[14] Down the rabbit hole we go

Term 1, Year 2 completed! This term has been quite light in that I have only had two modules, though as they both had group projects I found I spent a lot of time working with other students to get work done.

I’ve found the engagement at the start of the year to be a lot less, and while it wasn’t so daunting in terms of the modules – the gaps in knowledge regarding the project were quite big. Laurie has stepped in and managed to help get us over the line but I do feel like if we had the input earlier in the term I wouldn’t be feeling rushed.

I’ve done a lot of self-reading in the interim and am going to make myself a reading list for the Christmas holidays to help get moving with project bits and term 2 modules.

Text Remix – I found this course a bit disjointed. I think it is the nature of having it spread across 10 weeks with 2 hour lessons. The 2 hours feel quite rushed for the content the tutors want to cover but the interim weeks were quiet. I appreciate the attempt to teach us python but it was very narrow in its approach. However, I really enjoyed the course work. Both the group project and individual project were quite creative and fun. It really helped me think about how qual data could be chopped up to look for patterns and twists.
Democracy, Data and the Rule of Law – this course was so interesting. I took it because I was interested in how governments can use technology for better resource distribution. John was focusing mostly on how to legislate for the governments use of technology and the ethical and legal dilemmas but it opened up a lot of areas of thought. I think the reading list will be helpful when I approach my project.

Since I submitted the 1000 word blog reflection, I’ve had some very interesting conversations about my project.

Firstly I met a company that explore some of the ‘big questions’ around data/ai/tech infrastructure as documentaries. They are a mix of speculative futures and interviews, which is very interesting. The owner was interested in partnering with me to pull together content on some of the big ethical questions around AI. My main concern was if this is something that can be done by the time the project is due. However, after talking to Laurie, I’ve realised that really I want to step further away from data ethics and focus more on resource distribution and improving on inequalities that currently exist (and will be exacerbated) due to the adoption of technology like ML/AI.

That brings me onto my second conversation with Code for Africa. I met a Senior Programme Lead who has an interactive tool that looks at wealth inequalities in different countries and regions by gender/age etc. We discussed expanding the tool with additional datasets to help predict what the impact might be on wealth equality when factors are modified. This is strongly aligned with the three things I think I’ve nailed down what I want to focus on:

Helping governments, industry and third sector use technology to make good decisions on how to improve wellbeing for citizens
Cash transfer programs – how they can be used to improve wellbeing and offset impacts from a changing industrial landscape
Showing that if we de-center the concept of productivity from wealth inequality, greater social wellbeing can be achieved

My concern with this second option is that my data engineering skills are somewhat limited and so the team at Cfa would need to be working with me to build the product and I would be more focused on establishing what alternative datasets are useful, why and what parameters are needed in the different regions for ‘modification’ (e.g. South Africa has old age grants to pensioners, so if we increase/decrease this how does it impact the status quo). I will have a further meeting in early Jan to discuss more.

Failing that I do have the fall-back of PIP. I can look at it through the lens of digitisation of government services and how that can improve/downgrade access to cash transfer programs.

I think over December I will have to have two reading lists to tackle to balance both projects until I know for certain which one I can follow.