List items
Items from the current list are shown below.
Waste
All items from August 2022
28 Aug 2022 : Three years of waste data #
I've been collecting data on my recycling and waste output since August 2019, which means I've just hit the three year mark. That's honestly longer than I thought I'd manage to keep it up for. You can see graphs of the full set of data on my waste pages. It might be getting to the stage where it's worth checking for trends in the data, but I've not tried yet. Overall my waste output hasn't changed a huge amount over the years. It dropped down for two years, but I'm on track for it to go up slightly this year. Overall my average output over the three years has been 160.87 g/day, equivalent to 58.7 kg/year.
28 Aug 2022 : Waste data dump #
My latest waste data dump is now up on the waste page. Notably, this was my first recycling that included metal (foil cartons). Unfortunately I had a whole batch of bread go mouldy before it was opened, which bumped up my compost output. Most of the other categories were fairly average, leaving me with an average output for the fortnight of 134.14 g/day, slightly higher than my 129.46 g/day average for 2022, but better than I expected given the bread situation. Plus at least finally the graphs are actually rendering correctly again, which is a relief.
27 Aug 2022 : Histographs fixed #
In my last waste data report I lamented the fact that the histograph portion of my graph rendering algorithm was failing, leaving the graphs on my waste page unhelpfully blank. After much digging around in the code I eventually narrowed the problem down to the x-axis limits. Dates in python can be converted to and from ordinals. As the documentation explains, the ordinal values are "the proleptic Gregorian ordinal of the date, where January 1 of year 1 has ordinal 1". Today (27th August 2022), for example, is 738394. So it turns out that I was setting the axis limits to be dates, while the data point x values were being set as ordinals. This worked fine for matplotlib 3.1.2 on Ubuntu 20.04, but broke after upgrading to matplotlib version 3.5.1 on Ubuntu 22.04. The fix was easy once I'd figured out the cause: simply use ordinals throughout and add a custom renderer for the x-axis text so that the values are still shown as dates. Arguably it should never have worked and my original implementation was faulty, so I'm happy the code is cleaner and more correct now, but more importantly it also means my waste graphs are now shown correctly again on the waste page. And I have one less niggling annoyance to deal with occupying the back of my mind!
14 Aug 2022 : Waste data dump #
I've uploaded my latest waste data to the waste page. My daily average over the last fortnight of 141.13 g is rather high, and looking at the graphs it seems to be due to increased compost and plastic output. The increased plastic is explainable: I bought a 3 litre bottle of cooking oil which just ran out. But while the larger bottle gives a bump when it's eventually disposed of, I'm reasoning that overall it will help bring down my average, due ot the higher content-to-bottle ratio. The increased compost I can't explain though. It didn't even seem that much to me. Unfortunately while the numbers are in, the main graphs aren't. An operating system upgrade seems to have had a bad effect on my histocurve graphs, which I'm still trying to figure out. Hopefully with a bit of debugging I'll be able to return the graphs soon.