Silas Lee | Tidemason

I spent the last few months building too much in private. I built apps, prototypes, landing pages, domains, and tools. Some shipped. Most never reached real users. So I’m doing a reset.

今天做胃肠镜检查，我躺在手术床上，护士给我打了麻药。 我问：“我的手有点痛，这是正常的吗？” 然后护士拍打我的肩膀让我醒醒。 感觉瞬间就做完了，但我还没听到答案。😂

慢慢的，已经积攒了20多个屏蔽词.

Product Hunt 像一个学习宝库

My new rule: No more building in private for weeks before talking to users. I’ll share what I build, what fails, what gets archived, and what real users actually say. If you’ve also overbuilt before distribution, I’d love to compare notes.

So I cleaned up the domains and kept one public home: tidemason.com/?ref=x-reset Tidemason is now my public lab for small apps, archived experiments, and honest build-in-public notes.

I spent the last few months building too much in private. I built apps, prototypes, landing pages, domains, and tools. Some shipped. Most never reached real users. So I’m doing a reset.

把域名也清理了一下，只留一个主域名，其他的重定向到主域名，清爽了许多。

今天一天都在清理项目、归档。 以后不再闷头干了，从用户出发。 到现在为止已清理3个项目，几十万行代码归零。

I realized I kept building products before distribution. So I’m doing a reset: - Archived old experiments - Keeping only tidemason.com - Choosing one app to validate publicly - No new features for 14 days Goal: Learn if strangers care before I build more.

Kept 1.5 is now live. This update pushes Kept closer to what I wanted it to be: not a subscription tracker, but a calm place to make renewal decisions before recurring services charge again. Decide before it renews. apps.apple.com/us/app/kept-su… #buildinpublic #indiedev #iosdev

@tinyfool +1

来让我看看关注我的人里面谁是蓝v，请蓝v关注者留言，谢谢

I agree! But I think this phase is gonna pass pretty quickly. It's like the early days of cars—everyone was just tinkering around, figuring out how to build and drive them. Then once the standards came in (steering wheel, brakes, driver's license rules), the bar dropped big time. Regular folks didn't need to understand the mechanics anymore; they could just hop in and drive.I 'm betting programming will get standardized first, 'cause developers are the ones willing to pay right now. Their pain points are super clear—coding, debugging, testing—and feedback loops are fast. Once protocols like MCP settle into dev tools, using agents to code will feel as natural as using a regular IDE. But enterprise automation is where the real money is. Short term, the standards from programming will spill over and help enterprises. A lot of companies start by boosting their dev teams first, then expand to core business stuff. Long term, the bar for users will drop a lot—leaders just need to set the big goals and boundaries, and the system handles the rest. This shift is probably gonna happen faster than we think.

It is notable that we are all debating exactly which markdown files are most important to feed AI (skills, memory, tool instructions) and in which order to feed them to get the best output. Feels that this is likely a temporary state of affairs in the development of agents

Try Kept Cost here: apps.apple.com/us/app/kept-co… Website: tidemason.com What subscriptions are you struggling with? Or what's your biggest challenge as a solo dev? Reply below — happy to chat in English or 中文! 🚀 #IndieDev #BuildingInPublic #KeptCost

Why follow me? - Authentic indie dev logs - Practical lessons from shipping small AI products - Building in public transparency - Occasional Chinese/English posts Next small product is already in the works...