Chapter 11 Appendix: How to make the Chinese character generator for the heart of a Chinese computer

This appendix was decided to be added after the proofreading was completed. Originally, I planned to publish it separately. After reading the book, my colleagues expressed that the content was insufficient and needed to be rewritten.It is unlikely in the current busy situation, and it is the best policy to combine the two books into one. As an appendix, there is another layer of meaning. I made a high-profile statement earlier, but I didn’t provide young people with a direction to get started. It’s a sin!sin! Therefore, I am willing to disclose the exclusive technology of the Chinese vector group lettering that I currently rely on for my livelihood, and hope that young friends can come to study together.

However, there is a statement that it is one thing for me to disclose the technology, and another thing for my product rights. The product has already been listed and has patent rights. If anyone wants to copy and steal without thinking, please try it. I will fight at all costs (although I have already lost all my property for research over the years, and I am in debt of more than 6 million yuan), fighting for a reason and fighting for the rights and interests of all research inventors. As for providing technology, so that you can also research, I hope you can do better, that is my wish.If you add new ideas and do better than mine, I'll be the first to congratulate you and learn from you.Don't forget that Chinese culture is not for personal self-interest, nor is it someone's responsibility. Only when everyone contributes their energy and wisdom can we carry on the past and open up the future.

The vector group lettering has cost me eight years of hard work, and it has also invested countless precious time and wisdom of more than a dozen colleagues in our company. It can be today, and there are more silent unsung heroes who should be thanked. They and you and me They are all Chinese, and both love their own cultural traditions.I made these technologies public today, not as a meaningless sacrifice, but as a way to attract new ideas. It is said that in 1972, I was working as an artist for a large cultural publishing company CUTURAL ABRIL in Brazil.At that time, I had been wandering overseas for eight years and worked in more than 30 different industries.Alone, alone, every day I ask myself: Another day has passed, why am I alive?What is the meaning of life?Of course there is an answer, but it is out of reach.

Everyone pursues happiness in life, but for a Chinese, happiness in life is not just material enjoyment.Because, there are one billion compatriots in the vast territory, and because of the ambition of a few people, unprecedented catastrophe has come.In Taiwan, the land is narrow and densely populated, and the economic conditions were not ideal 20 years ago.Therefore, I left my hometown, thinking that it would not be long before I could carry the flag of the returned overseas Chinese leader and return home in good clothes! The problem is not so simple. In North America and South America, wherever I go, the industries that Chinese people can work in are laundry shops, restaurants, and Chinese and foreign groceries.Furthermore, due to estrangement, the Chinese are regarded as stubborn, mysterious, and unprogressive monsters.

Because it is not easy for Chinese people to make a living, they save small money to make big money by living frugally.After getting rid of poverty, he immediately tried his best to isolate himself from his compatriots, and he was ashamed to keep company with him.For many years in the Chinese circle, this is the way of carp jumping over the dragon gate. The old and the new alternate. It is no longer the fish that jump into the dragon gate, which makes the fish envious! I have made in-depth observations and experienced the mentality of each class, and the only conclusion I have come to is: Chinese people's upbringing is good in essence, but it is limited to knowledge, especially various professional knowledge in modern society, and socializing with foreigners The interface knowledge is too lacking, so that they first seek self-protection and thus become autistic, and because of autism, they feel inferior.In the end, even after making a fortune, they still have no sense of stability, happiness, or achievement.

This situation has many similarities with the early society in Brazil. Brazil is a paradise of vast land and abundant resources. 24 degrees Celsius, except for the extreme south, it is like spring all year round. Brazil has a mixed race and is a paradise for immigrants. Because immigrants have a higher level of knowledge and are relatively thrifty, they all achieve success, which arouses the attention of Brazilians and gives up on themselves. This phenomenon did not occur until April 1964, when a group of young military officers overthrew the left-leaning President Jiang Gula and carried out strict reforms. They believed that the modernization of the country must start with the improvement of education and knowledge, so the establishment of this A cultural publishing company.

In the year I worked for the company, the company's turnover accounted for the thirtieth place among all kinds of enterprises in Brazil. The company had 5,000 employees, and published 98 professional magazines every week, as well as numerous irregular books and publications! This corporate means of disseminating knowledge is indeed effective. In recent years, Brazil’s economic growth has surpassed that of my country. You can’t imagine that they were considered barbaric more than ten years ago, and 80% of the citizens were illiterate! One day, the supervisor of my small unit was vacant. Due to work, the editorial department sent me an urgent letter. It was a novel that had just been published in the United States and was a bestseller. After it was translated into Portuguese (Portuguese is used in Brazil), it was designated Available today.

I was stunned: listed today? how?Are you new here?The people in the editorial department looked suspicious. I don't dare to ask more questions. I just do it according to the standard work anyway. I have ten thousand doubts in my heart. It is based on my experience in China. It takes at least one or two months to publish a book! I first made a large sample, and then sent the manuscript to the typing room. I recorded the time as 8:40 in the morning, and I told the foreman: It will be listed today! The foreman is a blond girl, she said: Yes, it must be ready before noon!I was afraid that she misunderstood my Chinese Portuguese, so I emphasized it again: there are more than 200,000 words in total!

She smiled and said challengingly: How about it?I'll leave it to you at eleven o'clock. I couldn't believe it, and walked into the typing room with her. There were 20 terminals in it, and only 15 people operated them.The foreman went in and randomly picked up the draft paper, and then divided it into fifteen copies, distributed them on the typing frame, and said to the typists: Ladies!Show it to the Chinese!We hand in papers within two hours! .I am the only Chinese in the company, so it became my name. When the dozen or so little girls heard it, everyone liked to accept the challenge. They cheered up and their fingers were flying. Everyone entered the computer at an average professional speed of 100 words per minute before eleven o'clock.

Immediately afterwards, I took the manuscript printed by the computer and sent it to the proofreading team, and then we finished the manuscript. By 4:00 p.m., I had sent all the completed manuscripts to the plate-making department, and my responsibilities were all fulfilled! In order to know how efficient the company is, I looked for a relatively large newsstand at night and waited from seven o'clock. Looking at the watch, at about seven forty-five, a large truck painted with our company logo sped away. When I came over, I didn't even stop, a large bag of things was thrown from the car, I stepped forward to have a look, it was right!It was the best-selling novel!

This is a novel experience!If one of the most popular bestsellers can be introduced from another country to one's own country within a day, then all kinds of novel and meaningful knowledge can be brought to our people at the same time in the shortest possible time. They share!If we know for sure that what we get is the latest and most valuable first-hand information, what are we lagging behind?Why are we ashamed? Human nature has something in common, such as a basketball game!If the scores of the two teams are too disparate, the fighting spirit of each other will be affected.Those who have lost too much just want to step down as soon as possible, and those who win may want to pay more points, or they may save some face. No matter what, the schedule is unbearable.But when the gap is between a few points, it becomes tense, and no one will admit defeat. After the game is over, they are still gearing up: look at me next time! . If we ask a large number of international students to search for us in the United States or the world and translate them into Chinese (I even hope to have a computer translation machine), we will immediately use this similar cultural company to use the fastest method and the most reasonable price , and transmit these new knowledge to every corner of the country.If you know one, you want to know the other. A series, from all walks of life, of various levels, interesting, educational, and endless, are delivered one after another. Who is not curious?Who wouldn't want to accept something new that wasn't completely foreign to him? What a person rejects is what he does not understand, and it is human nature to reject what he doubts.A stranger comes to you, takes out a pill and says to you: Take it and you will live forever! , if you eat, you're an idiot!Who doesn't want to live forever?But we must understand the principle, and then prove its feasibility. The future of the country is the same. Some people say that the country will be prosperous and strong. If I don’t understand what he said, I would be an idiot if I believe it!In the same way, if I want to convince someone about something, I must first let him understand all the relevant knowledge. If we are facing compatriots all over the country, how can we convince so many people? This is the answer I see, an efficient cultural publishing company.However, there are still some technical problems. How can we print a book of 200,000 words in two hours? Having found a bright road, not only has the meaning of my own existence, but also benefit the people.I left the company because of this, and all the homework is clear, the key lies in Chinese input. There was no opportunity and environment in Brazil, so I returned to the country in 1973 (the 62nd year of the Republic of China), fully expecting that someone would support my plan.In the end, in order to make a living and to experience the publishing situation, I wrote two unpopular novels. From the beginning to the end, it took at least two and a half months for each book, which was seventy-five times slower than the Brazilians! (Of course they are organized practices!) One cannot live by ideals. On the one hand, I have a part-time job, and on the other hand, I started a long-distance running of Chinese input. Until today, it is a pity that my ideals have not been fully realized.Everyone sees me as a lunatic, even though this lunatic has proved time and time again that he has no wild ideas, but after a hundred successes (just successes in work, not authority successes), the one hundred and first idea is still considered insane by others So simple. What's worse is that I don't know how to make money, and I don't want to make cultural property, so I am forever doomed to lead our company's group of fools who don't know what fame and wealth are, and work hard silently! Cut the gossip!Now let me talk about the technical problems related to Chinese import and export from the initial conception to today, and their solutions.If you are not interested in boring technical issues, let's say goodbye!I have worked hard for you for too long! In the Kangxi dictionary, there are more than 40,000 characters in Chinese, each character is a unique individual, and there is no obvious relationship between characters. The texts that have been handed down today are continuously compiled, inductive and interpreted by scholars in the past dynasties, and it is only possible to have such a scale today.From Xu Shen’s Shuowen Jiezi in the Eastern Han Dynasty (the book consists of 14 chapters, with 9,353 characters, mainly in Xiaozhuan, divided into 540 parts according to the shape and meaning of the characters, and the meaning of each word is first explained , secondly explaining its shape, and then citing ancient scripts, 籀wen, Xiaozhuan and other popular scripts different from Xiaozhuan, with a total of 1,163 characters.) In the Song Dynasty, there was Zheng Qiao’s Liushulue, Classified according to the six books; Ouyang Xiu's Jigulu and Zhao Mingcheng's Jinshilu have made great contributions to the sorting and classification of words.Due to political factors in the Qing Dynasty, the study of philology reached its peak. Ruan Yuan wrote a collection of ancient Zhai tripod inscriptions, and the research was very detailed.In addition, Duan Yucai's commentary on explaining characters, and Zhu Junsheng's commentary on Wen Tongxun Dingsheng are both books with outstanding contributions in the history of philology.After the Republic of China, Luo Zhenyu, Wang Guowei, Zhang Binglin, etc. also had their writings.Today, Chinese characters are facing new challenges｜using characters in information tools｜the problem of Chinese input. I have emphasized before that to solve Chinese input, you must use a small keyboard. The text on the small keyboard only occupies 26 keys. How can 26 keys be enough?If you want to use a few more keys, you can consider: 1. Punctuation keys, 2. Number keys, 3. Uppercase and lowercase letters distinguishing keys. Punctuation keys should not be used for a simple reason: if you type an article, the most commonly used is punctuation.If the punctuation key is occupied, when using punctuation, it is necessary to identify under what circumstances it is a text with other control keys.At that time, the most commonly used symbols become the most troublesome.Therefore, this road is blocked, and the punctuation keys must not be occupied. Secondly, the number keys also have their own difficulties, because when processing data, there are often more numbers than words, which is the same as the reason why the punctuation keys cannot be used. Another example is to apply both uppercase and lowercase letters, and add 26 symbols. It seems much more ideal, especially when using Chinese, there is no reason for uppercase and lowercase letters, why not use them by the way?A problem is involved here. Chinese input is a fundamental plan, which is related to the prospect of Chinese information processing.We need to know that there are no lowercase letters on teletypewriters. Since we want to accommodate the existing system, we cannot give up the function of using Chinese teletypes. Therefore, it is best not to use uppercase and lowercase keys.In addition, the conversion of uppercase and lowercase letters requires pressing one more button, which is so inconvenient to use!Therefore, there is only one most difficult road to take!No matter what method is used, it must be completed with 26 keys. English is a pinyin system, and my country also has phonetic symbols, and no matter how many phonetic letters there are, can it be considered to use phonetic input? Of course, the telecom bureau has used it very successfully, but where it is used has a lot to do with its success or failure.Telecom bureaus are very suitable for number checking. Once you connect to 104 and tell the service lady a name, you will send her a voice over the phone. As long as the pronunciation is consistent, she will immediately complete the input procedure.However, since there are only 1,300 sounds in Mandarin, and tens of thousands of characters, the repetition rate is too high, so you have to explain the relevant definition of the name, and the service lady can make a second choice. The problem is in general data processing. We also mentioned that blind key pressing is one of the prerequisites. Let's take the application of 15,000 characters as an example!There are 1,200 sounds, and the repetition rate has exceeded 1,000%. Taking a word as an example, there are about 100 homophones in this word. What method is used for the second selection?with eyes?That does not meet the principle of blind button. The more serious problem is that unless they are professionals, few people can correctly spell the 16,000 Chinese pronunciations!To tell you the truth, out of the 5,000 Chinese characters I know, 3,000 of them are incorrectly pronounced. How much better are you than me?Even if you spell eight thousand characters correctly! (Experts are nothing more than that!) And half the words you can't use.Another example is that I hope to use all Chinese, more than 40,000!God knows if there is such a person in the world? Obviously, if Zhuyin was feasible, Chinese input would have been solved long ago. Others are not fools, and there are many Chinese people who use their brains! What about the four corner number?In the same way, if possible, there is no need to wait until today. There is only one most practical way in the world, and that is to work hard. Don't look for ready-made and simple ones, just think about it!Are you the only one who thinks Chinese input is important? My opinion is so unreasonable. I believe that many experts and scholars have worked hard and achieved a lot, but the fact is that there is no one way to declare success.I am far inferior to others in terms of ability and knowledge. If I have to refer to them and learn some skills, I will only walk on the road they have developed and always walk behind them. Therefore, I bought five small dictionaries in one go, and cut out the characters in them one by one, regardless of radicals or strokes, and I decided to take a new path. The original idea was, since words are used by people, where should people use them?Nothing more than natural things, man-made tools and equipment, human relations and miscellaneous categories that cannot be classified. Fortunately, a little knowledge of history helped me identify the most important set of letters, which are the Yin-Yang and Five Elements emphasized by Han Confucianism.This is the foundation of Chinese philosophy, physics and chemistry, and has a profound impact on culture.At that time, I also hoped that this effect would show a significant response in the structure of the text, so that I could reduce some unknown factors. This bit of luck has such a great impact that it can be regarded as the key to my success or failure. In the following years of hard work, fortunately, I have the unwavering confidence of these seven letters, just like the seven buckets in the sky, staying where they are With all the stars together, I can use various combinations to experiment and analyze, and gradually, one by one, I can find other letters. These seven letters are also a bit of a coincidence, they are the moon character. The moon is not used much in the text, and I regard it as the code of the Yin character of Yin and Yang. I don’t know if the ancestors had the foresight. The writing on the side is the same as the moon character (at least it has evolved to be exactly the same today), so the day character becomes the code of yang, and the other gold, wood, water, fire and earth are in place respectively. With seven letters, we began to classify, and first put together the words related to the meaning of these seven elements, and about a quarter of them have been solved.After that, I couldn't find any clues.I moved these characters around every day, cut out dozens of dictionaries, and arranged and combined them thousands of times. Finally, I barely made up the first set of alphabets, and named them Chinese Xingyizhizi. Published in May 17 and registered with the Ministry of the Interior. The first set of letters is like this: sun, moon, gold, wood, water, fire, earth, human heart, hands, feet, mouth, eyes and ears, Wang Shishan, insects, fish, dogs and horses, clothes, words, silk, grass and bamboo. In addition to letters, there are 120 auxiliary glyphs.Using these letters, we sequence the 8,000 words of the dictionary through coded sequences.If only get three codes for every word, repeated word is as high as 8%, if auxiliary font is used auxiliary key again (equal to increase 26 keys), repeated word then drops 0.5%. As for the arrangement of letters, I have also considered the cooperation with the frequency of use (will be further analyzed later), but the frequency of use itself is difficult to define.For example, according to the statistics of Jiaotong University, the most frequently used characters are based on articles in newspapers. If you use a telephone directory as an object, you may not be able to find the characters used by a few people as names.At the same time, in terms of data processing, articles are only a small part, and most of them are names, materials, and numbers. So, which one should be used as the standard for frequency of use?So, in the early days, I could only arrange by category. In any case, I know that I am still far from the ideal purpose. It is unacceptable to use 52 keys. With 26 keys, the repetition rate of 8% is too high, and I have no way to start with the frequency of use.At best, so far, it can only be proved that this road seems to be feasible, and the result remains to be worked on. In July 1967, through the recommendation of my friend Lin Junfu, Dazhou Construction Company agreed to support me in further research work, and hired Shen Honglian from the Chinese Department of National Taiwan University to assist me. This time, for the sake of practicality, we selected the vocabulary of the Mandarin Daily Dictionary as the standard, first compiled the vocabulary card, and then selected the letter group code. At this time, due to a very accidental opportunity, I saw the triangle number of Wang An Computer Company in Zhonghua Bookstore. It has nothing to do with the composition of the word itself, so only half of the problem can be solved, and the problem of the other half of the group of words will be a headache again. How can Chinese characters be combined?I assume that there is a set of basic fonts, just like typing in English, and these fonts are installed on one end of the lever. As long as I can control the lever with the input code, the characters can be pieced together in sequence.Assuming the above conditions hold, how do I start? This assumption is very bold, and in the end I finally gave up because it was too difficult to implement.But this assumption had a decisive influence on the later development of the vector grouping method, because without this assumption, I would never find the law of grouping characters.Simultaneously, if it is not for this assumption earlier, my code fetching method cannot be closely combined with the group word. Here's a law I've learned from experience: the infinite number of insoluble problems can only be solved by assuming that some of them have been solved, and then concentrating on the remaining ones. Assuming that there are already fonts, how many fonts are the least and most necessary?How many words are involved in this question?What is the standard of choice. I will make another setting. If you consider a typewriter, three or four thousand commonly used characters should be enough. Leave the more difficult characters aside for the time being, and start with the ones with a high combination rate. With the direction, the problem is simple. I first researched the characters from the second-generation group code, only to find that there was a lot of room for discretion in the classification method at that time. Because to combine glyphs, you must understand the structure of Chinese glyphs. Chinese is a square character, which was originally transformed from pictographs, which is due to the evolution of tools and the habits of users. As the name suggests, pictographic characters are obtained by the ancestors based on the image of things and simplified with simple lines. Everyone knows that drawing pictures is not for everyone, and requires the control of observation skills and skills.Therefore, when the application of characters becomes more and more popular, the elegant and beautiful graphics in pictograms will gradually become rigid and symbolic, and the circles will become squares, and the curves will become straight lines (you might as well try it, straight lines are far easier to draw than curves) ). Secondly, with the improvement of cultural level, the expression of characters is wider, and the demand for new characters arises as the times require. The ancients summed up a set of rules for character creation, that is, six books: pictograms, instructions, knowing, phonetics, Zhuanzhu, and borrowing.Among them, the pictophonetic characters contributed the most to the creation of characters in later generations. Phenomenon is half using shape and half using the function of sound to combine characters.Since the pictographic font itself has pronunciation, the pictophonetic characters do not increase the complexity of the glyph itself, but only provide the combination principle of the glyph. Today, the vast majority of Chinese characters are pictophonetic characters. Based on this principle, I also found a direction, which is to use the combination principle of pictophonetic characters to divide the font into two parts: the initial and the body. This is a new breakthrough. According to the original classification, among the 4,000 characters, I found about 100 prefixes and 300 characters.Using 400 radicals to form 4,000 characters is still troublesome, but it has made some progress.What excites me even more is that about 10,000 characters composed of these 100 prefixes and 300 prefixes have actually been found in dictionaries, but I have not collected them in the research materials. Encouraged by this, we worked hard to organize and try to find the best combination among them. One thing is certain, that is, with the twenty-six keys as the input code, how can the best combination be obtained when fetching the code?To cooperate with the rules of character formation, it is necessary to take into account that the number of codes to be taken is small while avoiding repeated characters, and the requirements for character formation must also be considered, because according to years of experience, as long as there is sufficient patience, manpower, and time, there will be a day An optimal letter code will be found.However, if the word grouping problem is not solved first, the code is determined and the font cannot be matched with the output, and the result is still nothing. When we finally made an analysis by telegram, I finally had a basis for my decision.The result of the analysis is that there are 240 prefixes and 1,200 characters (many characters are used only once or twice). Most of the characters that cannot be divided into prefixes and characters are the body itself, which is also a complete character. There are 2,500 such characters. The data proves that since there are 400 prefixes, if more characters are collected, the prefixes will inevitably increase, but it is estimated that the number of prefixes will not exceed 400.We have 26 keys, each key has a code, and 26 keys can only represent 26 words.Therefore, 240 characters must be 26^2. In other words, 26 keys must be used to fetch twice. There are 676 combinations of two codes, which are enough to contain 400 prefixes. Let me explain again, if there are 400 prefixes, the number of prefix codes must be greater than 400 to avoid repetition, so the prefix codes are one to two. Looking at the character body again, there are 1200 character bodies. Similarly, two codes are not enough. Only increase to three codes, and the combination is as high as 17,576. It seems a bit wasteful, but the problem cannot be solved without doing this. There is another advantage of using one to three yards for the body of the characters, because many of the characters are also a complete character, which can still be divided into the initial and the body.Let's assume it is the second prefix and the body!The second prefix can also meet the requirement of taking one or two yards for the prefix, which is very useful in forming characters. With this method, we can not only combine all existing characters, but also follow the rules of the ancestors to create characters. Combine tens of thousands of innovative characters.So far, Chinese characters are no longer dead characters, they can be used as flexibly as pinyin characters, and they are more powerful. As for the technical problem of forming characters, we will further explain, now we must go back to the principle of input, so that you will have a complete concept. The general principle is determined, one to two yards for the prefix, one to three yards for the body, and one to five yards for a complete font. In terms of coding technology, the most important thing is to be easy to learn and apply. To be easy to learn, you must follow a set of simple methods, and you must not rely on rote memorization.However, Chinese characters already exist and cannot be changed. No matter what method is used to sort them out, the fact of empty codes cannot be avoided. The only exception is to use the method of telegram clear codes, and code according to the order of radical strokes. Let's see what are the advantages and disadvantages of this rote method?Its advantage is to collect 10,000 characters, that is, 10,000 numbers, which will never be repeated.Telegram codes have been implemented for decades. If the problem can be solved, why wait until today to have a headache for Chinese information? Someone proposed a cure for headaches, which is recoding.There are too few words in the telegram, so adding more words is enough?How much do you add?Just add it to 16,000 words!As for whether 16,000 words are enough?According to household registration statistics, there are currently 16 million people in the country, and there are more than 18,000 characters in personal names, and the best part is that more than ten new words are born every month! New vocabulary?You haven't heard of it!The reason is very simple. Newborn babies need to be named. Some parents are well-educated people who take out the Kangxi dictionary to find words. Who can say that these are not Chinese characters?It's just that you haven't seen it!Some parents are illiterate, but they have the right to draw a word for their children. In today's society, everything is done according to the law. Who dares to say that this is illegal?In addition, the township office is busy with work, and all documents have to be copied by hand. If you are not careful, a word is added or a line is omitted, and the original document is in black and white. Who dares to change it?If it is changed, it may be a forged document! What is the case today will not be so good tomorrow. Unless the country has a set of writing standards, the vocabulary will only increase day by day.If a standard is set, who will set it?How many words do you order?Do we recognize the words in the Kangxi dictionary?It is very simple to admit the account, at least 43,000 more! Therefore, although recoding is imperative, it cannot be regarded as a standard based on a few subjectively selected words.In case a dear friend suddenly becomes a news figure, but his name is not coded, you can't call him XXX!Thinking a step further, one day we will use computers to sort out our inherent culture, but we find that more than half of the ancient characters are not recognized by computers. What should we do then?Is it because our culture is too ignorant? In addition, rote memorization is absolutely unworkable. At present, the professional training for 8,000-character telegrams is half a year. Not only is half a year a long time, but few people can bear the hard work of rote memorization.If 40,000 characters are continuously encoded, God!To learn it is probably not five times the plain code of the telegram, but fifty times! Some experts believe that at least there should be a set of exchange codes, which need not be memorized by humans, but only need to be stored in the computer memory. No matter what codes each computer uses, with the exchange code as the standard, they can communicate with each other and exchange information. This is also one of the strange phenomena of Chinese information. They are in the same national system and use the same language, but they need to use exchange codes. You may or may not have heard of this term.If we use an example to illustrate, you may also be dumbfounded. Let's assume that each computer represents a region. Foreign regions are very simple and all use English, but our Greater China nation has a particularly strong regional concept.Therefore, each region must use a language, that is to say, the people in each region cannot communicate in a language that can be understood by each other. This is the current Chinese computer world, and each has its specific input method.The input data cannot be shared by anyone except the same type of computer, which is a waste of time.The same work and the same data have to be done once, or even countless times, on different systems. Today's country has limited human and financial resources. How can this continue?Therefore, there has been the discussion of exchanging codes. But does the exchange code work?The National Science Council tried hard, but it was aborted because a certain person who thought he was superior insisted on using his method to rule the world.In fact, the exchange code is just to arrange the commonly used characters in a certain order.It cannot be used as an input code, and it also increases the space in the computer.However, if there is it, at least until the input method is not unified, it is a feasible temporary channel.However, it must have a prerequisite, that is, it must have the generosity to cover all Chinese words. It is called an exchange code just by copying 16,000 words from the dictionary. Please think about it, aren’t the other 30,000 words Chinese characters?We don't know because we don't know much, and we don't need it because we don't need it. However, how can the exchange code representing the country only consider the current needs and categorically deny all traditional characters because of the temporary lack of ability? Of course, our survival today cannot be ignored, but isn't there a better way?If everyone is willing to put aside their prejudices, disregard fame and fortune, discuss and research together openly, and be fair and comfortable, I believe there will be a reasonable result! In any case, I decided to use the natural coding method. The advantage of taking five codes is that it has more than ten million permutations and combinations, and any Chinese can be accommodated.Some people think that five codes are too wasteful. I disagree. I have calculated that the average code fetch of 35,000 words is 4.1 codes, and English is 4.5 codes according to general statistics. Why does no one say that English is a waste?At the same time, there are many ways to save storage and transmission. Phrases and abbreviations are the most effective methods. Among the more than 10 million combinations, we can make full use of this function (if you have used my Cangjie input method , you will find that I actually only use 24 letters, and the other two are used to do this kind of work. I am still developing a phrase database, which has 24^4 phrase functions, but only occupies 64KB!) . According to my ideal, Chinese should be the same as English, and the input code, output code, internal code and transmission are all integrated, so that it can exert its maximum function, and it is also a good way to cure the root cause once and for all. These principles are determined, and there is only one thing left to do, which is to select the codes that can represent the initials and characters, and finally pick out the letters. This step is not difficult, especially now that I have a computer, which can save a lot of time and manpower, but at that time, after each letter was assumed, the work I did was as follows: encode and proofread the 10,000-word cards one by one, After editing, arrange the cards in alphabetical order, find out all the cards of the same code, study the reasons for the same code, and the method of avoiding the same code, and finally order a set of letters. Repeatedly, it can be sorted up to 1.5 times a month. It is not difficult to reduce the repeated words of the same code. The difficulty is that all letters must have a classification meaning, and it cannot be decided just to avoid repeated words.In addition to the fact that the letters must have meaning, the auxiliary glyphs must have an easily associative relationship with the letters, otherwise it will be difficult to learn, difficult to accept, and reduce its value. Some people think that the method of grouping repeated words is the most ideal method. If the repeated words can be completely avoided through very reasonable rules, that is of course good, but it is debatable to cut the foot and fit the shoes just to eliminate the repeated words.Taking Wan and Mian as an example, they are both day in philology, and the codes are the same, only by looking at the position, one is on the left and the other is on the top.It is obviously a waste to add a left button to all the days on the left side for this reason, and only add the same word with code fetching, so what is the difference with the definition of repeated words? In order to save space, there is no need to mention the process. I only compare the second generation and third generation I have printed and published with the list I am using now: The first generation of sun, moon, gold, wood, water, fire, and earth, with hands, feet, mouth, eyes and ears, Wang Shishan has insects, fish, dogs and horses, and his clothes speak of silk, grass and bamboo. The second generation of sun, moon, gold, wood, water, fire, and earth, people's mouth, ears, hands, and bamboo, The oblique points of the cross button are vertical and horizontal, and there are ten divinations in the twenty fields and mountains. The third generation of sun, moon, gold, wood, water, fire, and earth, people's hearts, hands, mouths, fields, grass and bamboo, Cross the vertical and horizontal buttons and point them obliquely, and the western mountain of clothes will be divined again. Now the sun, moon, metal, wood, water, fire, and earth, cross the vertical and horizontal hooks at oblique points, People's hands and mouths, sideways and upwards, the new square divination. It can be seen from the table above that, except for the sun, moon, gold, wood, water, fire, earth, and human heart, hand, and mouth, there were different choices for others.Simultaneously, from the second generation onwards, a new idea has been added, which is the stroke, and to the last generation, the stroke has been expanded to become a font. This transformation is obtained through statistical analysis, but it also happens to cooperate with the theory and becomes a very natural result. Theoretically speaking, characters are used by people to express thinking concepts through symbols. Human thinking concepts are firstly philosophical thoughts, and secondly life. Regarding these two points, we have sorted out the sun, moon, gold, wood, water, fire, and earth, and the heart, hand, and mouth as letters.On the other hand, Chinese uses strokes and shapes to express its differences. In the strokes, we chose oblique dots crossing the vertical and horizontal hooks. In terms of shape, we chose the side and upside-down squares, and the application is not satisfactory. The current 26 letters leave the X and Z keys for expansion. In my idea, there are three expansion methods: one is repeated code words, and I use the X key (on Acer’s Tianlong series, I use the Z key) ）來處理，目前我們重複字約1％，但若來者不拒全部收齊，可能會增至2％，一鍵可夠應付。其次為縮寫字，譬如說金屬氧化半導體這七個字在應用上極不方便，我們可以視為金氧半的縮寫，寫成金氣，簡單明瞭。金屬結晶寫成金晶也是順理成章又有何不可？ （當然我們不能越俎代庖，這只是舉例而已！） 前述這些字是我們的字形產生器第二代功能，它能組合二百至三百萬個怪字，但絕對符合取碼規則，這些字一點也不佔空間，用不用隨你。 第三種擴充的方式就必須增加一片64KB的模板，我們要預先選定所有的專有名詞，常用的片語等字數在二字以上的，多多益善，用最精簡的資料結構方式，我估計至少可以存24^4種，而在使用時當作一個單字用，只是首碼為Ｚ而已！ 這一來，中文的傳輸及儲存將是精簡無比！ 再談談字母各鍵的安排與人體工學的關係，由於所收的字很難有個標準，有人主張以最常用字（據新聞報紙用字統計，約有三千字），也有人主張以學生字典（約八千字）。不論一種，都有缺點！因為對象的不同，字彙的使用頻率就亙異。一種編碼方法，要用各種情況，就不能以偏概全，我認為面面俱到是不可能的，只要能符合大前提的要求，已經是難能可貴的了，我的前提是這樣的： 1、字母的分類牽涉到原理、記憶與使用，絕不能因為任何單獨的理由而令之支離破碎。在分類中，基本類有哲理及筆劃，次元的有人體及筆形。要變動只能在各類的字母中，前後調整。 2、要使字母朗朗上口，就必須顧及音韻，尤其在未來使用時的字母排序上，要充分利用字母的背誦性，才能順利地發揮其功能。舉例而言，如果你不會背九九乘法表，對做算術一定很不方便。同樣地，如果你不會背英文字母，查英文字典也會增加不少困擾。因此，中文字母一定要會背誦，要能背誦就必須顧及其音韻的抑揚頓挫。 3、在人體工學而言，食指用得最方便，其次為中指，最難控制的是小指。對左右手而言，絕大多數人慣用右手。因此，右手手指又比左手手指方便。在鍵盤位置而言，文字鍵有三排，中排最佳，因為打字時，係以中排定位，字鍵使用的頻率高低應以前述因素作通盤的考慮。 有了前面三個前提，再去根據組碼規則，選擇取碼的樣品，一一分析比較，其過程之艱辛，不足為外人道。更由於我選擇的樣品是康熙字典中的35,000字，而這35,000字的取捨也有待進一步的斟酌。雖然最後確定了字母的順序，我仍然覺得不夠理想。 為了測試字母順序與人體工學上的配合是否妥當，現在且用教育部頒訂的4,803字為準，編碼統計於後，以供參考。（註：由於編碼小有疏忽，多收了九個字，致成為4,812字，但無礙全部取碼的正確性，故未加更正。） 在表中，比較不符人體工學的有Ｎ鍵（右手食指下排）以及Ｋ、Ｌ兩鍵相反，這是基於前述第二個前提的考慮所致，Ｏ、Ｐ、Ｑ、Ｒ四鍵係分類的限制，其餘尚差強人意。 字母一碼二碼三碼四碼五碼小計使用頻率 日A 96137139121305232.70％ 月B 2202353362098510856.61％ 金C 1281161011351065863.03％ 木D 194191164143387303.77％ 水E 2677862124736043.12％ 火F 116293100111967163.70％ 土G 126173159158526683.45％ 竹H 3663474252163913937.21％ 戈I 2722832292198510885.63％ 十J 186247158133747984.13％ 大K 91142152180446093.15％ 中L 132196224161347473.86％ 一M 3083144152705713647.06％ 弓N 165186206119367123.68％ 人O 3133232551657811345.87％ 心P 124123177109375702.95％ 手Q 235836358244632.39％ 口R 26429922625611511606.00％ 尸S 156154170120396393.30％ 廿T 271177179121387864.06％ 山U 551741302631157373.81％ 女V 2337198133525873.03％ 田W 69132172106184972.57％ 重X 8724202371610.83％ 卜Y 33828922481269584.95％ 小計： 4812478745843734139819315 每字平均取碼數： 4.01 | 至此中文輸入的說明暫告一段落，否則我們的篇幅就不夠了。 我前面說過，輸出之假定是用槓桿打字頭，很明顯的其中技術困難重重，幾百種字鍵倒可以解決，如何使字鍵放大縮小，改變形狀，改變位置呢？ 我當初用的只是一種工作的方法｜各個擊破｜並不是說將未來的希望全寄托在打字的技術上。 我另外有一個構想，是用光學透視的方法，可以將各種字形經過反射透視鏡組合起來，用在打字機並不困難，後來發現採用縮底片，幾千個字都可以濃縮在一片底板上，再經輸入碼改換成座標位置，即可將字形輸出，技術上確無困難。事實上我也開始動手設計，但問題出在輸出之印字頭，必須用半導體技術來解決，也就是利用半導體將光轉化為熱，再用熱印在熱帶上，熱色帶受熱會將字熔在普通紙上。 研究本身是相當大的投資，我費盡心力希望做出便宜而經濟的打字機來，要達到這個目的，我研究出來還是不行。必須能工業化大量生產。我先做好實驗，再找人投資，一百個投資者有一百個人要求專利保障。因之，我拿熱字頭及熱色帶去申請專利保障，不幸被批駁了。理由是樣品做得不夠好，不夠成熟。 大家都知道熱印字頭目前是最便宜最快速且無噪音的印字方法，但是卻無法普遍被人接受。原因無他，只為了熱字頭必須用熱感紙。假如一台打字機只能用一種特殊紙，它的性能再好，用起來也不方便。因此熱色帶如果得到專利，它本身就是一件極有市埸的商品，它可以使熱打字頭取代目前大量被採用的衝擊式打字頭。 friends!我已心力交疲，我申請專利的目的是希望不被一些唯利是圖的商人壟斷，但是既然得不到專利，我寧願你們快想法做出來，至少不要落在外國人手中，再來賺我們的血汗錢！ 熱色帶的試驗很簡單，當然商業用品要考慮其配方及多方面的性能，但試驗其可行性卻簡單不過：用最薄的塑膠紙（最理想是鋁箔，是熱的良導體），大約0.01MM至0.02MM厚，上面塗一層易熔的色素如同腊，或者用奇異墨汁（那種易燃的），用時可將鉛字加熱，在背面輕輕一壓，字形就印在紙上了。 如果你有錢，希望你快先申請國外專利（你有了專利，不妨賣給我），國外的有了，國內或許有希望。不過你要注意，一般熱印字頭只有攝氏60度的瞬間溫度，你必須多方試驗，降低色素熔點。再不然就得提高熱印字頭的溫度，那就要用到半導體的技術了。 再談中文字形吧！由於前述專利受阻，無人敢投資，我又無力生產，因此，中文打字機一直無問世。幸而微電腦適時大行其道，我便將全力放在這方面。 英文的字形出現在電腦上，是利用座標點的掃描完成的。一般說來，電腦顯示器上有512*256點，或者是小一點的256*192點，再大一點640*512點，這代表些什麼呢？我們叫它顯示密度，也就是指我們所看到的光點數。以512*256為例，512點是指顯示幕上從左到右可以顯示512個光點，相當於有512個小燈泡，可以控制其明暗。256則是指上下共有256排燈，換句話說，就是一共有256排，每排512個小燈的顯示幕。 那麼英文字怎樣控制呢？最普通的就是用5*7的燈陣來表示，我們再用前面用過的0與1的方式，寫一個Ｅ字，只是為了讓你看得清楚，我們用.代表0，用鬱字代表1，其結果是： 鬱鬱鬱鬱鬱 鬱.... 鬱.... 鬱鬱鬱鬱鬱 鬱.... 鬱.... 鬱鬱鬱鬱鬱 我相信你一定看得出這整體的視覺反應，是個英文字母Ｅ字，在顯示幕上，表示1的地方燈就會亮，0則不亮（當然也有反過來的做法）。 由於英文字母很簡單，5*7的點陣就能把明暗的特徵表達出來。其原理是，每一筆劃與另一筆劃之分辨，全賴二者之間的反差現象。明確地說，字形之辨認，端視對比色的排列訊號與我們視覺感應所產生的效果。英文字母之特徵在於不論橫向縱向，其對比差異不超過五次，以筆劃論則不超過三劃，因此，他們只要5*5的點陣就夠了。 你會問，能用5*5為何還要用5*7呢？那是由於美觀及避免誤認所致，由於英文字母中有斜線也有弧形，這些用單純的橫直點來表示不夠美觀。再加上還要用阿拉伯數字，其他符號等，太簡單易生誤會，5*7的變化多了七分之二，辨識性就理想得多。 當然，5*7並非最理想的，只是最經濟的而已，點陣密度愈大愈美觀，辨識性也愈強。現今美日高級的電腦，也已逐漸走高密度點陣的路了。尤其是日本，他們為了要與漢字共用，英文字母已用到11*24。 中文呢？中文該用多大的點陣？這又得分幾方面來說了。首先我們要知道，最經濟而又能分辨大多數中文的點陣，以多少最理想？其次，我們把全部中文收羅進來，又應該是多少？這樣我們才能做詳盡的分析。 最經濟的條件是要配合電腦的結構，我們前面說過，電腦的最基本辨識單位是字元，也就是八組的開關，在此我們將之正名為八個點位或位元。電腦顯示幕上的光點，是電子束在陰極射線管中，受到水平向及垂直向的控制後，所產生的座標位置。正常情況都是先水平掃描再一行行地向下移（也有將陰極射線管橫放，以致先見到垂直動作的），由於電腦以八個點位作為一單位，因之橫向最好用八的倍數，縱向雖比較不重要，但如能用八的倍數，則更方便。 why?由於電腦設計之初，以英文為主，英文只要5*7的點陣就可以了，而且英文及符號只要94個，全部放在一起，直接用點陣儲藏，其空間要多大呢？ 94*5*7是3290個點，而八點為一位元，那麼3290/8為412個位元，也就是相當於0.4千筆可以放到一片小小的積體電路器中。 這片積體電路器都是放在顯示器線路中的，一旦某一字母的訊號傳到，立刻即可將相關位置上的光點訊號送出，直接了當。中文不然，我們無法用英文的這種方法，原因是字太多了，顯示器本身是個附屬機構，一切作不了主，主系統那邊要遙控的話，也有一定限度。像中文儲存，一萬個字最起碼要320千筆，誰也控制不了。 因此，還得想別的辦法來處理，方法之一，也是目前最理想的方法，就是利用繪圖的功能，把中文當作一個圖形來處理，這一來，就必須是個能繪圖的電腦了。（注意市面上有很多電腦都號稱能繪圖，其中有很多是假的。這種稱為SEMI GRAPHIC，他們是先設計好一些圖形碼，視作英文符號之一，然後可以藉積木方式拼出圖形來，而實際上，並不是真正的圖形處理。） 有繪圖功能的電腦，在能力上就比普通電腦要強，當然成本就高，其中必須加一片螢幕記憶板（SCREEN MEMO）。它的功能是先要把圖形畫在其中，然後再用掃描方式傳到螢光幕上。 很顯然，把記憶中的圖像搬到螢幕上，就必須配合電腦傳遞的方式，電腦既然用位元為基礎，你用一個，兩個或三個位元都沒有多大問題。但是如果用1.3個或2.5個位元，那可就麻煩了。 舉個例子來說，你上學時總要揹書包吧！書包是根據書本的大小設計的，你今天如果設計一種書，當然要遷就書包大小，如果你設計的書比書包大，別人帶起來不方便，除非你不在乎別人用不用，否則不是你替別人改書包，就是你改書本的尺寸。 其次，我們還要考慮中文字的筆劃到底要用多少點才適合？我們只能這樣說：上下十六點可以分辨的字約佔80％，十八點可以辨識的字佔85％（上下可以不採八的倍數），廿四點可以分辨的字佔99％。左右八點能分辨的字佔30％，左右十六點可分辨的字佔93％，廿四點可分辨的佔99％（以上數據僅憑我們發展各種字體所得的實際經驗，尚有待學者們整理）。 點數愈多自然愈理想，但是成本相對提高，在我個人的考慮中，我選了16*16作為低成本的用途，24*24作為一般高級電腦之用，還有32*32，可供自動排版印刷用。 現在我就來解釋16*16的作法與相關的技術原理。 16*16實際上不能真正地令左右打十六點。各位知道，中文字形有很多是左右對稱的（這完全符合人的生理學，因為人的眼睛一左一右，形狀大小相若而方向相反，但上下則不然，所以你如果深入研究下去，會發現美感之產生與生理息息相關。），我們以一個十字為例，橫高一點、低一點影響不太大，但直卻一定要在中間。 如果直在中間，我們且假定直佔了橫向中間一點，左邊的橫再假定為七點，右邊亦然，總共是十五點，還剩下一點怎麼辦？加在右邊？左邊？或者把直加粗？ 單線體是不允許任何一劃比別的筆劃粗的。因此，我們為了左右對稱，只能用十五點，至於上下，我們仍用十六行，因為橫在正中間反而不好看。一般說來，在中文的美學觀念上，上密下疏較為好看，你看看人的面孔就知道，如果眼睛眉毛橫在臉的正中間，是個什麼樣兒？ 市面上有幾種電腦字體，左右還是用十六點，那些字是美術專家寫的，一個一個地寫的，一筆一劃都很講究。但是你看了會不知為何，有些怪怪的感覺，說穿了，就是因為左右不對稱，你不習慣！ 一行多出了一點，十六行就十六點，這不僅不是浪費，卻正是我能用最小空間，收容幾萬個中文字形的最大秘密。把這一點學去了，我這八年心血也就變成你的了。 在武俠小說中，常有什麼武術秘笈的爭奪，無論成名多年的大俠，獨霸一方綠林，以及後生小輩，莫不摶命以求。我年輕時也做過夢，滿以為得到了一個訣竅，立刻可以學貫古今，其實那真是夢，而且是癡人說夢！訣竅是有的，懂了可以節省很多無謂的重複浪費。但是，不去下苦功，把這個訣竅與其他的相關知識融會貫通，那麼訣竅還是訣竅，它不會自已變成學問的。 因此，你如果真有心，我希望你看了下面的介紹後，還要多下功夫，進一步追求各種相關的知識，功夫下得愈深，你所得的就愈多，才能靈活運用。 前面介紹過排列組合的概念，不過還不十分清楚，我們再簡單解釋一下：當我們設定每行有十六行時，我們就有16*16個點，也就是256點。換句話說，我們有了256個位置。由於我們要在這256個位置中畫圖、要寫中文，我們就必須賦與每個位置一個名稱，以便處理。 命名的方法很多，顯然在這裏用數字最適合，而且要用座標的方式先取Ｘ軸的點位，再取Ｙ軸的點位，我們可以順次命名為第一行第一點，第一行第二點。 再把文字省略掉，我們可以寫11、12、13，不過你看得出來超過10的點位就變成110、111、112，我們無法知道11是指Ｘ軸或Ｙ軸，更好的辦法是，統一取二位數，如0101、0102、01030111、01120201、0202。 這是十進位的寫法，對電腦而言，十進位非常浪費，一個位置要四個字元來表示！ 其實，有種方法，在16*16的點陣中，每個位置剛剛好只要一個字元就可以表示，要節省就要充分利用其有利的特徵，這正是我們採用16*16的最大理由！ 這種方法是二進位的延伸，叫做十六進位，是瞭解電腦最重要的一步。因此，我不厭其煩地把十進位，二進位及十六進位的對照表列在後面，以供比較及參考： 十進位二進位十六進位 0000000 0100011 0200102 0300113 0401004 0501015 0601106 0701117 0810008 0910019 101010A 111011B 121100C 131101D 141110E 151111F 你不難發現，在0～15，二進位只佔四個位置，或稱四位元，十六進位是一個數字，而十進位則要兩個數字。（實際上，十進位通常要佔一個位元即八個位置來表示，這又牽涉到其他枝節，這裏恕我不多做介紹。） 電腦雖然是用二進位的開關方式操作運算的，但是一大堆的0與1，由人去辨識很容易混淆（記得我們前面提到的中文與英文字母之比較吧？變化太少不見得適合人類複雜的頭腦），所以人就把十六進位給自已用，而讓電腦它們去玩0與1的開關把戲！ 各位也看出來了，前面我們說的16點，正好可以用0到F來表示，16行也用0到F表示，那麼我們就可以寫成01、02、030F、FE、FF。每對數字一個位置，同時由於每個16進位數字只佔四點，所以每對數字正好八點。換句話說：16*16的點陣上，每個位置正好以一個位元來表示。這時，一切都明朗了，我們若要劃一橫，可以指定由某個位置到某個位置，直也可以，斜也可以，只是要注意：斜只能畫對角線，否則不好看，也畫不出來。（有種畫法是把斜分成幾段橫或直，見仁見智，但我沒有這樣做。） 你不妨用下面的附圖，按其中的位置寫出你要寫的字來，然後把座標值讀出，就是電腦字！ ┌｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┬｜┐ ︱00︱10︱20︱30︱40︱50︱60︱70︱80︱90︱A0︱B0︱C0︱D0︱E0︱F0︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱01︱11︱21︱31︱41︱51︱61︱71︱81︱91︱A1︱B1︱C1︱D1︱E1︱F1︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱02︱12︱22︱32︱42︱52︱62︱72︱82︱92︱A2︱B2︱C2︱D2︱E2︱F2︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱03︱13︱23︱33︱43︱53︱63︱73︱83︱93︱A3︱B3︱C3︱D3︱E3︱F3︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱04︱14︱24︱34︱44︱54︱64︱74︱84︱94︱A4︱B4︱C4︱D4︱E4︱F4︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱05︱15︱25︱35︱45︱55︱65︱75︱85︱95︱A5︱B5︱C5︱D5︱E5︱F5︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱06︱16︱26︱36︱46︱56︱66︱76︱86︱96︱A6︱B6︱C6︱D6︱E6︱F6︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱07︱17︱27︱37︱47︱57︱67︱77︱87︱97︱A7︱B7︱C7︱D7︱E7︱F7︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱08︱18︱28︱38︱48︱58︱68︱78︱88︱98︱A8︱B8︱C8︱D8︱E8︱F8︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱09︱19︱29︱39︱49︱59︱69︱79︱89︱99︱A9︱B9︱C9︱D9︱E9︱F9︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱0A︱1A︱2A︱3A︱4A︱5A︱6A︱7A︱8A︱9A︱AA︱BA︱CA︱DA︱EA︱FA︱ ├｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┼｜┤ ︱0B︱1B︱2B︱3B︱4B︱5B