anti-spam plugin I thought I’d better try and investigate. I had thought it was some kind of server setting, but I installed the site on my test server and it gave the same errors, even though another WordPress site I had on the server worked fine. I tried a variety of approaches, such as updating the version of WordPress, replacing the data with data from a different instance and deactivating each plugin in turn, and I eventually figured out that it’s something within the wordpress options table that’s causing the problem. If I replace this with a default version the connection works. However, this table contains a lot of information about valid WordPress plugins so I’ll have to carefully go through it to identify what has caused the problem. I’m fairly certain it’s one of the plugins that has somehow managed to block external connections. I’ll need to continue with this next week.
I met with Gary Thoms this week to discuss the technical aspects of the SCOSYA project. The .ac.uk domain name has still not come through yet so I contacted Derek Higgins, who is the person who deals with JANET, to ask him what’s going on. Happily he said JANET have now approved the domain and are awaiting payment, so we should be able to get a project website set up in the next few weeks at least. In the meantime I set Gary’s laptop up so that it could access the test version of the site I developed last year. This now means that he can use the content management system to upload and edit survey data and things like that.
I also tried to help Fraser Dallachy out with a problem he was encountering when using the command line version of the SAMUELS tagger. When running the script on his laptop he was just getting memory errors. I updated the command he was running and this at least got the script to start off, but unfortunately it got stuck loading the HT data and didn’t proceed any further. Fraser spoke to Scott at Lancaster about this and he thought it was a memory issue – apparently the script requires a minimum of 2Gb of RAM. Fraser’s laptop had 4Gb of RAM so I wasn’t convinced this was the problem, but we agreed to try running it on my new desktop PC (with 16Gb of RAM) to see what would happen. Surprisingly, the script ran successfully, so it would appear that 4Gb of RAM is insufficient. I say it ran successfully, which it did with a test file that only included ‘the cat sat on the mat’. Unfortunately, no matter what we did by way of changing the input file, the output resolutely produced the output for ‘the cat sat on the mat’! It was most infuriating and after trying everything I could think of I’m afraid I was stumped. Fraser is going to speak to Scott again to see what the problem might be.
I spent the rest of the week getting up to speed with TEI and Oxygen for the People’s Voice project. Although I have a bit of experience with TEI and XML technologies I have never been responsible for these aspects on any project I’ve been involved with before, and to become the resident expert is going to take some time. Thankfully I found some very handy online tutorials (http://tei.it.ox.ac.uk/Talks/2009-04-galway/) aimed at complete beginners, which I found to be a very useful starting point, despite being a few years old now. With a sample poem from the project in hand and Oxygen opened on my computer I managed to make some pretty good progress with transcribing and figuring out how to cope with the variety of content that needed to be marked up in some way. The thing about TEI is there are often several different ways something could be encoded and it’s difficult to know which is ‘right’, or perhaps most suitable. Thankfully I had Graeme Cannon around to offer advice, which was hugely helpful as Graeme has been using these technologies for a long time now and knows them all inside out. By the end of the week I had familiarised myself with Oxygen as a tool, had created a RelaxNG schema using the TEI P5 Roma tool, had created a stylesheet for the Author View, had transcribed the poem to a standard that Graeme was happy with and had begun work preparing the workshop I’m going to be leaving for the project team next Wednesday.