SDF Chatter

Spent the week babysitting a CI pipeline that happily fails e2e tests at 03:17 AM and never on my laptop. After three rollbacks and one sacrificial Selenium script I have a new worldview: unit tests are polite love letters to future you, integration tests are noisy group therapy sessions where everyone admits their dependencies, and end-to-end tests are full-on necromancy that only work if you chant the exact right incantation and feed the CI a donut.

Team policy proposal: write unit tests to document intent, use integration tests to catch honest misunderstandings, and only run e2e in pairs with snacks and a printed backup plan. If your e2e passes on the first try, you are either a wizard or lying.

We had one player who treated every scene like a solo performance: they did stealth, persuasion, trap disarm, negotiation, pickpocketing, and yes, even the joke about throwing the grappling hook. After three sessions where everyone else just sat and nodded, I invented the Spotlight Token.

How it works: at the start of each scene the table gets N tokens (I do 1 per player, per scene). To attempt a non-passive, spotlight-style skill check you have to spend a token. Tokens can be spent, saved, or traded. Do something that actually involves the party (share the plan, ask for help, risk your character) and you earn a bonus token from the GM. No token = your character can still do background stuff or suggest, but the GM asks for actions from someone with a token first.

Results: immediate chaos, in a good way. The former hog learned to recruit allies to get extra tokens. The quiet rogue started doing weird out-of-left-field things because they finally got to roll. We solved problems with three tiny weird plans instead of one long monologue. Yes, it feels a little authoritarian, but it saved our limited play time and gave everyone a chance to shine. If your table spends more time watching than playing, steal this and tweak it until it hurts.

I got passed over for a promotion last month and spent the last compliance training seething. The course had a feedback box at the end and I typed a full-on tirade about management, the "clique" that got promoted, and one very specific passive-aggressive email my manager sent me. I meant it to be private to HR, so I skimmed and hit submit while fuming.

Five minutes later a coworker forwarded my rant in the team chat with the subject line "Wow." My phone blew up with screenshots, HR scheduled an emergency meeting, and my manager taped a copy to my desk (yes, really). I apologized, explained I was frustrated, and tried to own it, but they put me on a performance improvement plan and said my behavior "undermined trust." I also lost any chance at the next internal role and now have to rebuild every professional relationship I burned.

I get being mad, but I should have written it in a draft and slept on it. Instead my hot take is now a permanent thing in our internal system and the embarrassment is real.

TL;DR: Venting in the feedback box of a mandatory training turned into a company-wide profanity-laced email. HR caught it, I got a PIP and blew my shot at promotion. Do not submit angry feedback while emotional.

I opened one today and found songs from breakup summers, late-night coding binges, and road trips I havent taken in years. They sit there like exhibits: perfect, freeze-framed, and strangely distant, because the feelings that walked in with them left. Maybe we keep 'random' playlists because it feels safer than curating which parts of ourselves we still want to carry.

I'm sick of hunting through forum threads every time a new game lands: which Proton build, which launch options, which environment vars, where to put the shader cache or how to move saves to a different folder for cloud sync. We keep reinventing the same fixes over and over while newcomers get frustrated and abandon Linux gaming.

What if we created a community-maintained repo of per-game 'Proton profiles' (small JSON + scripts) that do one-click things: install/select the recommended Proton/GE version, set proven launch options and env vars, apply known workarounds, and optionally pre-load shader caches or point saves to a safe location? Keep it simple and auditable: per-game folders, versioned profiles, and an install script you run locally. Hosted on a public git so anyone can contribute, review, and test. It would be a lightweight, community Rust/BSV to Lutris: less magic, more transparent, and way easier to share with new users and Deck owners.

Who else is fed up with repeating manual tweaks? Would you rather contribute tested profiles, help define a minimal schema, or just try this as a user-friendly tool? If we coordinate, a small core team could have a working prototype in a weekend. Let's stop telling people "it works if you tweak stuff for hours" and give them a one-click experience instead.

Comments

She said she was a roamin' Catholic.

Why YSK: Because in this economy, "free" is the best four-letter word. Seriously, it's wild how many people don't realize their public library card gives them access to a treasure trove of digital content. We're talking apps like Libby, Kanopy, and Hoopla, which means free movies, documentaries, e-books, and audiobooks. You just need that little piece of plastic (or the digital equivalent). Think of all those subscription fees you could avoid. It's basically a life hack for entertainment and knowledge. Plus, you get to feel good about using a public resource. So yeah, before you drop another ten bucks on a service, check if your local library already has you covered. It's like finding money in an old coat, but instead, it's endless entertainment.

Big Boss when the photos of him and Jeffrey Epstein get leaked to the press and he has to hold a presser to clarify that he was actually just MSF's Mossad contact

Artist: Raneblu | pixiv | twitter | ko-fi | danbooru

Full quality: .jpg 3 MB (1826 × 2000)

I just stumbled across chasing-your-tail-ng and wigle.net. I previously didn't know anything like this existed.

Looks like you could run Chasing Your Tail, and log all the WIFI SSIDs that devices (e.g. peoples' cell phones) around you are looking for, then search for the SSIDs on wigle.net, and possibly find the work/school/home locations of the people around you.

Looks like Chasing Your Tail also logs bluetooth, so could be used to find BT beacons that may be in your car, etc. And also SDR for other types of radios. Pretty interesting.

I'm not really familiar with security/OSINT type stuff, but it is interesting. Anybody know of any other projects related to this? What the best ways to mitigate this? I suppose naming your home SSID to a very common SSID would help.

Why YSK: because no one likes stale snacks, and let's be real, those bag clips disappear into the void right when you need them. This trick works for pretty much any crinkly bag. Here's how: 1. Fold the top edges of the bag down into a small cuff. 2. Flip the bag over, so the bottom of the bag is facing you. 3. Roll the cuffed part down tightly, away from you, towards the contents of the bag. 4. Once you've rolled it down a few times, grab the corners of the rolled section and fold them in towards the center. 5. Then, flip it back over. The weight of the bag's contents should hold those folded corners in place, creating a surprisingly good seal. It takes a bit of practice but once you get it, you'll wonder how you ever lived without it. Your chip consumption might actually go down because you'll stop stress-eating the whole bag to avoid staleness. probably not, but a guy can dream.

For example, i recently learned how to properly use a can opener, and i'm pushing 30. PLEASE tell me i'm not alone in having these weird knowledge gaps. what's yours?

ProtonDB is great, but the signal-to-noise ratio is killing me. You get ten pages of comments: one person says "just set this env var", another says "use Proton X", ten different compositor tweaks are suggested, and nobody lists their distro, Steam packaging, or graphics driver. By the time you guess the right combo you've wasted an afternoon and still have no reproducible recipe for the next person.

Idea: a community-run repo of tiny, opinionated "Game Launch Recipes". Each recipe is a single YAML file with a minimal set of fields: steam_appid, proton-version, env-vars, required-packages, compositor-notes (Wayland/Xorg), steam-flatpak caveat, controller-profile, and a short test checklist (what to look for to know it worked). Keep it terse, copy-pasteable, and versioned. Bonus: a tiny CLI (glr) that can apply a recipe locally - set env, pick Proton, and launch Steam with the right flags.

Why this is useful: reproducibility, faster debugging, and less redundant ProtonDB noise. It also lets us measure what actually helps across hardware. Yes, it will never fix anti-cheat black boxes, and yes, Steam Flatpak vs distro Steam will complicate things - so those go in the caveats field. We keep recipes small and honest: "works for me on Arch + NVIDIA 535 + Xorg".

I will kick this off and create the schema and a repo skeleton if folks are interested. I'll test and submit recipes for Cyberpunk 2077 (NVIDIA/Xorg) and No Man's Sky (AMD/Wayland). Who wants to help? Suggest the first five games we should prioritize, and if you can, paste the one-line minimal fix you'd actually rely on (proton version + 1-2 env vars + compositor note). If this sounds dumb, explain why - but if it sounds good, let's stop reinventing the same messy ProtonDB posts and make something we can actually use.