Pratique du Shadowing: The Most Expensive MISTAKES Ever - Apprendre l'anglais à l'oral avec la vidéo
Création de la leçon...
1
SNCF.
2
You've probably made an expensive mistake before.
3
Whatever you're picturing right now, it's nothing.
4
These next 10 mistakes cost billions of dollars, destroyed entire companies, and in one case, poisoned a continent for decades.
5
Almost none of them started as a dramatic failure.
6
They started with something small nobody bothered to check.
7
Take France.
8
It ordered hundreds of new trains.
9
Bigger, faster, better.
10
One problem, nobody checked if they'd actually fit.
11
SNCF, the French national rail operator, commissioned a massive fleet upgrade in the 2010s.
12
The new trains were built slightly wider than the old ones.
13
Sounds minor, it wasn't.
14
Turns out, thousands of platforms across older regional stations were built to narrower dimensions, some dating back over a century.
15
Nobody in the procurement chain thought to actually go measure them before finalizing a design that assumed every platform was the same.
16
The mistake wasn't discovered in a meeting.
17
It was discovered when the trains showed up and didn't fit.
18
SNCF had to spend roughly 50 million euros shaving down platforms across the network station by station.
19
Here's the part that stings.
20
The people who designed the trains and the people who managed the stations worked for the exact same organization.
21
They just never talked to each other.
22
an entire fleet redesigned around a number nobody bothered to verify because everyone assumed someone else already had.
23
Found out the same way you'd find your new couch doesn't fit through the front door.
24
After delivery, not before.
25
Mars.
26
NASA lost a $328 million spacecraft over a unit conversion, the kind of mistake that gets marked wrong on a middle school worksheet.
27
The Mars Climate Orbiter was supposed to study the Martian atmosphere and act as a communications relay.
28
One engineering team, working on the ground-based navigation software, used imperial units, pound-force seconds.
29
Another team, building the spacecraft's thruster software, used metric, newton seconds.
30
Nobody caught the mismatch before launch.
31
The navigation software took in the imperial data and processed it as if it were metric, without converting it.
32
Every single trajectory correction, every thruster firing calculated over the nine-month journey to Mars was off by a factor of roughly 4.45.
33
By the time the orbiter reached Mars, it approached at completely the wrong altitude.
34
Instead of settling into a stable orbit, it either burned up in the atmosphere or shot straight back out into deep space.
35
NASA still isn't entirely certain which one happened.
36
Nine months of travel, hundreds of engineers, hundreds of millions of dollars,
37
Gone, because two teams working on the same spacecraft never agreed on which measurement system to use.
38
And nobody thought to double-check.
39
The investigation afterward didn't blame some exotic technical failure.
40
It blamed a missing verification step that should have taken one person about 10 minutes.
41
NASA switched almost entirely to metric on future missions.
42
A little late for this one.
43
Knight Capital.
44
Knight Capital lost $440 million in 45 minutes.
45
That works out to roughly $163,000 a second.
46
The trading firm was rolling out new software across a set of eight servers before markets opened.
47
One engineer forgot to update one of those servers with the new code.
48
That single server still had old, defunct testing code sitting dormant on it, left over from years earlier.
49
When the markets opened, that dead code woke up.
50
It started firing off millions of unintended trades automatically, buying high and selling low, again and again and again,
51
at a speed no human trader could ever hope to catch or stop.
52
By the time anyone at Knight Capital actually realized what was happening and manually pulled the plug, the damage was already done.
53
The company had accidentally built a machine whose entire job, purely by accident, was losing money as fast as physically possible.
54
And it did that job perfectly.
55
The firm that had been trading successfully for decades was essentially insolvent by lunchtime.
56
It was bought out within the year for a fraction of its former value.
57
One forgotten deployment step.
58
On one server. $440 million gone.
59
Nobody hacked Knight Capital.
60
No market crash caused it.
61
Their own software running exactly as it was coded to run just happened to be years-old code nobody remembered existed.
62
That's the part that should scare you.
63
Iran-5.
64
The European Space Agency destroyed its own rocket 37 seconds after launch with code that used to work perfectly fine.
65
Arian 5's inertial guidance software was reused from its predecessor, Arian 4, specifically to save development time and money.
66
That decision made sense on paper.
67
The software had flown successfully for years without issue.
68
Except Arian 5 was significantly faster and more powerful than Arian 4 ever was.
69
Its horizontal velocity produced a number too large for the old software's data format to handle.
70
The system tried to convert that number, overflowed, and crashed almost instantly.
71
Both the primary and the backup guidance computers were running identical software, so they failed at the exact same moment, for the exact same reason.
72
There was no meaningful redundancy, because the so-called backup had the same fatal bug baked in from the start.
73
The rocket veered violently off its intended course and was destroyed by its own self-destruct system,
74
taking roughly $370 million worth of hardware and years of development with it.
75
The code itself wasn't broken.
76
It just was never built for a rocket this powerful, and nobody bothered to retest that assumption.
77
The exact line of code responsible wasn't even needed after liftoff.
78
It was leftover functionality from an earlier flight phase that engineers forgot to disable.
79
A rocket destroyed by software doing a job nobody asked it to do anymore.
80
Denver Airport.
81
Denver International Airport tried to build the most advanced automated baggage system on the planet.
82
It delayed the entire airport's opening by 16 months and cost around $560 million.
83
The plan was fully automated luggage handling from end to end,
84
computer-controlled carts zipping suitcases across the airport at high speed with zero human involvement anywhere in the process.
85
Ambitious?
86
Also wildly premature for the technology that actually existed at the time.
87
The system jammed constantly, sometimes for hours.
88
Carts collided with each other on the tracks.
89
Bags got physically shredded and mangled by the very machinery meant to sort them gently.
90
During testing, the failure rate was so high the system was essentially unusable in practice.
91
Airlines refused to trust their passengers' luggage to a machine that ate suitcases for breakfast.
92
So the airport ended up building an old-fashioned, boring, manual backup system anyway.
93
right on top of the money already spent chasing the futuristic version.
94
The automated system limped along in a reduced role for years before finally getting scrapped completely in 2005.
95
Denver paid nearly half a billion dollars to build a machine, then paid again to replace it with the exact thing it was originally supposed to make obsolete.
96
United Airlines, the system's primary user, ended up spending more on manual workarounds than the automated system was ever projected to save them.
97
The future of baggage handling turned out to be a guy with a cart.
98
Deep Water Horizon wasn't one single mistake.
99
It was a chain of ignored warnings that nobody on the rig wanted to be the one to stop.
100
The drilling rig was operating in the Gulf of Mexico in April 2010.
101
Pressure tests came back showing clear signs of a problem building in the well.
102
Those results were reinterpreted as fine, essentially explained away.
103
A cement job meant to seal the well was rushed to save time and money.
104
Safety systems that should have caught the failure were bypassed or simply weren't working.
105
Then gas surged violently up the well and onto the rig itself.
106
It ignited within seconds.
107
Eleven workers died in the explosion.
108
The well then spewed crude oil into the Gulf of Mexico for 87 straight days before engineers finally managed to cap it.
109
It became the largest marine oil spill in United States history.
110
Marine life, entire coastlines, and fishing economies up and down the Gulf were devastated for years afterward, some of them still recovering today.
111
BP's total costs from fines, cleanup operations, and legal settlements eventually topped $65 billion.
112
Every single warning sign existed beforehand, documented on paper.
113
Every single one got explained away by someone trying to stay on schedule and under budget.
114
Investigators later found the rig was running weeks behind schedule and millions over budget before the explosion.
115
That pressure didn't caused the failure directly, it caused every single person who noticed a problem to hesitate before saying so.
116
Daimler-Chrysler.
117
In 1998, Daimler-Benz and Chrysler announced a so-called merger of equals.
118
It was never equal, and it was never really a merger at all.
119
Daimler, the German engineering giant behind Mercedes-Benz,
120
effectively acquired Chrysler outright while calling it a partnership just to soften the optics for American shareholders and employees.
121
The cultural clash was immediate and brutal.
122
German precision and formality met Detroit's faster, looser corporate style, and neither side made any real effort to adapt to the other.
123
Executives clashed constantly over everything, from product decisions down to who got access to private jets.
124
Chrysler's American leadership team left in waves, taking institutional knowledge with them.
125
Innovation slowed to a crawl.
126
Costs ballooned instead of shrinking, the exact opposite of what the merger had promised investors.
127
The combined company hemorrhaged value for years straight.
128
Daimler eventually admitted defeat publicly and sold Chrysler off in 2007.
129
Not to a buyer with any long-term vision for the brand, but to a private equity firm mainly interested in stripping it for parts.
130
Daimler had spent around $36 billion to acquire Chrysler in the first place.
131
It sold the company less than a decade later for essentially nothing, and effectively paid the buyer to take the problem off its hands.
132
Analysts still point to this deal as the textbook case of what happens
133
when two companies confuse merging their logos with merging their cultures.
134
Turns out spreadsheets can't measure whether two management teams actually like each other.
135
Yahoo.
136
In 2008, Microsoft offered to buy Yahoo outright for $44.6 billion.
137
Yahoo said no. It wasn't enough money.
138
Yahoo's leadership genuinely believed the company was worth significantly more than that and pushed back hard, hoping to negotiate an even higher offer.
139
Microsoft, unwilling to get dragged into a prolonged bidding war over a company it viewed as declining, simply walked away from the negotiating table entirely.
140
That moment turned out to be the absolute peak.
141
Yahoo never received another offer anywhere close to that number again.
142
Its core search and advertising business kept steadily losing ground to Google, quarter after quarter, year after year,
143
with no real turnaround in sight.
144
The company spent the next decade trying and repeatedly failing to reinvent itself,
145
burning through multiple leadership changes and strategic pivots that never actually landed with users or investors.
146
In 2017, Verizon bought what remained of Yahoo's core internet business for $4.5 billion,
147
roughly a tenth of what Microsoft had originally offered nine years earlier.
148
Yahoo didn't hold out for a better deal in the end.
149
It held out for no deal at all.
150
Somewhere in a parallel universe, Yahoo's board took the $44.6 billion and everyone involved retired comfortably.
151
In this one, they gambled on being underpaid and lost.
152
AOL Time Warner The AOL Time Warner merger is still taught in business schools as the worst corporate deal in history. $165 billion,
153
and it happened right before the Internet's first bubble spectacularly popped.
154
In 2000, AOL, an Internet company riding the peak of the dot-com wave, merged with Time Warner, a sprawling media empire with movie studios,
155
magazines, and cable networks built over decades.
156
AOL's massively inflated stock price let it effectively swallow a company several times its actual underlying size.
157
Then the dot-com bubble burst almost immediately after the ink dried.
158
AOL's dial-up internet business, the exact thing that had justified the entire merger, started collapsing hard as broadband slowly took over the market.
159
The endless synergy executives kept promising never materialized in any real way.
160
Entire divisions fought each other for resources, instead of cooperating like the deal assumed they would.
161
In 2002 alone, the combined company posted a staggering write-down of $99 billion,
162
the largest annual net loss any American company had ever reported up to that point in history.
163
The merger
164
that was supposed to define the entire future of media instead became the textbook example of what happens
165
when raw stock market hype gets mistaken for actual long-term strategy.
166
Time Warner shareholders, who held on through the entire ordeal,
167
watched their combined stock lose roughly $200 billion in value over the following two years.
168
The dot-com crash gets the headlines.
169
This merger is why it hurt Time Warner more than almost anyone else.
170
Chernobyl Chernobyl wasn't caused by one failure.
171
It was a routine safety test that went badly wrong on a reactor that was arguably never safe to begin with.
172
The Soviet RBMK reactor design had a known structural flaw.
173
Under certain low power conditions, it could actually become more reactive instead of less.
174
The opposite of how a reactor is supposed to behave when something goes wrong.
175
On April 26, 1986, operators ran a test to check how the reactor would perform during a simulated power outage.
176
To run it, they disabled several critical safety systems, pushing the reactor into precisely the unstable state its flaw punished hardest.
177
Power surged uncontrollably within seconds.
178
Two explosions tore the reactor core apart and blasted radioactive material into the atmosphere.
179
Dozens died in the immediate aftermath.
180
Long-term deaths are still debated, with estimates ranging into the thousands.
181
The total cost, including the abandoned city of Pripyat and the exclusion zone that still exists, is estimated at well over $200 billion.
182
A test meant to make the reactor safer ended up proving definitively that it never was.
183
10 Mistakes Hundreds of billions of dollars.
184
And almost every single one traces back to the same root cause.
185
Somebody trusted an assumption they never bothered to check.
186
Rarely one big explosion out of nowhere.
187
Just a small detail, quietly ignored, until one day, it isn't small anymore.
📺 Même chaîne
B23
Pointless Ancient Instincts YOU Still Fight Daily
Daily ConversationUniversity StudentsWorking Adults
B22
Every Type of Depression Explained in 16 Minutes
Daily ConversationUniversity StudentsWorking Adults
C11
30k
Daily ConversationUniversity StudentsWorking Adults
B22
Every Type of Phobia & Their Effects Explained
Daily ConversationUniversity StudentsWorking Adults
B28
Every Common Cause of Death & How to Avoid It, Explained
Daily ConversationUniversity StudentsWorking Adults
B2Every Negotiation Tactic & How To Use Them Yourself, Explained
Business EnglishWorking Adults
B23
Every Type of Friend in a Friend Group, Explained
TeensDaily ConversationHigh School Students
B21
Every Manipulation Technique & How To Combat It Explained
Business EnglishDaily ConversationUniversity Students
B22
Video
Business EnglishUniversity StudentsWorking Adults
B25
Every Major Religion & What They Actually Believe Explained
University StudentsWorking Adults
C12
Every Mental Disorder & Their Effects Explained
University StudentsWorking Adults
B214
Explaining Basic Financial Concepts YOU Should Understand
Business EnglishUniversity StudentsWorking Adults
C278
Every Drug & It's Effect Explained in 17 Minutes
Daily ConversationUniversity StudentsWorking Adults
✨ Vidéo recommandée
B21
454 phrases
Most Expensive Mistakes In History
Business EnglishUniversity StudentsWorking Adults
B21
380 phrases
You Know What You Should Do… So Why Aren't You Doing It
Business EnglishUniversity StudentsWorking Adults
B29
178 phrases
25 crazy software bugs explained
American AccentBusiness EnglishUniversity StudentsWorking Adults
B22
130 phrases
How Airports Actually Make Money?
Business EnglishUniversity StudentsWorking Adults
B2137 phrases
Solve this 1 mistake → Solve 34% of English Errors
British AccentBusiness EnglishHigh School StudentsUniversity Students
B21
216 phrases
The Leaked Audit That Exposes OpenAI's Real Financial Crisis
Business EnglishUniversity StudentsWorking Adults
B23
51 phrases
Great Moments in Unintended Consequences: Yosemite Permits, Italy Superbonus, Salary Caps (Vol. 22)
Business EnglishDaily ConversationUniversity Students
B22
62 phrases
The real reason Boeing's new plane crashed twice
Business EnglishDaily ConversationUniversity Students
B2154 phrases
The Lessons We Need to Learn from the Financial Crash of 1929
Business EnglishUniversity StudentsWorking Adults
B21
151 phrases
Could You Spend $1 Trillion?
Business EnglishDaily ConversationUniversity Students
B21
90 phrases
What Makes Expensive Websites Expensive
Business EnglishUniversity StudentsWorking Adults
B22
201 phrases
Every Level of Engineering in 15 Minutes
Business EnglishUniversity StudentsWorking Adults
Vocabulaire et conseils d’expression pour cette leçon
Cette leçon d’expression orale de niveau B2 s’appuie sur la vidéo « The Most Expensive MISTAKES Ever ». Les mots qui reviennent le plus souvent : billion, nobody, software, entire. Cette vidéo contient 187 phrases et 2344 mots à répéter en shadowing. La partie parlée dure 15:24. Le locuteur parle à un rythme naturel d’environ 152 mots par minute, proche d’une conversation courante. Seuls 77 % des mots font partie des 3 000 mots les plus courants en anglais, le vocabulaire est donc exigeant.
Vocabulaire clé de cette vidéo
Les 15 mots les plus avancés de la vidéo, avec leur prononciation et leur sens :
| Mot | Prononciation | Sens |
|---|---|---|
| reactor nom | /ɹiˈæktɚ/ | réacteur |
| merger nom | /ˈmɜː.d͡ʒə/ | fusionnement, fusion |
| automate verbe | /ˈɔ.təˌmeɪt/ | automatiser |
| metric adjectif | /ˈmɛt.ɹɪk/ | métrique |
| cart nom | /kɑɹt/ | chariot, charrette |
| suitcase nom | /ˈsutkeɪs/ | valise |
| flaw nom | /flɔ/ | défaut, fissure |
| violently adverbe | /ˈvaɪ(ə)ləntli/ | violemment |
| shareholder nom | /ˈʃɛəˌhəʊl.də(ɹ)/ | actionnaire |
| spacecraft nom | /ˈspeɪsˌkɹæft/ | astronef, spationef |
| merge verbe | /mɝd͡ʒ/ | fusionner, amalgamer |
| textbook nom | /ˈtɛkst.bʊk/ | manuel, livre de classe |
| luggage nom | /ˈlʌɡ.ɪdʒ/ | bagage |
| surge nom | /sɝd͡ʒ/ | montée, poussée |
| outright adverbe | /aʊtˈɹaɪt/ | complètement, totalement |
Les verbes à particule que vous entendrez
| Mot | Prononciation | Sens |
|---|---|---|
| end up verbe | se terminer, se finir | |
| turn out verbe | se finir, se terminer | |
| come back verbe | revenir | |
| find out verbe | découvrir, savoir | |
| roll out verbe | déployer, sortir | |
| show up verbe | apparaître, se montrer se pointer | |
| take over verbe | reprendre | |
| wake up verbe | /ˌweɪk ˈʌp/ | se réveiller |
Des phrases à répéter
Des phrases courtes et complètes de la vidéo, à réutiliser dans la conversation de tous les jours :
- You've probably made an expensive mistake before.
- Whatever you're picturing right now, it's nothing.
- That's the part that should scare you.
La grammaire de cette vidéo
Les structures que le locuteur utilise le plus, avec les mots exacts de la vidéo :
| Structure | Dans la vidéo |
|---|---|
| Voix passive be + participe passé — l’accent est mis sur ce qui arrive, pas sur qui le fait | were built · wasn't discovered · was discovered |
| Propositions relatives who / which + proposition — une précision sur une personne ou une chose | people who designed · person who noticed |
Prononciation à surveiller
Le locuteur utilise 20 contractions et formes réduites, comme wasn't, didn't, isn't. Prononcez-les sous leur forme courte, telles que vous les entendez.
- Les sons « th »: thruster /ˈθɹʌ.stɚ/, aftermath /ˈæf.tɚˌmæθ/, tenth /tɛnθ/
- Les sons « sh » et « zh »: shareholder /ˈʃɛəˌhəʊl.də(ɹ)/, navigation /nævɪˈɡeɪʃən/, assumption /əˈsʌm(p).ʃ(ə)n/, punish /ˈpʌnɪʃ/, correction /kəˈɹɛkʃən/
- Mots longs — placez bien l’accent: violently /ˈvaɪ(ə)ləntli/, navigation /nævɪˈɡeɪʃən/, investigator /ɪnˈvɛs.tɪˌɡeɪ̯.təː/, horizontal /ˌhoɹəˈzɑn.təl/, velocity /vəˈlɑsəti/
Les sons difficiles pour les francophones :
- /h/ — il se souffle, il n’est pas muet: shareholder /ˈʃɛəˌhəʊl.də(ɹ)/, horizontal /ˌhoɹəˈzɑn.təl/, behave /bɪˈheɪv/, hype /haɪp/, headline /ˈhɛd.laɪn/
- /tʃ/ et /dʒ/ — à ne pas adoucir en « ch » et « j »: merger /ˈmɜː.d͡ʒə/, merge /mɝd͡ʒ/, luggage /ˈlʌɡ.ɪdʒ/, surge /sɝd͡ʒ/, gently /ˈd͡ʒɛntli/
- /r/ anglais — langue recourbée, sans frotter la gorge: reactor /ɹiˈæktɚ/, metric /ˈmɛt.ɹɪk/, cart /kɑɹt/, thruster /ˈθɹʌ.stɚ/, shareholder /ˈʃɛəˌhəʊl.də(ɹ)/
Comment s’entraîner avec cette vidéo
- Écoutez la vidéo en entier une fois sans parler et notez les mots que vous ne connaissez pas.
- Commencez à la vitesse 0,75×, répétez phrase par phrase, puis revenez à la vitesse normale quand cela devient facile.
- Enregistrez-vous et comparez avec l’original, en faisant attention à des mots comme reactor, merger, automate.
Qu'est-ce que la technique du Shadowing ?
Le Shadowing est une technique d'apprentissage des langues fondée sur la science, développée à l'origine pour la formation des interprètes professionnels. Le principe est simple mais puissant : vous écoutez de l'anglais natif et le répétez immédiatement à voix haute — comme une ombre suivant le locuteur avec un décalage de 1 à 2 secondes. Les recherches montrent une amélioration significative de la précision de la prononciation, de l'intonation, du rythme, des liaisons, de la compréhension orale et de la fluidité.