mitchm
This has always been the worst part about the web, now amplified.
Simple, declarative RUM and E2E tests for web apps
Show HN: Uno – Us vs. Them. Humans vs. AI version of the card game
I love Uno, so for fun I made a web based multiplayer version of the game. Completely free, no ads, no accounts. Has some of my favorite custom “house” rules like stacking, shuffle hands, and numbers only!Play friends or play AI. Chat with opponents. You can even bring your own AI
Show HN: Turn E2E tests into observability signals
Show HN: A real-time multiplayer card game built with Go and Htmx
This started out as an exercise in what’s possible building web apps without JS UI libraries.It’s a real-time multiplayer version of the classic card game Uno. Heavy use of the SSE extension for HTMX, which powers realtime game updates, chat, and the live replay feature for spectators. Lots of granular OOB swaps.Figured it would be a better demonstration than a todo list. Happy to answer any...
Using an LLM friendly api with a snapshot of game state and calculated heuristics, legal moves, and varying levels of strategy in working out nicely. They can play a web based game via curl.
Production for the frontend is the user’s device, and we all know things can go wrong without an explicit error. With the current state of frontend observability tools, how do you tell when something should have happened but didn’t?
Too bad we can’t just have our users run `it('opens checkout modal on click', …)`, right? What if we could..
Maybe the world doesn’t need more HTML attributes, but it’s my best attempt so far at improving frontend observability. I think the ends justify the means.
What if your users ran your E2E tests?
Show HN: Fault Sense – An Application Feature Monitor for Web Apps
If you’ve ever been asked, “Conversions for XYZ are down. Did something break on the site?” this might resonate.Fault Sense is the library I wish existed to answer the kinds of questions product analytics and tracing tools can’t: when something should happen, but doesn’t. Tracing tools get close, but they’re heavy and invasive to instrument.Fault Sense stays lightweight. It’s language and...
YouTube is an interesting one. Like the Internet itself, there is some really great stuff, but letting them roam is a recipe for disaster. YouTube Kids has gotten better over the years but nothing's perfect. On one hand, I didn't want to block YT completely but I also wanted to be able to approve anything they watch. So, I did what any other programmer dad would do in this situation and built what I wanted.
They now have a permission layer on top of Safari (as an extension that cannot be disabled when coupled with Safari's content moderation setting). I set Safari to only allow YouTube and any channel they try and view that I have not approved will require them to send a permission request to my phone. That "permission request" is a button + them coming to me with "Hey Dad, I sent you a request". It's worked quite well in our home and has opened a nice dialog about what they are trying to watch and why. I made it public if any other parents are interested -- https://www.sunscreenapp.com/
YC NYC Challenge: Hack your way to the top
USvsTHEM (http://usvsthem.com) is a site that I co-created - and still very early alpha; but adds a twist that centers a debate around people and their network of support. There is no browse-able stream yet, as the alpha release was meant to test the social utility.
Good Luck!
At first i was cautious since escrow.com looks like it straight out of 1996 but it's legit.