Deep dives into tools,
techniques & the web.
Practical guides covering developer tools, health metrics, SEO, converters and more — written to be actually useful.
✦ AI-powered — understands natural language
Your List Still Has Duplicates After Removing Duplicates
Geo-Redirects vs Hreflang: How Automatic IP-Based Redirects Can Silently Break Your International SEO
A search engine crawler trying to access your French-language URL might get auto-redirected to your English version, based on the crawler's own IP location — meaning the URL your hreflang tags point to is never actually reachable by crawlers. Here's why IP-based geo-redirects and hreflang conflict, and the suggestion-banner and crawler-exemption approaches that resolve it.
String Repetition and ANSI Escape Codes: How Progress Bars, Borders, and Terminal UIs Are Built
A terminal progress bar, a box-drawn menu, and colored CLI output all rely on the same underlying operation: repeating a character a calculated number of times, often combined with ANSI escape codes for color and cursor control. Here's how string repetition builds progress bars and borders, why Unicode box-drawing characters connect seamlessly at corners, and how terminal width affects repeated-character output.
Percentage Points vs Percent: Why "Rates Rose by 5%" Could Mean 4%→4.2% or 4%→9%
"Interest rates rose by 5%" and "interest rates rose by 5 percentage points" describe a rate going from 4% to either 4.2% or 9% — completely different outcomes from the same starting point. Here's why "basis points" exists specifically to eliminate this ambiguity for interest rates, why the confusion is rampant in tax-rate and polling reporting, and why percentage points can't "compound" the way percentages do.
BIMI Display Adoption: Why Your Perfectly-Configured Logo Might Still Not Show Up
A perfectly-configured BIMI setup — valid SPF/DKIM/DMARC, correctly-formatted SVG, even a VMC certificate — can still show no logo in most inboxes, because each mailbox provider independently decides whether and when to display BIMI logos. Here's the provider-by-provider adoption patchwork, why SVG Tiny PS (BIMI's restricted SVG profile) commonly rejects design-tool exports, and why "specification-compliant" and "logo actually displays" are genuinely different claims.
Muscle, Metabolism, and Aging: Why Sarcopenia — Not "Slowing Down" — Explains Metabolic Change
A landmark study tracking 6,400 people found metabolism is remarkably stable from age 20 to 60 — what actually changes is muscle mass. Here's the Pontzer 2021 findings, how sarcopenia drives the "skinny fat" phenomenon, why protein needs increase with age, and why resistance training is the primary evidence-based intervention.
Why WHOIS Now Says "REDACTED FOR PRIVACY": GDPR, Default Redaction, and the Shift to RDAP
A WHOIS lookup that once showed a domain owner's name, address, and phone number now typically returns "REDACTED FOR PRIVACY" — the result of GDPR-driven policy changes that affected WHOIS globally, not just for EU registrants. Here's why this happened, what information remains visible after redaction, how RDAP is replacing the decades-old WHOIS protocol with structured JSON output, and whether paid "WHOIS privacy" services still add value.
UUID v4 vs UUID v7 vs ULID vs NanoID: Which Identifier Format Should You Use?
UUID v4's random bits fragment database B-tree indexes, causing write amplification. UUID v7 adds a millisecond timestamp prefix to fix this. ULID is sortable and URL-safe without hyphens. NanoID is compact and customisable. Here's how each works and when to choose each format.
Pagination, Infinite Scroll, or "Read More": Three Answers to "This Content Is Too Long" and When Each Fits
"Read more" links, pagination, and infinite scroll are three different answers to "content is longer than the space available" — each with different SEO, accessibility, and UX trade-offs. Here's where each fits best, why infinite scroll's footer-reachability problem is well-documented, and how hybrid approaches (load-more buttons, infinite scroll with History-API pagination underneath) combine their benefits.
Bcrypt's 72-Byte Limit: Why Two Different Passwords Can Hash to the Same Value
Bcrypt silently truncates input at 72 bytes — meaning two passwords sharing the same first 72 bytes but differing afterward produce the identical hash. Here's why 72 specifically, how multi-byte Unicode characters (emoji especially) reach this limit far sooner than "72 characters" suggests, why pre-hashing with SHA-256 is a common mitigation, and why this isn't actually a meaningful security concern for typical passwords.
Base64 vs Base64url: Why `+`, `/`, and `=` Break URLs, and Why JWTs Use a Different Alphabet
Standard Base64's `+`, `/`, and `=` characters all have special meaning in URLs — which is why a Base64-encoded JWT pasted directly into a URL can silently corrupt, and why Base64url exists as a separate, URL-safe alphabet. Here's exactly which characters differ, why JWTs require Base64url specifically, how padding works and why Base64url commonly omits it, and how to tell the two variants apart.
Ketogenic Diets and Nutritional Ketosis: What the Research Shows About Weight Loss, Performance, and Who Should Be Cautious
Ketogenic diets shift the body's fuel source from glucose to fat and ketones — a measurable metabolic state with the strongest evidence base in drug-resistant epilepsy. Here's how nutritional ketosis differs from dangerous diabetic ketoacidosis, what weight loss research actually shows over longer timeframes, and the mixed evidence on athletic performance.
Harris-Benedict vs Mifflin-St Jeor vs Katch-McArdle: Why BMR Formulas Disagree by 100-200 Calories
BMR calculators don't all use the same formula — Harris-Benedict (1919, revised 1984), Mifflin-St Jeor (1990), and Katch-McArdle (lean-mass-based) can disagree by 100-200 calories for the same person, which represents 20-30% of a typical weight-loss deficit. Here's how each formula was developed, why they diverge more for people at the extremes of body composition, and why any formula's output should be treated as a starting point rather than a precise target.
