End-to-end hardware and software solutions for community-based revitalisation are a tricky proposition and remain a major barrier for community-based revitalisation projects to get on with their valuable work. Over the last few months I've come to realise that this is actually what I'm researching. This whole period has resulted in a lot of agonising, reading papers, looking at countless different software solutions, and writing and re-writing my research objectives. That's why this blog has been quiet. Finally, I understood that my research is focused on how communities can adopt existing tools and, by extension, discovering the 'holes' or what tools ought to exist.
The emerging plan for 2016 was to come up with a reasonable set of off-the-shelf tools, build the minimal components needed to do work (usually transferring from formats and so on) and then evaluating that in Taiwan. These few words don't do justice to the confusing and agonising process of working out what to investigate and how. In the end, I resolved to come up with something modest and then test it on not one, but several different environments in Taiwan. That's the only way to yield good data that can be generalised outside of a single fieldwork environment. A 13-ICAL paper was influential on my thinking, "Collaborating with Speech Communities in Language Documentation", by Cruz and Gonzalez. The used the BOLD method (an audio-based method to cut down on detailed transcription in the field) which immediately drew my eye.
Cruz & Gonzalez analysed three language documentation projects on the Klata, Isinai & Guinaang Kalinga languages in the Philippines, with a view to identifying factors that helped or hampered collaboration. Their focus was collaboration in general, but I realised that the adoption of technological tools ought to be examined in a similar way. How do we know if digital tools will be helpful in particular settings? What factors must we address, both in our work with the community and perhaps even embodied within our software? The other important aspect of this approach is that it becomes more straightforward to figure out how to investigate. Any research student will tell you that one of the biggest problems is getting clarity on research methods. This was a bit of a hallelujah for me, given that clear steps replaced a lot of what I called 'handwaving" in my last blog post.
This bit of good fortune was met with yet more in early 2016...
Steven asked me if I was interested in working on building a web-based tool to produce annotations for spoken language? I'd been thinking about this for some time. The need for such a tool is great indeed and it's a surprise that no one has done it (well) before. This situation is akin to this; imagine you are planning on a big trip to pick up a buried treasure you know about but the road is washed out and you're not really sure how you're going to get past. You've been whining about the washed out road for ages. Then someone offers to pay you fix the road so that you can get to the pot of gold. It's pretty much the dream scenario. That's where I've been since February. I've been working very hard on building an annotation system based on web technologies, specifically a Chrome App so that it can be 'installed' on computers to work offline.
It's going brilliantly, but it's hard work and long hours filled with the feeling that it's not going fast enough. I've got just two weeks until I head back to Taiwan. I plan to spend a couple of weeks in Taipei and iron out the bugs and properly organise the internationalisation (into Chinese), then head up to the mountains and try it out on at least one language, Saisiyat, and hopefully Atayal as well. It's great to have clarity on what I will do in Taiwan on this trip, but also this time the ball is firmly in my court to get things ready. This often boils down to deciding what features in the app, working title Annoweb, need to be complete to undertake this fieldwork. Many of the big ideas wont make the cut this time, like cloud-storage, social media sharing, working on video, and more. Yet getting this thing to the line for a beta also marks the point that I can flick the GitHub repository onto public, aim to recruit some other participants with luck.
Somehow I also need to find time to address the logistics of fieldwork and research methods, and do some writing of words that will actually appear in the academic record. Something that up to now has been too far on the backburner. The joy about this plan is that I'm clear about what I can report in the academic record, at conferences and other opportunities. There's a digital tools symposium being hosted by my other supervisor, language documentation & tools guru Nick Thieberger, soon after I get back. At that stage, I will be well and truly battle hardened to this challenge and I'll be extremely interested in connecting to like-minded people in the linguistic digital tools environment.
The next fieldwork trip will be much easier than the last one. It's hard to overstate the errors I made, the lessons I learned, but I'm virtually counting on hitting the ground running this time. Let's hope my hard-won Mandarin skills will spring back quickly, I'll need them soon enough!