🥰

Gots to gots to gots to crash.

#threegoodthings

1) My beloved has had a hard brushfire-laden week, & there were some days I was very concerned about her. She partied some yesterday, & seems in good spirits today.

2) I am looking forward to work tomorrow. See another thread for the part I think will bring me joy to do.

3) Molly the dog, my precious friend, snores, ever so gently, and it makes me remember the prose poem "The Desiderata":

"doubtless, the universe is unfolding as it should".

Happy Pride Month!

I’m a proud LGBTQ+ artist working in raku and traditional pottery — and I’d be so honored if you share my work.

This Pride, I’m reflecting on something personal: I never thought I’d make it to this age.

But I did.

And I’m still here.

Creating. Grateful.

Alive 💖

#pride #pottery #lgbtq

(The last time I spoke of TDD, some well-meaning soul explained to me what TDD really meant. I'm in an okay mood tonight, but I would not try your luck, cuz I can turn right quick.

I have been studying, doing, learning, and teaching test-driven development for longer than all but a handful of people, longer than many folks in our trade have been potty-trained, and I can get pissy about it.)

But those are the big 3: 1) "TDD is a quality technique." 2) "Large-scale tests are good." 3) "Design first, test later."

And it doesn't help, of course, that fifth-rate epigones get mad consultant bucks to spew this bullshit far and wide, preferably with sexy slides and no actual experience whatsoever.

There are other issues: a foolish and over-simple purity that ignores the role of judgement. a lotta shu-ha-ri/training wheels shit that somehow never advances beyond shu, a lotta failure to explain how to jiggle designs to get easy testability, a *gigantic* over-emphasis on auto-mocking tools, bullshit moralizing about what a bad person you are if you don't do it, the idea that one writes a test w/no notion of how it might be passed.

3) The notion that one can design and create code w/o respect to tests and testability, and then readily test it.

This is nonsense. Code that was not written to be tested *can* be tested, but typically only at very great expense.

In true TDD, tests and testability are first-class participants in design. Ignoring that idea, which is called the "Steering Principle", results in code that is, once again, expensive to test, if it can be tested at all.

That's the #2 reason developers don't TDD.

2) The notion that large-scale tests -- typically not-quite-end to not-quite end -- are desirable TDD tests.

This is the #1 reason that most developers *hate* what their shops require of them re: testing. Such tests are extremely expensive to create, and generally, tho I freely admit not always, a waste of time: they are hard to write, slow to run, and when they break, they're too big to know what went wrong where.

1) The notion that the reason we do TDD is to prove our application works.

The real reason we do TDD is because it makes us more productive. This misunderstanding is what makes just about every middle manager on god's green earth tell their team to skip the tests to get the feature faster.

I have strong views about how to do TDD and how to learn to do TDD. Over the last 20 years, I've introduced new coinages, reframed some concepts, and written dozens of articles about how I do what I do and why tt works so well.

Most of the TDD instruction I see is, in my view, shite.

I believe this is a big part of why TDD isn't usually practiced in the trade, even, sadly, in the shops that self-describe as doing it.

Three issues I see over and over again.

Sunday night, pushin' towards 1 am. We'll have a beer and a shot of rye.

Tomorrow -- I always start thinking of work on Sunday night, and I don't even have a full-time job -- one of my tasks is to make the initial draft of a new TDD module for the students.

I wrote about half of Industrial Logic's Testing & Refactoring Workshop almost 20 years ago. Since then, of course, I have learned a whole lot more, about TDD, about mircrotest TDD, and about creating curriculum.

Looking forward to it.

I toast this beer to Judy. I hope, wherever she is, she is well.

It's a cover, of course, of a very famous song by someone else.

But I have known and loved this take for not quite four decades.

Emmylou Harris, "The Bottle Let Me Down".

https://www.youtube.com/watch?v=lrs0KUHPeKg

Sometimes I write a joke that I love very much but it just falls flat. In those moments, I get a glimpse of how God must feel about me.

This is a take for the ages, in my view.

Sly and the Family Stone: "If You Want Me to Stay".

https://www.youtube.com/watch?v=gZFabOuF4Ps

Disinformation kills

When stars reach the show in pop, we tend to focus on their charts and sales.

Knight & the Pips certainly dominated the charts, and made mad bucks from the sales.

But when you think about it that way, you miss how fucking *good* she was as a singer and performer.

Knight is incomparable.