This week I had the joy of prepping a talk for the LangVIEW pre-conference workshop for #MPaL2020
I talked about Design Thinking for data collection in multilingual environments
1/

Two key concepts in design thinking are
the co-evolution of the problem and the solution
an iterative design process with multiple stages of feedback - grounded in empathy for the end user of the tool (eg the parent completing the survey, the child doing the task)

When we try to adapt tools designed in one context, it’s typical to assume that ‘translation’ will do the job of localization, but some core features of the original context are hard wired into the resulting tools...

Our @BLIPLabNTU in Singapore has struggled with some of these challenges in our multilingual context, complete with high levels of multilingual inputs for babies!

(oops - that reference to @FeitingW ‘s Masters dissertation should read 2018)

Our tool development process sometimes requires going back to first principles, rather than adapting existing tools... this is a lengthy but worthwhile process
(so many steps!)

All to overcome the features of the research context that are baked into existing standardised tools...

One of the hardest challenges is how languages can (or should) be treated as discrete for people in massively multilingual communities with high levels of code switching/mixing/trans-languaging

Perhaps words are various fishes in an ocean of linguistic possibility

Also, we must recognise that resources are not equitable distributed across research contexts: The environments with the lowest resources often have the highest complexities

So our community of scholars with an interest in language development has work to do finding meaningful ways of supporting research in resource poor contexts, with empathy for the participants in, and beneficiaries of our research at the core of our work practices /.