– We don't write tests. – Why? – Because we don’t have time for it. – Why? – Because there is too much work and pressure. – Why? – Because we don’t move fast enough. – Why? – Because changing software has become difficult and risky. – Why? – Because we don’t write tests.

The discount on $STNE stems from a deterioration in its loan portfolio, not because Brazil has been forgotten. Today’s bullish candle has already proven the latter part of that statement wrong. It closed at 9.55 on Friday. Today it opened at 10.88, with the low matching the opening price and the high at 12.06; it hovered around 11.6 in the afternoon, up about 21%, with trading volume exceeding 20 million shares. The entire pre-market range of 9.10–10.00 was left far behind. Both 10.00 and 10.80 were breached. The 52-week low of 8.95 was not retested. Results from the first round of Brazil’s presidential election were announced last night. Conservative candidate Flávio Bolsonaro received approximately 47.3% of the vote, while incumbent Lula received approximately 44.8%. The market interpreted the right-wing lead as signaling a shift in tax rates and economic policy, triggering a broad buying spree in Brazilian assets. This is an “election beta,” not a decline in loan loss provisions. Second-quarter revenue was 3.59 billion reais, up only 2.5% year-over-year. Total payments rose 4% year-over-year, but customer retention has yet to improve. The loan portfolio doubled year-over-year to 3.8 billion reais, with loan income up 14%; adjusted gross profit remained largely flat at 1.56 billion reais, as provisions ate into growth. Delinquencies between 15 and 90 days rose from 2.51% a year ago to 6.00%, while those over 90 days stood at 8.60%. The cost of risk accounted for 21.5% of the loan portfolio. Retail deposits totaled 10.8 billion, up 22% year-over-year, with 4.8 million active customers. Full-year adjusted earnings per share guidance remains at 10.8–11.4 reais; the company itself states that, given high interest rates, it prioritizes meeting the lower end of the range. The price-to-earnings ratio is approximately 4x, with a market capitalization of roughly 2.2–3.0 billion. The average sell price is 14.30, and the consensus rating is Hold. Last Thursday and Friday, trading volume surged in call options expiring October 16 with a strike price of 9.47, with open interest jumping from about 3,200 contracts to over 14,000. That was a short-term bet near the low point, not an order to double the investment in three years. Today, the price has already moved away from that strike price. The election could bring Brazil back for a day. Past-due loans and loan loss reserves are still on the loan books. The next chart to watch will show whether the rate stays at 6% over the next 15–90 days—not today’s bullish candle. This is not investment advice. #QQQ #STOCK #STNE

Strategy 1: Trying to be *irreplaceable* via knowledge/skills hoarding — you're bad for business, they *want* to get rid of you, but they can’t because they’re stuck. Losing you hurts temporarily, benefits permanently.

Whatever you are not changing, you are choosing.

You can have a PhD and still be an idiot! Education and intelligence are two different things.

Refraining from writing code is more challenging than actually writing it.

a good pro-HTML site/article: html-first.com/ the hypermedia renaissance is building!

"International Tax Competitiveness Index 2023" is misleading because the world is much bigger than the OECD. It tricks people into thinking that Estonia and Latvia have the best tax code internationally. Among OECD countries – well, maybe. Internationally – hell no. On that list, there is not a single tax-friendly country that I would consider setting up a company: UAE, HK, Singapore, Malaysia (Labuan), Georgia, and the list goes on. taxfoundation.org/research/a…

Code is cheap. Show me delivered value.