khlilo

你无法坚持一件事,是因为太喜欢“学习”

2026-04-16

能成功的人总是少数,而绝大部分普通的人都认为自己拥有远超自身成就的认识。

在我的视角里,我很少听到失败,我听到更多的,是压根就没有开始。

其因为何?我们太多人,都倒在了准备阶段,耗尽了热情与兴趣。百度网盘里放着几百G的视频,收藏夹里也躺满了各行各业的起步教程,而实际上没有一个能验证的、拿得出来的demo或者成果。

不要把手段当目的。 大部分人无法坚持一件事,不是因为不够努力,而是因为太喜欢“学习”。这种做法不仅无效,甚至是反效的。

此学习非彼学习。这里的“学习”不是认知升级和自我提升,而是一种心理上的“前期准备”,是心理防御机制的产物。面对你心中自认为困难的事,人本能的会想要退缩。为了让这种退缩显得心安理得,你选择用无休止的‘学习’来填补时间。你企图用完美的前提准备来消灭一切试错的风险,但这绝无可能。无限学习带来的唯一作用就是让你止步不前。

要做什么最好最快的方式就是直接做,而大多数人多使用了相反的做法——在做一件事之前,总是想着先去“学”。比如想要写文章,先要去学习一堆“如何写好文章”的课程;想要使用ai,先想着要去看一堆如何使用好ai的视频;想要利用大语言模型构建自动化工作流,却天天在看别人写的“神级 Prompt 汇总”,自己连一个基础的 API 接口都没调通过。

AI泛滥,信息爆炸,在这个时代10小时内能入门任何一个行业,了解到基础的运作机制。但大多数人将这个时间扩展到了无限大,实际上,这种学习存在边际递减效应,即超出10小时后,脱离实际操作的纯理论学习,其收益几乎为零(单纯深耕某一领域的理论知识除外)。于是所有人那些想做而未完成的事都停在了学这一步。

当你真正热爱并了解了做事的底层原理后,你不再会有这些思考,你的想法只有做事本身。

最后我想要提醒的是,不要把学习当作先决条件,要建立“即时调用”的思维。缺什么补什么,遇到具体的错误再去查具体的文档,把学习变成解决问题的插件,而不是前提。

任何没有产出的学习本质上都是心理安慰,都是在消耗时间。掌握“项目导向”式的学习逻辑,直接上手做,哪怕很糙,你也是在构建一个可以持续迭代的资产。

Only a few people actually succeed, yet most average folks walk around believing their insights far exceed their actual accomplishments.

From where I stand, I rarely hear stories of failure. What I hear, overwhelmingly, are stories of people who never even started.

Why does this happen? Too many of us die in the preparation phase, draining our enthusiasm before the real work begins. We hoard hundreds of gigabytes of tutorial videos on our hard drives, our bookmarks are stuffed with "getting started" guides for every conceivable field, but in reality, we don't have a single working demo or tangible result to show for it.

Stop mistaking the means for the end. The reason most people can't stick with anything isn't a lack of effort; it's because they are too addicted to "learning." This approach isn't just ineffective—it's actively counterproductive.

But let's be clear about what kind of learning this is. The "learning" I'm talking about isn't about cognitive upgrades or genuine self-improvement. It's merely psychological "prep work," a byproduct of your own defense mechanisms. When faced with something you perceive as difficult, your instinct is to retreat. To make this retreat feel justified and guilt-free, you fill your time with endless "studying." You attempt to use perfect upfront preparation to eliminate the risk of trial and error, but that's a pure illusion. The only thing infinite learning achieves is keeping you stuck exactly where you are.

The best and fastest way to do anything is simply to start doing it. Yet, most people do the exact opposite—they always think about "learning" before doing. Want to write an article? First, they take a bunch of courses on "how to write well." Want to use AI? They watch hours of tutorials on "how to master AI" first. Want to build an automated workflow with LLMs? They'll spend every day reading "God-tier Prompt Collections" without having ever successfully called a basic API endpoint themselves.

We live in an era of abundant AI and exploding information. Today, you can get the gist of any industry and understand its basic mechanisms within 10 hours. However, most people stretch these 10 hours into infinity. The reality is that this kind of learning suffers from severely diminishing returns. Beyond those initial 10 hours, pure theoretical study divorced from actual hands-on practice yields near-zero value (unless you are deeply researching the theory of a specific niche). As a result, all those unfulfilled ambitions die right at the "studying" step.

Once you find genuine passion and grasp the underlying principles of how things actually get built, you stop having these hesitations. Your only thought becomes the work itself.

My final piece of advice: stop treating learning as a prerequisite. Build a "Just-in-Time" mindset. Fill the gaps as you find them. Look up specific documentation only when you encounter specific errors. Turn learning into a plug-in for solving problems, not a precondition for starting.

Any learning that doesn't produce an output is, at its core, just a psychological pacifier—a sheer waste of time. Master a "project-oriented" learning logic. Get your hands dirty immediately. Even if your first draft or initial build is incredibly rough, you are already building an asset that can be continuously iterated upon.