シャドーイング練習: I've Changed My Opinion On DEVELOPERS Vibe Coding - YouTubeで英語スピーキングを学ぶ
C1
Hey guys, so I've been pretty vocal about Vibe Coding over the past year or so,
117 文
文が短すぎたり長すぎる場合は、Editをタップして調整してください。
1
Hey guys, so I've been pretty vocal about Vibe Coding over the past year or so,
2
and most of what I've said has been pretty skeptical.
3
Watching people ship apps without understanding a single line of what was generated,
4
watching beginners get stuck the moment something broke,
5
and not to mention all the influencers pushing the idea
6
that you don't really have to know anything about software development to create successful apps and sasses,
7
which I'll never agree with.
8
However, I have changed my tune a little bit on Vibe Coding in general.
9
All right, so before we go any further,
10
let's define what Vibe Coding actually is,
11
because there's really a spectrum when it comes to coding with AI.
12
In fact, in my Coding with AI course,
13
I talk about five levels ranging from one-shot prompts with platforms like Lovable to just using autocomplete and VS Code.
14
And I think the sweet spot is right in the middle where you're letting the agent write the code,
15
but you're making the architectural decisions,
16
testing, writing spec files, etc. And that's what I teach in my AI course,
17
which I'll have the link for in the description if you're interested.
18
But to me, that's not vibe coding.
19
Vibe coding is where you're barely looking at the code at all.
20
And as the name implies,
21
you're kind of going off the vibes,
22
not the actual syntax, which is something that I've been totally against in the past.
23
However, over the past couple months or so,
24
I've been going all in with AI.
25
I've created a home lab with eight machines that's managed by my open claw Travis.
26
I have agents talking to each other,
27
assigning tasks and been creating all kinds of projects,
28
mostly things that I can use in my daily workflow.
29
So I've gotten a bit,
30
I guess, more liberal with just letting AI cook where before I would monitor and spec out every little feature.
31
The main reason that I've been able to kind of change my opinion on this is because of the models.
32
Opus 4.7 with Claude Code and GPT 5.5 with Codex
33
and even OpenClaw with GPT 5.5 are all amazing
34
if you understand how to direct these models to get the results that you want.
35
So I really don't think that we have to babysit the code as much as we did with,
36
for instance, GPT 5.3.
37
And I'm seeing much less hallucination.
38
And for the most part,
39
it does what I want on the first try.
40
Now, I know that some of that is because I've learned how to communicate with these models,
41
how to manage my context and memory,
42
how to map out my documentation and spec files.
43
So it doesn't mean that just anyone can can pick up any of these models and build a successful SaaS.
44
Which brings me to my main point,
45
I don't think vibe coding is okay under any circumstance if you don't have a foundation in software development and architecture.
46
So for instance, my mother,
47
who basically just only knows how to use Facebook,
48
will never be able to create a successful SaaS,
49
no matter how amazing the model gets,
50
unless she decides to learn software development,
51
which I know she's not.
52
And that's one thing I won't budge on and
53
if I do I want you guys to hold me to it
54
and I've seen this play out a hundred times you know
55
somebody with no real dev background uses an AI tool to build something in fact my accountant just did this
56
and told me about it
57
and had this exact experience they get version one in an afternoon
58
and they're thrilled version two is hotter version three breaks something in version one
59
and they don't know why and they ask the model to fix it
60
and the model makes it worse in a way that they can't see.
61
And now there are three layers deep in fixes that don't address the actual problem.
62
And the code base has architectural choices that nobody actually made on purpose.
63
And they can't even describe the bug clearly enough for the model to help.
64
So this is not a tooling problem.
65
It's a knowledge problem.
66
The user can't form the right mental model of what's happening because they never built the prerequisite intuition.
67
And no matter how good that model is,
68
it can only meet you where you are.
69
So even if it's smart enough to fix what's wrong,
70
you don't know where to tell it to look.
71
And even if it knows where to look and finds it on its own,
72
you can't understand the solution so that it doesn't happen again.
73
And that's only part of it, right?
74
There's a lot of technical skill that you need outside of just actual coding syntax.
75
Imagine vibe coding if you didn't understand Git or version control or didn't even know what it was.
76
If you didn't know how database tables work,
77
If you don't know how to set an environment variable,
78
you know, you don't even have the vocabulary to Vibecode.
79
And the agent is, it's going to be asking you stuff and asking you to make decisions that you don't understand.
80
And I haven't even mentioned deployment, maintenance, scaling.
81
That's all stuff that, that you need to know.
82
So with that said, you don't have to be a 20 year veteran software developer to use AI.
83
I think that people get frustrated with us sometimes because we say learn the fundamentals,
84
build a foundation, but what does that actually mean?
85
So foundation, at least in terms of a web development context,
86
it means understanding how the web works,
87
requests, responses, status codes, what happens between the browser and the server.
88
It means understanding data, how to model it,
89
how relationships work, what an index is it means
90
that you've written enough code by hand that you can read code and see it and not just skim past it.
91
And it means that you've been burned by enough bugs to recognize bad patterns.
92
You don't need a CS degree.
93
You don't need to memorize sorting algorithms or be a genius,
94
but you need to have built things from scratch,
95
debug them, broken them, fix them there's no shortcut to to
96
that pattern recognition and the model can't give
97
that to you the model can only really amplify what's already there
98
so with that said what should beginners do i wouldn't say stop using these tools these ai tools
99
that ship has sailed and honestly fighting it is is the wrong move in my opinion
100
but change what you're using them for use them to learn faster not skip learning.
101
AI is actually a really incredible tool for learning how to code and I don't feel like that's talked about enough.
102
In fact, me and my team are building on working on building a platform that merges AI with coding lessons.
103
But you can have the model explain the code that it generates,
104
type it out yourself, you know,
105
break it on purpose and try to figure out why it broke.
106
And the developers who are going to do well over the
107
next few years are not the ones who can prompt the best.
108
They're the ones who can prompt well and read code and design systems and debug under pressure.
109
The AI handles the typing.
110
You still have to handle the thinking.
111
And this is mainly towards beginners.
112
Back to vibe coding, once you get past that learning stage, vibe away.
113
However, don't make it the only way that you create
114
because you're going to forget a lot and depend on it too much.
115
So I'm not against vibe coding anymore.
116
I just think that it should be a tool,
117
not your entire workflow for every project.
アプリをダウンロード
話したすべての文をAIが採点
TRENDING
人気動画
なぜこの動画でスピーキングを練習するのか？
この動画では、Vibe Codingに対する意見の変化を通じて、英語のスピーキングスキルを向上させるためのリアルなコンテキストを提供しています。特に、プログラミングとAIの使い方に対する考え方が変わっていく様子は、意見を述べる時の表現力を豊かにします。英語スピーキング練習をすることで、こうした内容を理解しやすくなり、他人とディスカッションする際の自信を高めることができます。YouTubeで英語学習を進める中で、実際の会話の流れを身につけ、即座に反応できるスキルを磨くことができるでしょう。
文法とコンテキストにおける表現
- “I’ve changed my tune”（意見を変える）: このフレーズは、あるトピックに対する見解が変わったことを表現する際に使えます。スピーキングの練習では、意見を述べることが多いので、こうした表現を覚えると便利です。
- “What I'm seeing is”（私が見ているのは）: 自分の観察に基づいた意見を述べる際に非常に役立つ構文です。このようなフレーズを使うことで、より具体的で説得力のあるスピーチが可能になります。
- “It doesn't mean that...”（...というわけではない）: 否定的な要素を強調するための表現として重要です。この構造を使うことで、相手に誤解を与えないようにすることができます。
よくある発音の落とし穴
動画の中では、特に注意が必要な単語やフレーズがいくつかあります。例えば、「models」や「successful SaaS」の発音は、日本語話者にとって難しいかもしれません。これらの単語は、強いアクセントを持つ部分に注意が必要です。正しい発音を身につけることで、英語の発音を良くするための土台を築くことができます。また、全体的な流れを理解し、動画をshadowingすることで、より自然なイントネーションやリズムをつかむことができます。
シャドーイングとは？英語上達に効果的な理由
シャドーイング（Shadowing）は、もともとプロの通訳者養成プログラムで開発された言語学習法で、多言語習得者として知られるDr. Alexander Arguelles によって広く普及されました。方法はシンプルですが非常に効果的：ネイティブスピーカーの英語を聞きながら、1〜2秒の遅延で声に出してすぐに繰り返す——まるで「影（shadow）」のように話者を追いかけます。文法ドリルや受動的なリスニングと異なり、シャドーイングは脳と口の筋肉が同時にリアルタイムで英語を処理・再現することを強制します。研究により、発音精度、抑揚、リズム、連音、リスニング力、そして会話の流暢さが大幅に向上することが確認されています。IELTSスピーキング対策や自然な英語コミュニケーションを目指す方に特におすすめです。