https://voiddo.com/blog Notes on scrapers, trading bots and small tools. How they are built and how they fail. en https://voiddo.com/blog/spring-cleaning-what-we-retiredhttps://voiddo.com/blog/spring-cleaning-what-we-retiredTue, 20 Oct 2026 12:00:00 +0000We asked three questions of everything we had published: does it work, do we use it, is it true. Here is what did not survive, and what the cull taught us. https://voiddo.com/blog/best-of-a-hundred-coin-flipshttps://voiddo.com/blog/best-of-a-hundred-coin-flipsTue, 20 Oct 2026 12:00:00 +0000Try enough strategies and one of them will look brilliant, even when every one is a coin toss. The number of tries is part of the result, and it is the part nobody prints. https://voiddo.com/blog/robots-txt-readers-disagreehttps://voiddo.com/blog/robots-txt-readers-disagreeTue, 20 Oct 2026 12:00:00 +0000Python's built-in reader takes the first matching rule and ignores wildcards. The standard takes the longest match. We ran both on 12,000 random files and listed where they split. https://voiddo.com/blog/cron-does-not-mean-what-you-thinkhttps://voiddo.com/blog/cron-does-not-mean-what-you-thinkTue, 20 Oct 2026 12:00:00 +0000Seven ways a crontab line does something other than what it reads like: step values that restart, two day fields that mean "or", a percent sign that eats your command. https://voiddo.com/blog/the-price-is-a-stringhttps://voiddo.com/blog/the-price-is-a-stringTue, 20 Oct 2026 12:00:00 +0000Shops print the same number a dozen ways. A careful parser reads what is clear, leaves the rest empty and never fills a gap with a plausible guess. https://voiddo.com/blog/slippage-is-a-fee-nobody-printshttps://voiddo.com/blog/slippage-is-a-fee-nobody-printsTue, 20 Oct 2026 12:00:00 +0000The exchange prints its fee. It does not print the spread, the thin book or the stop that fills worse than its trigger. A test that leaves them out is flattering itself. https://voiddo.com/blog/the-key-is-already-publichttps://voiddo.com/blog/the-key-is-already-publicTue, 20 Oct 2026 12:00:00 +0000Deleting the commit is not the fix. Revoke the key first, check who used it, then clean up and put a scanner in front of the next push. https://voiddo.com/blog/a-timestamp-without-a-zone-is-a-guesshttps://voiddo.com/blog/a-timestamp-without-a-zone-is-a-guessTue, 20 Oct 2026 12:00:00 +0000Scraped dates arrive as "2026-03-08 02:30" with nothing else. Some of those times never happened, some happened twice, and the gap between two cities moves during the year. https://voiddo.com/blog/retry-after-is-a-sentencehttps://voiddo.com/blog/retry-after-is-a-sentenceTue, 20 Oct 2026 12:00:00 +0000When a site says "slow down", it often says how long to wait. Reading that line, backing off with jitter and knowing when to give up is most of polite scraping. https://voiddo.com/blog/start-from-the-taskhttps://voiddo.com/blog/start-from-the-taskTue, 20 Oct 2026 12:00:00 +0000A trading bot project begins with what the bot is for, what it may lose and what it must never do. The signal comes last, and it is the smallest part of the work. https://voiddo.com/blog/what-apify-taught-ushttps://voiddo.com/blog/what-apify-taught-usTue, 20 Oct 2026 12:00:00 +0000Seven things we learned about publishing on the Apify Store: the automated tester, identity checks, rate-limited visibility changes, pay per result and the first "users". https://voiddo.com/blog/a-stop-in-your-bot-is-not-a-stophttps://voiddo.com/blog/a-stop-in-your-bot-is-not-a-stopTue, 20 Oct 2026 12:00:00 +0000When the process crashes, the server reboots or the connection drops, a stop that exists only in the bot's memory does not exist. The stop belongs on the exchange, at entry. https://voiddo.com/blog/how-a-scraper-job-goeshttps://voiddo.com/blog/how-a-scraper-job-goesTue, 20 Oct 2026 12:00:00 +0000Read the site, build, test on real pages, ship one clean schema, check it every day, rebuild when the site changes. Most of the work is the last two. https://voiddo.com/blog/did-my-order-fillhttps://voiddo.com/blog/did-my-order-fillTue, 20 Oct 2026 12:00:00 +0000A timeout is not a failure. It is a third state between success and failure, and the safe answer to "still unknown" is to stop opening new positions. https://voiddo.com/blog/robots-txt-firsthttps://voiddo.com/blog/robots-txt-firstTue, 20 Oct 2026 12:00:00 +0000Every URL is checked against the site's robots.txt before it is fetched. When a site puts up a wall, the run stops and says so in plain words. https://voiddo.com/blog/the-fee-line-eats-the-gridhttps://voiddo.com/blog/the-fee-line-eats-the-gridTue, 20 Oct 2026 12:00:00 +0000Grid and DCA bots advertise small, steady profits. Each step's gross edge is often smaller than the round trip fees plus the spread, and a backtest without fees can get the sign of the result wrong. https://voiddo.com/blog/what-a-scraper-can-and-cannot-dohttps://voiddo.com/blog/what-a-scraper-can-and-cannot-doTue, 20 Oct 2026 12:00:00 +0000A plain list for anyone thinking of buying one: what is realistic, what is not, and what we will tell you no to. https://voiddo.com/blog/testing-a-bot-against-a-fake-exchangehttps://voiddo.com/blog/testing-a-bot-against-a-fake-exchangeTue, 20 Oct 2026 12:00:00 +0000The dangerous code paths in a trading bot are the rare ones: timeouts, partial fills, rejects, duplicate ids. A fake exchange that fails on command lets you hit each of them on purpose, every time the tests run. https://voiddo.com/blog/proxy-ladderhttps://voiddo.com/blog/proxy-ladderTue, 20 Oct 2026 12:00:00 +0000Direct first, then datacenter, then residential. Residential traffic is billed by the gigabyte, so the ladder is a cost control as much as a technique. https://voiddo.com/blog/every-strategy-has-to-beat-a-coin-tosshttps://voiddo.com/blog/every-strategy-has-to-beat-a-coin-tossTue, 20 Oct 2026 12:00:00 +0000Run the same sizing, stops and fees with seeded random entries across many seeds and you get a distribution of luck. A strategy is interesting only if it clearly sits outside it. https://voiddo.com/blog/what-we-will-not-scrapehttps://voiddo.com/blog/what-we-will-not-scrapeTue, 20 Oct 2026 12:00:00 +0000A list of sites we decided not to build for. The reasons are general, which is the point: most of them are not about technique. https://voiddo.com/blog/walk-forward-vs-in-samplehttps://voiddo.com/blog/walk-forward-vs-in-sampleTue, 20 Oct 2026 12:00:00 +0000Optimizing on all of the history finds the best past, not a good future. Fit on one window, judge on the next untouched one, roll forward, and count how many variants you tried. https://voiddo.com/blog/kill-switches-people-actually-rearmhttps://voiddo.com/blog/kill-switches-people-actually-rearmTue, 20 Oct 2026 12:00:00 +0000Soft pauses and hard halts: which conditions may clear themselves, which need a person, and why a re-arm should leave a written reason behind.