Week Beginning 10th February 2014

I spent a lot of time this week working with the Mapping Metaphor interface in preparation for Friday’s demonstration at the Digital Humanities event in Edinburgh.  I added in the facility to download the visualisation as an SVG image.  I had previously been hoping to provide facilities to export the visualisation in a variety of formats including vector (SVG) and raster (PNG) but as documented in an earlier post, I had encountered some major difficulties in converting the exported SVG image to a different format, either through the client’s browser using a Javascript library or at the server side.   So I’ve decided to leave it as ‘download SVG’ – users can then convert the file to a different format themselves if they want!  It actually took longer than I had expected to get the ‘download SVG’ option working because the SVG is generated purely on the client side by Javascript.  It turns out getting a browser to generate and save a file on the user’s file system is trickier than it looks at first glance, and instead I made a script that would post the SVG data back to the server to then be made available for download as a file.  It’s a bit of a clunky approach, but it works fine and the resulting SVG looks very nice.

I also made some further browse options for the ‘drilldown’ view of the data, which are now accessible from the ‘metaphor card’.  When viewing the connections between two major categories if you click on a line joining up two categories the card now gives you the option of viewing all the other categories that are connected to either one or both of the categories in question.  It’s a nice little feature that allows visualisations comparable to Ellen’s previous ‘Beauty and Light’ diagrams to be viewed.  I have also implemented the option to select the type of metaphor to view, which now appears as an option in the left-hand box.  You can select whether to view strong, weak or both types of metaphor and the visualisation updates immediately to reflect your choice, both through the aggregated diagram and the drill-down one.  I also moved the colour-code ‘key’ information to its own little box as the left-hand box was starting to get a bit cluttered.

The final update I made was to begin working on the timeline view of the data.  I decided to use the timeglider API for this, and created a PHP script that would automatically spit out data in the correct JSON format depending on user selections.  At the moment the script only spits out data for three ‘streams’ (Society-World, Mind-Society, World-Mind) and you can’t provide different options, but this is just a start.  The metaphor data gets displayed on the timeline as an exact point in time (i.e. not rounded to 50 year chunks) with different coloured icons for the three streams and different sizes of icon depending on whether the metaphor is strong, weak or unclassified.  Even with the test data it is obvious that there is too much data to be properly displayed.  If you zoom out so you can view 100 years on screen at one time the screen is swamped in different coloured dots.  I’m going to have to think about how this data could be displayed a little more effectively.

As previously mentioned, I attended a Digital Humanities event in Edinburgh on Friday.  It was an all-day event and on the whole it was very useful.  Andrew Prescott gave a hugely interesting talk about the current state of digitisation and digital editions and there was a great session where attendees could discuss their project in 5 minutes.  I gave a talk about the Mapping Metaphor visualisations, and it was great to hear the other speakers too, especially the other projects there were also dealing with visualisations.  The organisers are hoping to launch a Scotland-wide Digital Humanities Network soon, which I think has a lot of potential.

Other than the above I did a few other tasks this week, including a little work on the Choral Burns project website, a little work on the Cognitive Toponymy website and I also was in contact with Peter at SLD about the API he has been working on.  He has now reached a point where he is ready to give me access to the API.  I’ve had a quick look at it and it seems pretty straightforward to follow, which is very good.  I aim to start working on connecting the front-end to the API next week, all being well.