Software reviews, code tips, and net culture.
I spent three weekends scanning and re-typing my grandfather's paper workshop manuals into a local-only knowledge base, testing Obsidian, Logseq, and plain Markdown along the way. The software mostly held up. The real lesson was about what documentation is actually for.
I pulled every cord, dongle, and power brick out of my house, tested them one by one, and tallied up the damage. The result: a small fortune in forgotten accessories and a sobering look at how much e-waste one American household quietly generates.
Most developers spend more time picking a font than picking a license. I gave one hobby project three different open source licenses, then tracked what happened when strangers, bots, and one very confident startup got their hands on it.
I capped my home internet at 56 kilobits per second for seven days to see what breaks first — and the answer says a lot about what modern software actually does with your bandwidth. Spoiler: it's mostly not the things you chose to load.
My grandparents' Windows machine was a ransomware scare waiting to happen, so I built them a Raspberry Pi desktop instead — auto-updates, a locked-down browser, remote admin, and a laminated cheat sheet. Six hundred miles later, I'm convinced this should be a product.
I built a real small business website using strictly 1996 design rules — table layouts, a visitor counter, zero JavaScript — and ran it head-to-head against a modernized twin for 30 days. The bounce rate data made me question everything I believe about 'good' web design.
Clicking 'I Agree' on your router's end user license agreement means accepting things most people would never guess. I spent a weekend with eleven consumer router EULAs, a highlighter, and way too much coffee. Some of what I found was boring. Some of it genuinely should worry you — and possibly your ISP.
I spent three months running my own mail server on a $6 VPS because apparently I hate myself. Here's what broke, what got silently blacklisted, and why I'd still tell you to try it.
Deleting an app should take one click. Instead, most of us spend a full workday hunting through settings menus like we're defusing a bomb. I talked to five developers — off the record — about who's actually responsible for the maze, and none of them said what I expected.
I dusted off the crude, keyword-honest ranking logic of a 1998 search engine and ran today's SEO-optimized marketing pages and AI-generated blogs through it. The results suggest the old link graph was a lot harder to game than the engagement metrics we've got now.
For one month, every software download and subscription decision went to the crowd: forum threads, Discord recs, and YouTube comment sections. Some picks were revelations. Others were clearly just the same five apps being recommended to each other in a loop.
I tracked 25 broken things across a month — printer drivers, router firmware, a Windows update that ate my Bluetooth — and asked two questions for each: one to official support, one to Reddit. The strangers won, and not by a little.
What happens when a regular person with a router from 2019 decides to run their home network like a small business IT shop? VLANs, a firewall rule spreadsheet, and one smart fridge that refused to cooperate. Here's what actually improved security — and what was pure theater.
I spent a month cross-referencing the update notes of 25 popular apps against what actually changed under the hood. Some developers write changelogs like honest humans. Most write them like lawyers hoping you won't read the contract.
I kept exactly 50 browser tabs open for a full month without closing a single one, then audited which ones I actually touched again. The results say less about my productivity and more about how the modern web weaponizes 'maybe later.'
I signed up for 15 free trials with zero intention of canceling them, then tracked every sneaky charge, guilt-trip email, and cancellation maze for two months. The total damage: $312.73 — and a masterclass in how subscription businesses actually make their money.
I dug up a bookmarks folder older than some of my coworkers and started clicking. Most of it was gone — but what survived, and how, says a lot about who actually owns the web. Spoiler: it's mostly Google, and it's mostly not you.
We dug through the privacy policies and data flows of a dozen popular personal finance apps to find out how they actually pay the bills. The answer, for half of them: your transaction history is the product.
I spent two weeks digging through privacy policies, corporate filings, and ownership trees for eight of the most-downloaded free VPNs in the US App Store. What I found: most of them aren't privacy tools at all — they're data collection businesses wearing a privacy tool costume. Two of them, though, actually hold up.
I spent twelve months spying on the things spying on me — a $300 vacuum, a smart plug, and a video doorbell — using nothing but a Raspberry Pi and free network tools. The destinations will not surprise you, but the volume might.