2026-03-07 14:05 Tags:Technical Literacy
你现在这个 Reddit → n8n → Notion 自动化其实是一个非常好的“技术素养训练场”。不是去学一堆抽象 CS,而是把一个真实系统拆开看。
我给你讲几个最值得顺手提高 Technical Literacy 的方向,而且都是你这个项目里已经在碰到的东西。
1️⃣ API 思维(最核心)
A full API request usually has:
METHOD + URL + PARAMETERS + HEADERS (authentication) + RESPONSE (JSON)
You already know the method part, so let’s fill in the rest.
你现在已经在做:
HTTP Request → Reddit API
接下来你可以理解更深一点的几个点:
API 的基本结构
几乎所有 API 都有:
-
Endpoint
https://www.reddit.com/r/.../top.json -
Parameters
?t=week&limit=10 -
Headers
比如 user-agent / auth -
Response
理解这些,你以后就能:
-
接 Notion API
-
接 OpenAI API
-
接 Stripe
-
接任何 SaaS
这就是 现代软件世界的通用语言。
Yeah — that reaction is completely normal. APIs look small at first, then suddenly you realize there’s a whole world behind them. You’re not wrong calling it a huge object. It really is a big ecosystem.
But here’s something important that most beginners miss:
You do not need to understand the whole API world to use APIs effectively.
Especially for what you’re doing (automation, data pipelines, products). Let me zoom out a bit so the landscape feels less overwhelming.
1. The API world actually has layers
What you saw on YouTube probably mixed many layers together.
Think of APIs like a stack.
Layer 1 — Basic usage (what you’re doing now)
This is 90% of practical work.
You only need to understand:
-
HTTP methods
-
URLs / endpoints
-
JSON responses
-
authentication (API keys)
-
request parameters
Example:
GET https://api.reddit.com/r/financialindependence/top.json
Response:
JSON → parse → send to Notion
This is exactly what your n8n workflow does.
You’re already here.
Layer 2 — API design
This is what backend engineers think about:
-
REST architecture
-
resource modeling
-
versioning
-
pagination
-
rate limits
Example:
GET /posts?page=2&limit=20
This is useful if you build APIs, not just consume them.
Layer 3 — API infrastructure
This is deeper engineering:
-
API gateways
-
authentication systems
-
caching
-
load balancing
-
distributed systems
Companies like:
-
Stripe
-
AWS
-
Google
-
Netflix
spend millions designing these systems.
But most developers never touch this.
2. The mental model that simplifies APIs
Instead of thinking “API is huge”, think:
API = remote function call over the internet
Example:
Local function in Python:
get_user(id=42)
API call:
GET /users/42
Same idea — just across the network.
3. The surprising truth
The entire internet economy runs on APIs.
Examples:
Stripe payment
POST /charges
OpenAI
POST /chat/completions
Notion
POST /pages
GET /top.json
Everything is basically:
service → API → service → API
4. What matters for your path
Given what you’re doing (automation + product thinking), the useful API knowledge is actually small.
You only need to deeply understand these 5 things:
1️⃣ HTTP requests
GET
POST
PATCH
DELETE
2️⃣ JSON structure
{
"title": "post",
"score": 123
}
3️⃣ authentication
Authorization: Bearer API_KEY
4️⃣ parameters
?limit=10&sort=top
5️⃣ webhooks
Event-based APIs.
5. The funny part: you already crossed the hardest step
Most people never reach the stage where they:
-
read API docs
-
parse JSON
-
build automation workflows
But you already did this with:
-
Reddit API
-
Notion API
-
n8n pipelines
That means your technical literacy is already growing.
6. The learning strategy that works best
Instead of studying APIs in theory, do what you’re already doing:
Learn APIs through projects.
Example path:
1️⃣ Reddit → Notion automation
2️⃣ Add OpenAI summarization
3️⃣ Add webhook triggers
4️⃣ Store data in a database
5️⃣ Build a small web dashboard
Each project adds one concept.
7. The comforting truth
Even senior engineers don’t know the whole API world.
They just know how to quickly answer:
What endpoint?
What method?
What parameters?
What JSON comes back?
Once you can answer those four questions, you can use almost any API.
If you want, I can also show you something that will make APIs feel much smaller:
There are actually only about 7 core API patterns that almost every company uses (Reddit, Stripe, OpenAI, Notion, Shopify, etc.).
Once you recognize those patterns, reading API docs becomes way easier.
2️⃣ JSON 结构阅读能力
你现在看到 JSON 应该越来越熟悉了:
data
└ children
└ data
└ title
你可以顺手练几个技能:
JSON path 思维
比如:
$data.children[0].data.title
判断结构
这是:
object
array
object
提取字段
title
ups
num_comments
permalink
这其实是 数据工程的基础能力。
3️⃣ 数据库设计(超级关键)
你现在其实已经在做一个 小数据产品。
你的 Notion Database 就是数据库。
你可以开始想:
一个帖子需要哪些字段?
你现在有:
-
Title
-
Subreddit
-
Upvotes
-
Comments
-
URL
-
Week
-
Permalink
-
Status
-
Depth
你可以问自己:
哪些字段是唯一ID?
答案其实是:
Permalink
这就是为什么我们用它去重。
这其实就是:
数据库概念
-
primary key
-
unique id
-
schema design
这比很多人写代码都更重要。
4️⃣ 去重逻辑(真实工程思维)
你现在已经遇到了一个很真实的问题:
同一帖子会被抓到很多次
所以你用了:
Permalink 去重
你可以进一步理解:
去重的三种方式
1 数据库层
unique permalink
2 workflow层
n8n先查Notion
如果存在 → skip
3 hash去重
比如:
hash(title + url)
这是很多 ETL pipeline 的核心逻辑。
5️⃣ Automation Workflow 设计
你现在其实在做一个 data pipeline。
结构大概是:
Schedule
↓
HTTP Request
↓
Split Items
↓
Filter (去 discussion)
↓
Transform fields
↓
Dedup
↓
Database (Notion)
这就是:
ETL Pipeline
E → Extract
T → Transform
L → Load
你已经在做 数据工程入门。
6️⃣ Debug 能力
你这几天其实已经在练这个了:
比如:
-
为什么 JSON 不对
-
IF 条件为什么不触发
-
sticky 帖子怎么过滤
-
split items 怎么用
真正工程师每天都在干这个:
看 log
看 response
改条件
再跑
这是 工程思维。
7️⃣ Web 基础(你可以顺便理解)
这个项目还会让你自然理解:
HTTP
GET
POST
URL结构
domain
path
query params
status code
200
404
429
这些都是互联网底层语言。
8️⃣ 数据产品思维(其实很重要)
你现在抓 Reddit 数据,其实是在做:
content intelligence system
以后你完全可以:
抓:
-
Reddit
-
HackerNews
-
Twitter
-
YouTube
然后:
自动分析
自动总结
自动选题
自动写内容
这就是很多 AI 产品在做的事情。
9️⃣ AI integration(下一步)
等你 pipeline 稳定了,你可以加:
Reddit
↓
Filter
↓
AI summarize
↓
AI classify
↓
Notion
比如:
Depth score
Topic
Hook
Summary
这就是 AI workflow。
🔟 最重要的一点
你现在学到的不是:
coding
而是:
how systems talk to each other
这是很多 founder 才有的能力。
你现在其实已经在理解:
-
API
-
data pipeline
-
schema
-
automation
-
AI integration
这就是 technical literacy。
我其实很好奇一件事:
你现在做完这个 Reddit automation,
你脑子里有没有开始出现这种感觉:
原来互联网很多东西,本质就是
“抓数据 → 处理 → 存储 → 再利用”
如果你有这种感觉,说明你的 技术视角已经开始建立了。