Also, to make sure there are no linguistic roots left over on your system that it might grow from again, add --no-preserve-root.

I never push to production 😎

I give my users a dev preview and ask for feedback before finalising, they never respond and I find out later they’re using it daily.

I once worked on a codebase where the reset function had a hardcoded default password

Aren’t they fractions rather than floating point decimals?

I’m thankful I don’t do software dev (I did two years as a working student, that was enough), but working in Data Engineering / Analytics* doesn’t make things better. I’ll overengineer the database, ETL and reporting, define a dozen measures I’ll never use, prepare a dozen ways to slice and view the data I’ll never look at and build a whole data warehouse I’ll never look at.

Eventually I remember that it exists, realise that I’ve answered all my questions by directly querying the database, except for “What am I running out of?”, which I answer by looking in the cabinet because I never update my inventory anyway.

*I don’t even know where the line is anymore and how much of my responsibilities is on either side of it

“6 for the beer, 9 for the longdrink, 4.20 for the water… That’s a total of 694.20 please.”

He uses a pencil, paper and a mechanical calculator to tally up the bill, which I absolutely understand when your career is in IT.

When the alternative is either having to search, evaluate, compare, select and configure an application for that purpose that you’re never quite happy with, or to scope, design, develop, test, deploy, maintain, eternally find things you wish you’d done better, refactor, realise you’re spending your free time on doing more of your job, regret your life choices, resolve to only make this last improvement and then call it good enough, renege on that promise to yourself a week later, burn out, curse that damn app for ruining your hobby…

…yeah, using the most trivial low-tech solution possible does look rather sensible.

My new data structure:

Given a heuristic for determining data quality, it homogenises the quality of its contents. Data you write to it has pieces exchanged with other entries depending on its quality. The lower the quality, the higher the rate of exchange.

If you put only perfect data, nothing is exchanged. Put high quality, you’ll mostly get high quality too, but probably with some errors. Put in garbage, it starts poisoning the rest of the data. Garbage in, garbage out.

“Why would you want that”, you ask? Wrong question, buddy - how about “Do you want to be left behind when this new data quality management technology takes off?” And if that doesn’t convince you, let me dig around my buzzword budget to see if I can throw some “Make Investors Drool And Swoon”-skills your way to convince you I’ll turn your crap data into gold.

I think it’s a symptom of the age-old issue of missing QA: Without solid QA you have no figures on how often your human solutions get things wrong, how often your AI does and how it stacks up.

ActiveSheet? Please no

Butterfly gang

Ooh that’s a good one