Technology invades the modern world

Chapter 149 Professor's miraculous healing touch brings Georgetown back to life (5k)
MIT Radiation Lab Series

Most people probably haven't heard of this name.

In short, this is an encyclopedia covering the field of radio.
In 40, in preparation for World War II, America transferred the nation’s top physicists and engineers to MIT and established the Radiation Laboratory.

The laboratory's projects cover radio, microwave, and radar; most of the radio and radar equipment used in World War II was produced here.

After World War II ended, the government stopped providing funding.

The researchers received an ultimatum, with only six months left.

We need to use this time to organize our existing research findings into a tutorial. Then the lab will close, and everyone will go back to where they came from.

Simply put, it now encompasses almost everything related to radio frequency electronic circuits, laying the foundation for America's later electronic technology.

Watson narrowed his eyes: "Professor, I don't quite understand."

This set of books was published by McGraw-Hill back in 47, and many university libraries have copies. Borrowing a set shouldn't be difficult for someone of your status.

The MIT Radiation Laboratory Series was published in 1947, but America realized something was wrong and stopped publishing it after 1953.

Watson is right; many universities have collections of these.

However, to view the content, the borrower's identity needs to be verified.

Given Lin Ran's status, obtaining this security clearance would naturally not be a problem.

“Of course, but what I want to borrow is the one IBM has.” Lin Ran smiled and raised his wine glass.

That's right, the universities received a censored version.

Not to mention universities, even up to 2020, the "MIT Radiation Lab Series" that you can find in public places is a censored version.

It covers most of the theoretical aspects, except that the volume on "Microwave Magnetrons" might have less content.

However, the content available in the app is severely censored.

Many of America's core radar applications during World War II were cut.

Only a very few organizations have the complete version.

One of them is IBM.

Although IBM focused primarily on punch card machines and early computing devices during World War II, IBM engineers had extensive technical exchanges with the MIT Radiation Laboratory, particularly in electronic signal processing and computing support.

After the war, IBM quickly entered the field of electronic computers, absorbing many achievements from radar technology, such as high-speed circuit design.

Lin Ran even knew that this collection of books was kept at IBM's headquarters in Yorktown Heights, New York.

Although it has MIT in its name, the full version doesn't even have MIT.

Complete copies of this collection are only available at Lincoln Laboratory, Lawrence Berkeley National Laboratory, America National Institute of Standards and Technology, Los Alamos National Laboratory, and IBM.

Of course, IBM is the easiest for Lin Ran to get.

While promoting artificial intelligence is one aspect, obtaining the complete "MIT Radiation Lab Series" from IBM is an even more important one.

The fact that this series of books can be sent back to China will save a lot of time in every aspect.

This series of books from the 1940s has unique value, not to mention in 1963, even sixty years later.

For a country developing radar, missile guidance, and communication technologies, this book provides a systematic theoretical framework, design methods, and engineering practices that can significantly reduce trial-and-error costs.

“Professor, this is probably not something you should know,” Watson continued.

Lin Ran pressed down with his hand: "Watson, calm down. You must know about my relationship with Jenny Hearst."

As fellow members of the publishing industry, isn't it normal for the Hearst family to know that McGraw-Hill published two sets of this series back then?

Is that strange?

Hearst was also a media empire, and McGraw-Hill needed to place advertisements in the Hearst family's newspapers or magazines to promote their textbooks or professional books.

Given the business dealings between the two parties, it's not surprising that they knew there were two versions of the "MIT Radiation Labs Series".

Lin Ran continued, "I know what you're worried about, but your worries are completely unnecessary."

If you send it to Redstone Base, I will only read this series of books in Redstone Base and will not take it out of Redstone Base.

You could even get McNamara to testify for me.

No matter how clever you are, you can't guess the existence of the door.

Watson immediately relaxed upon hearing this: "Professor, you should have said so earlier."

Lin Ran meant to follow the proper and aboveboard procedures.

First, IBM should submit an application to the Department of Defense. After the Department of Defense approves the application, the books can be taken to Redstone Base.

Lin Ran will not have any contact with the entire process, and will not be outside the jurisdiction of the Ministry of National Defense.

After all, although Redstone Base belongs to NASA, a large part of its security is still handled by the Department of Defense.

Watson immediately relaxed. I thought you were going to do something illegal.

For example, I could send this set of books to your home, and then, inexplicably, the Russians would use a miniature camera to put it into a miniature film roll, which would then eventually reach Moscow.

As long as you don't intend to play like that, it's fine.

Watson thought to himself, "Professor, no problem, I will definitely do it."

Lin Ran smiled and nodded, thinking to himself, "You'd never guess that I have a way."

Two months passed in the blink of an eye, and Lin Ran commanded the entire team with the efficiency of a tyrant.

Although he only spends half a day or less at the Georgetown project each day.

However, he forced Calvin to collect Russian and English scientific and technological literature, count high-frequency word pairs, and personally guided Cuthbert and Peter to modify the IBM 7090 program to add a frequency-based word selection module.

Whenever someone made a mistake, he showed no mercy: "You can't even do this right? Are you engineers or just useless?"

Two months have passed in the blink of an eye.

The inspection team was personally led by Secretary of State Rusk and Secretary of Defense McNamara.

Those members of the Georgetown project who had prior knowledge of the news couldn't help but exclaim, "The professor truly lives up to his reputation!"

"If it's not a professor, it's at most Colonel James Reed coming to inspect our project."

"Yes, only a professor could have such influence."

"The Secretary of State and the Secretary of Defense, this is the treatment afforded to manned spacecraft."

Originally, at most a colonel would have come, but now a civil official comparable to a five-star general has arrived.

In a laboratory at Redstone Base, the air was thick with tension and anticipation.

Fluorescent lights cast a bright yet cold light, illuminating the main subject in the center of the room: a massive translation machine with several IBM machines connected in parallel, its metal casing covered with knobs and switches.

If it weren't for the experience gained from DeepBlue last year, we really wouldn't have been able to get it done in just two months.

A low humming sound surrounded the machine, mixed with the heat emanating from the electronic equipment and the unique smell wafting from the air conditioner in the distance, which was used to maintain a constant room temperature. All of this testified to the team's continuous improvements and adjustments to the machine over the past few days.

Calvin stood beside the machine, his posture upright, his eyes revealing focus and confidence.

As the leader of the improvement work, he knew the machine's performance inside and out.

Today, he will present his findings to this special group of visitors: officials, staff and experts in the field of translation at the Ministry of National Defense.

They stood on one side of the laboratory, each with a different expression—some curious, some doubtful, and some even indifferent.

Many of them were old acquaintances.

Those who work as translators have known about the Georgetown project for a long time.

Everyone had high expectations from the beginning.

Of course, we hope this thing will reduce their workload.

As a result, they waited and waited, but the Georgetown translation machine was never provided to them.

They had also visited Georgetown University before, and their assessment was that it was a complete waste of space and unusable.

The reason everyone came this time was entirely because of Lin Ran.

They wouldn't have come if it weren't for the professors. Of course, McNamara is the most confident in revitalizing the Georgetown project.

"Professor, why don't you introduce it yourself?" McNamara asked Lin Ran, who was standing next to him, in a low voice.

"I only played a small supporting role for them; the credit actually goes entirely to the researchers from Georgetown University and IBM," Lin Ran said.

McNamara feigned displeasure: "Professor, you're too modest."

I've seen all the statistical methods you designed for them. Without your involvement, those stupid guys at Georgetown, and those guys at EM, probably wouldn't have come up with anything even in ten years.

While Lin Ran was whispering with McNamara, Dr. Paul Gavin, a linguist working for the Department of Defense, frowned slightly in the crowd, tapping his pen lightly on his notebook. He had always been skeptical of the machine's actual capabilities.

Even renowned professors personally participate in it.

Colonel James Reed, who was next to him and had been in charge of approving the Department of Defense's budget for Georgetown and tracking the project, stood with his arms crossed, expressionless, and in his crisp military uniform.

Rusk, holding a notebook, kept glancing between the machine and Lin Ran, wanting to interject between Lin Ran and McNamara, but hadn't decided what to say.

Calvin cleared his throat, his voice steady and firm: "Gentlemen, today you will witness the latest advancements in machine translation. Under Professor Lin's guidance, we have significantly improved the system's accuracy and its ability to handle complex language structures."

Dr. Gavin adjusted his glasses and asked first, "An improvement? The previous version performed terribly in handling idioms and syntactic ambiguities. How did you address that this time?"

Calvin confidently stated, "We introduced a frequency-based statistical model, and by analyzing bilingual corpora, the system is now better able to distinguish word meanings and handle context-dependent translations."

Colonel Reed interjected coldly, “That sounds nice. Let’s see the results. Time is of the essence; we need results.”

Calvin nodded and turned to walk towards the machine's control panel: "Of course, we'll begin immediately. Let's start with a simple demonstration."

He skillfully typed a Russian phrase on the keyboard: "Мыговоримомире".

The machine immediately started operating, its internal parts emitting slight clicking and humming sounds. A few seconds later, the screen displayed the English translation: "We speak about peace."

A murmur rippled through the room. Dr. Gavin raised an eyebrow. "Not bad, but that's just a simple sentence. What about something more difficult?"

Calvin was prepared: "Of course. Let's try ambiguous sentences."

He typed in another Russian phrase: "Ончитаеткнигу". The machine quickly output: "He reads the book."

Dr. Gavin nodded, but was still not satisfied: "What if the context is different? For example, what if 'книга' refers to an account book instead of a book?"

Calvin slowly tapped his fingers on the keyboard, typing a new sentence: "Бухгалтерчитаеткнигу".

After a brief pause, the machine displayed: "The accountant reads the ledger."

A gasp of astonishment filled the room. Colonel Reed slowly lowered his arms, a flicker of surprise in his eyes. This was completely different from the Georgetown machine he had seen before: "How does it make that judgment?"

Calvin explained, “The system can now make inferences by combining contextual clues and probabilistic models. 'Бухгалтер' (accounting) hints at the financial context, so 'ledger' was chosen instead of 'book'.”

The staff member who was previously in charge of translation thought, "This could be revolutionary for intelligence work."

Dr. Gavin approached the machine, his eyes showing interest: "I want to see how it handles more complex sentences, such as multiple clauses or idioms."

“Let me try.” Gavin sat down in front of the input device.

Calvin nodded and said, "No problem."

Dr. Gavin pondered for a moment and typed: “Времялетит,когдатывеселишься”.

The machine quickly output: "Time flies when you're having fun."

Colonel Reed remained unmoved: "It's good, but we need it to handle military terminology and technical vocabulary. Our needs are very specific."

Calvin looked him straight in the eye: "Dr. Calvin, you can try some more technical terms. We've started compiling a technical glossary, including military and scientific terms."

After a moment's thought, Dr. Gavin typed: "Ракетаоснащенаинерциальнойнавигационнойсистемой".

The machine translation reads: "The missile is equipped with an inertial navigation system."

A flicker of emotion crossed the colonel's eyes, and a rare hint of approval appeared: "Quite accurate."

The room fell silent for a moment, and everyone felt the weight of this moment.

The officials exchanged glances, their skepticism gradually turning into optimism.

A defense expert, a sharp-looking woman, spoke up: "Professor Calvin, your work may change the course of our actions."

The speed and accuracy of this machine have indeed improved significantly compared to the past.

How long will it take before it can be widely adopted?

Calvin paused for a moment before cautiously replying, "With more resources available, we can begin expanding the system within six months. However, we will still need to continue optimizing the algorithm and expanding the language database."

McNamara clapped his hands and turned to Watson, saying, "Mr. Watson, congratulations, you've secured a contract from the Department of Defense."

Then he said to Lin Ran, "Professor, I knew you could do it."

Finally, Calvin said, "Professor Calvin, the Ministry of Defense will contact you later."

Lin Ran said:
"Don't be too optimistic."

Georgetown can only be described as usable; it's still a long way from being used effectively.

IBM needs to improve the machine's performance; the better the performance, the higher the translation accuracy.

Secondly, Georgetown University should continue to follow the technological path I designed and explore its potential.

However, this technical approach to machine translation has its limitations.

Only through updates and improvements in statistical methods can better translation results be achieved.

However, once you've explored the limits of this technical approach, I suggest that Georgetown University recruit more PhDs in statistics and theoretical mathematics to participate in this project.

Finally, regarding the Ministry of Defense, when using the Georgetown machine for Russian translation, it will still rely on manual review.

It can help you improve efficiency, but the degree of improvement is limited; you cannot rely entirely on machines.

In fact, Georgetown's translation is somewhat similar to the earliest translations, a purely mechanical one.

It will not be optimized according to language characteristics.

It's just that English and Russian are both related to the Latin language family, but only related; the correlation isn't low.

But it's definitely much simpler than translating from English to Chinese.

"The current version of the Georgetown translation machine performs better in the specific fields of military and technology, but its error rate is higher in other fields, as well as in complex sentences and non-professional contexts."

Russian is a highly inflectional language, with nouns having six cases and verbs having complex tenses and aspects, while English grammar is simpler, relying only on word order and prepositions.

The Georgetown machine still faces considerable challenges in accurately handling case changes and word order adjustments.

Currently, we have only mitigated the polysemous word problem through probability tables and context windows, while the rule system ensures the accuracy of basic grammar.

Simple sentences can achieve an accuracy rate of about 80% to 90%, but complex sentences, such as nested clauses, are still easy to make mistakes in.

It should work well for clearly structured texts such as news releases and diplomatic statements.

However, it struggles to accurately capture subtle differences in conversations involving a large amount of spoken language, and the error rate can reach 50%.

I think everyone needs to maintain a cautiously optimistic attitude.

After Lin Ran finished speaking, Rusk said, "Professor, that's enough."

For the current needs of the White House and the Department of Defense, it is already very useful if it can process written texts in batches, such as news and reports.

We didn't expect him to provide accurate translations for a conference call anyway.

Indeed, as a project that was almost sentenced to death, was canceled a year later, and was even considered an AI winter, it was saved by Lin Ran in just two months and now has practical application value.

This is already a leap forward.

Visitors left one after another, their excited whispers echoing in the corridor.

McNamara insisted on going to Lin Ran's office to sit for a while.

"Professor, I have approved your request to borrow the 'MIT Radiation Lab Series'."

IBM should approve it as early as the day after tomorrow. I'll have Lincoln National Laboratory provide you with information including some radar design plans from that time.

"You should also take a look at this kind of content and see if you can offer us any inspiration," McNamara said while sipping his tea.

Lin Ran thought to himself, "Damn, what a good person!"
(End of this chapter)