Last week

Priorities

I chatted to Anil about priorities, as I had a new thing to work on this week that wasn’t on my list last week, and so my current focus is:

1. LIFE visualisations
2. TMF if that’s blocked and needs me to get involved
3. FSArk stuff

LIFE visualisations

Most of the week was dedicated to a task I didn’t have on my list when I wrote last week’s notes: prototyping some interactive visualisations for the LIFE metric. JS/React really isn’t in my wheelhouse, but I think that’s true of all our group, so I did as best as I could in the time I had: we had a deadline to get something for Friday so Anil had something to show for an important meeting on Friday.

The first hurdle was just updating the quantify.earth codebase as it hadn’t really been touched for a year. In that time the JS world moves forward quite a bit, and so I spent a day just updating libraries and fixing the breakages from that. This was both good to do from a security type thing, but also I needed some newer features on the deck.gl package we use to let me put things on a globe view.

In the past Patrick’s had success using globe.gl rather than deck.gl, but globe.gl doesn’t have quite as many layer types as deck.gl, and I needed to be able to switch between many raster layers whilst keeping a base map layer underneath. In the end I got something working that lets me take the LIFE data rasters, turn them into map tiles, and then have a quick way to turn them on and off.

As an aside, we might want to reflect that any practiced web developer will look at what I’ve done and feel unwell and want to fix it all up, as I brute force my way to a solution as best I can given that 1) I have a tight deadline and 2) am not invested in web development and just want to get it out of the way so I can go back to what I’d like to be doing. You can now perhaps replace me in this scenario with an ecologist trying to get results out of out computers and the web developer with me doing what I’ve done for the last year. Doesn’t paint me in a good light, but I suspect it’s an embarrassingly true comparison.

Late last year I went to a climate unconference and had a chat with someone working in interaction design about how can we make it easier for ecologists to Do The Right Thing™ CS wise, and she pointed out that I was operating a bit too low in the scale - not that I was wrong, but really there was a lot of work to be done above me. The problems she argued were systemic in the system: we don’t have the right mix of people to do the job right, so we end up with people just bodging through as best they can and that’s bound to be sub-optimal. I feel this week was a nice reminder of this from being inside looking out rather than outside looking in. Whilst I don’t disagree there is a lot of CS work that can be done to make it easier to do the right thing, getting people to want to use these will require some higher order systemic change.

My thanks to Patrick for help with this - in particular getting it deployed, which was also a somewhat painful process because we’ve not had to deploy new tile sets in over a year and both the code and our memories of now we do that had atrophied.

Having started with this, I’m going to chat to @Tom this week about what we actually want on this visualisation tool, as the email thread we started last week didn’t really start much discussion.

PROPL

I’ve done an initial read through all the talk abstracts for PROPL that I have, and need to find time this week to sit down with Patrick and plan out the session we’re chairing/discussion leading.

I kinda wish that Ali was going to be there, as I feel it’d be great to have someone on the other side of the fence there to counter the sense I get of the CS people trying to shoehorn their pet research area into something else. I feel sharing the pain we’ve had on the TMFEMI work as a spark to discuss the session papers in context is how I’d like us to go.

Sysadmin

I had to take down and bring back up all the 4C servers for the power outage at WGB this weekend. This mostly was plain sailing, except for the DB server, which had ran out of disk space due to excessive Postgres logs, so I did a bit of work there to fix that and get it updated again.

This coming week:

• Prep for PROPL on Saturday
• Work out what it is we want on LIFE visualisations
• Meeting with Cyrus on Monday