Got to love Altibox. They sent a message to warn me that they were doing maintenance so my internet would be going offline. Ten minutes after it came back on.

Left my walking boots on the garden wall to dry. Came outside to find the neighbours' dog growling at them.

For short term testing and other pragmatic purposes, I am doing calculations that are both (a) bad code, and (b) not entirely scientifically correct. This is making me very sad.

Excessively humid weather for hiking today, so just a short one. Found some spooky dead trees though.

It's 9:45pm. My neighbour just came round to borrow some eggs. 🤷‍♂️

Calling DevOps' bluff during system reconfiguration:

DevOps: Do you want to do it yourself or should I?
Me: I'll give it a go. What could possibly go wrong?
DevOps: Wait a second.

Watching a presentation describing how fast data access can improving forecasting efficacy. The illustration using BBC-era John Kettley was a bit of a surprise.

And of course I have Tribe Of Toffs going round my head now...

John Kettley is a weatherman,
a weatherman,
a weatherman,
John Kettley is a weatherman,
And so is Michael Fish.

Only an Oxbridge college would have an entry on its Wikipedia page titled "Customs" which starts with the phrase "Every hundred years..."

There are two sets of characters for Chinese numerals: one for everyday writing, and one for use in commercial or financial contexts.

The latter arose because the characters used for writing numerals are geometrically simple, so simply using those numerals cannot prevent forgeries.

A forger could easily change the everyday characters 三十 (30) to 五千 (5000) just by adding a few strokes. That would not be possible when writing using the financial characters 參拾 (30) and 伍仟 (5000).


First experiments with tree supports with - they're giving me some distinct baobab vibes!

I'm seriously impressed with the algorithm's ability to navigate around parts it shouldn't touch, but it did end up leaving some fairly large areas unsupported. Perhaps I pushed it too far. I think if I mess with the layout I can get those parts into places where they won't be seen, so it should still turn out well.

Time to decide which dumb data format to deal with next. Options are:

1. One file per record, different fields stored as separate files in separate folders
2. CSV, but columns are only unique if you merge three column header rows together
3. Date is given as decimal day of year, but the file contains no year. Which is critical.

One of those magic eye pictures, but the hidden image is a migraine.

Google is now tracking your visitors for advertising purposes even when you’re not using Google Analytics or Google Ads.

I took a closer look at Google’s FLoC initiative and how you can fight back and opt-out as a web user and a web developer

Just read a hot take by someone who fancies himself a great leader explaining how he expects his employees to work extra hours to make up for the time lost on public holidays.

joy: just received a data file with the column header line:


The file has a header explaining what the column headers mean. 🤦‍♂️

Show older

Server run by the main developers of the project 🐘 It is not focused on any particular niche interest - everyone is welcome as long as you follow our code of conduct!