Notices by Greg Wilson (gvwilson@mastodon.social), page 5

"In the future we won't need programmers - just people who can describe to a computer precisely what they want it to do." (original source unknown)

When the signal came, there was nothing subtle about it: every radio telescope on the planet redlined. First primes, then simple arithmetic, then basic vocabulary, each burst richer than the one before.
Finally humanity received the most complex message yet. Thousands of researchers and ad hoc internet communities raced to decode it.
"Are you crabs yet?"
Hesitantly, humanity replied, "No."
"Oh," came the reply. "Sorry to bother you."
Silence followed.

I have now rewritten two projects to use the PyPika query builder (https://pypika.readthedocs.io/) instead of an object-relational mapper, and the query builder approach had many fewer WTF moments. Caveat: they are small projects with relatively simple data models (a dozen to a couple of dozen tables, no need for recursive queries) and I haven't had to maintain the code through schema changes yet.

The average human being is about 733 farts in size. (66 L per person / 90 ml per fart) You're welcome, and good night.

Wendorff's "Politics in Software Development" contains a lot of useful nuggets, but I'm struggling to find an explanatory arc (the teaching equivalent of a story line), and the author has somehow managed to make a fascinating and contentious subject feel a bit dull. ☆☆

Once more with feeling: I am _not_ a "thought leader". I am at best a thought recycler, and most days I'm more of a thought raccoon.

This just in from the Ministry of Subtle Bugs:

def f():
print("false")
return False

result = all(func() for func in [f, f, f])
print("just comprehension", result)

only prints 'false' once before printing "just comprehension", but

result = all(list(func() for func in [f, f, f]))
print("with list", result)

prints 'false' three times before printing "with list". I understand why, but damn, this one was hard to spot.

@eaton thanks - do you have any pointers to open-access lessons that demonstrate this approach for security and/or devops, preferably designed for one-day workshops?

I have tried twice to put together a hands-on full-day "security for data scientists" tutorial but am finding it just as hard as a "systems programming/devops for data scientists" tutorial: the topics are so scattered and disjoint, and require so much random background knowledge, that the material doesn't hang together and will probably just frustrate participants. I would be very grateful for pointers to open-access lessons on either topic that actually work - thanks in advance.

Prediction #1: the Republicans will use every legal trick they can think of to prevent Harris (or anyone else) from using the $90M that Biden has raised.
Prediction #2: thanks to their 40-year focus on politicizing the judiciary, they will succeed.

iNaturalist is looking for a Head of Engineering https://app.beapplied.com/apply/gotyyjtuxa must be US-based, and the "technical expertise" section reads like "please know everything", but if you care about conservation and open science, this is a chance to make the world a slightly better place.

I am standing for election to the board of the Python Software Foundation - you can read my pitch and those of the other candidates at https://www.python.org/nominations/elections/2024-python-software-foundation-board/nominees/ . Cc @ThePSF

" Ford government closing Ontario Science Centre today after report found roof in danger of collapsing: Friday is the last day of operations for the once-popular tourist attraction that successive Progressive Conservative and Liberal governments have allowed to deteriorate for years." https://www.thestar.com/politics/provincial/ford-government-closing-ontario-science-centre-today-after-report-found-roof-in-danger-of-collapsing/article_3e7a8442-2fd8-11ef-9c00-03276c11fe83.html

Pioneering computer scientist, electrical engineer, and transgender activist Lynn Ann Conway has died at the age of 86. Every modern silicon chip incorporates her ideas, and her courage was an inspiration to many. https://en.wikipedia.org/wiki/Lynn_Conway Long may she run...

I look forward to the day when American billionaires pumping money into reactionary Canadian front organizations is investigated as "foreign interference".

I think a book comparing Big Tech to drug cartels would be awesome. Both make money from selling addictive products and shrugging off externalities; both needed new offerings to bring in new revenue (AI is to Google what meth was to the Juarez cartel); both regard the legal system as merely an operating cost; and both tend to be run by egomaniacal sociopaths. (I expect Musk and Escobar would have gotten on well.) If I knew enough, I'd start today; if you do, please start tomorrow. 2/2

I _really_ enjoyed Wainwright's "Narconomics", which explores the economics of the illegal drug trade. It turns out that (for example) HR is more of a problem than you'd think: good accountants know they can actually make more money on Wall Street where crime is legal. Marketing, dispute resolution, quality control: it's a very fun read. But it sparked a thought… 1/2

@powersoffour @yabellini the problem I have with worldbuilding is the same problem I have with writing tools for books instead of books themselves: I could spend _forever_ "getting ready" and never get around to what I was supposed to be supporting.

@yabellini I can't think of anything I already know well enough to teach that I haven't taught - if I want to teach something new, I'll have to learn it myself first.

@yabellini like worldbuilding, that's something that other people have written about far better than I could.