4chan archive /sci/ (index)
2015-01-09 02:57 6998083 Anonymous Encyclopedia of all human knowledge (universe.jpg 1000x1000 522kB)
Dear /sci/, When you want to learn a certain subject, and look for the "best textbook", you will always find several. They are written in different style, with different ideas of what's didactically sensible, and the union of their knowledge is usually larger than the knowledge of any single one, meaning: If you're reading only one, you're missing out. But if you're reading several, you will waste your time on redundant information. Another problem is, that universities will usually regard books written by their profs as superior, just for the ego boost. However, there are some textbooks, that most people in the field agree on to be the most concise and rigorous like Baby Rudin or SICP. Unfortunately, they are incomplete, and don't fit neatly with textbooks on follow-up topics. What would it be like to have a complete, concise, rigorous, and constructive collection of all human knowledge? A collection that would have maximum regard for logical structure and no/little for didactics, and where the union of the knowledge of any two books would be empty? Yet something that you could learn from, if you were dedicated enough. The top experts and authors would collaborate to write the textbook for a single subject, and it would get updated. Every textbook would reference the other books, or chapters of other books, it builds upon, the whole would be constructive. Topics would include everything from maths, to science, and other topics that can be treated in a scientific manner, like chess, or even fishing. This gigantic knowledge base could also serve as a reference for: Books written with a more didactive focus and for university curricula. I know that making all these authors put their egos aside and work for a common purpose would be very hard, especially the softer the science.

4 min later 6998087 Anonymous (1417856997582s.jpg 122x125 4kB)
What's the point in wasting years (likely multiple decades) in attempting to construct a book that exists purely for the sake of convenience? All the information in the books is already available (for free usually) on the internet. Furthermore, the cost of constructing such a book would be unimaginable. Hiring experts from EVERY field and area of study in the entire world wouldn't just be a waste of time, it'd be a waste of money. Also, I think you are heavily underestimating the amount of knowledge out there.

9 min later 6998091 Anonymous
>>6998083 It could also be used in school, to make education modular and to stimulate talented children to advance more qucikly through topics. Books could be accompanied by didactic supplements in the shapes of websites or books, that would contain examples, graphs, and others. Those didactic supplements could be written by anyone in the community.

17 min later 6998100 Anonymous
>>6998087 >multiple decades I don't know what makes you think that. I would be completed in 2 years max. Books are written all the time, so the only cost would be coordinating the effort, which is almost nothing. The only hurdle would be to make the people collaborate. While human knowledge might be vast, it can be enumerated.

28 min later 6998122 Anonymous
Wikipedia.

31 min later 6998127 Anonymous
You're basically thinking of an Encyclopedia, but more systemized. >>6998100 > so the only cost would be coordinating the effort, which is almost nothing. I take it you've never worked in edition, or organized an event bigger than your birthday party. Just so you realize the magnitude of the task: currently mathematicians alone publish around 20,000 theorems a year (that's actually a figure from the past year, it's probably higher now). The body of results of medical research in China is believed to be already bigger than that of the Anglosphere, yet we have little access to it. Merely gathering enough experts to cover most fields of human knowledge would probably take years, and coordinating a handful of thousand of experts from various countries, while ensuring that knowledge is properly systematized (so that there is no redundance and every part refers t the previous) would be anightmare. You're handwaving coordination as if it were an afterthought, but that's actually the key issue here.

34 min later 6998130 Anonymous
>>6998083 What is more, it would make bibliographies of books much shorter and more sensible, since you would list fewer books, and no irrelevant ones. It could also be used as a reference for scientific papers and for any new book you write you could differentiate between whether or not this is already included in a certain version of the encyclopedia. Appart from university curricula, it could also be used in exams, thus making the grading system universal and you could make statements such as "Student A performed 76% in the chapter C1 in Book B1" which contains much more information than giving marks over one-semester subjects. It could also differentiate geniuses from just straight-A students, and help with employment. The end result of all of this would be that people would not just know subjects, but they would also know what they know.

37 min later 6998137 Anonymous
The information amount published in a single day is, even expressed concicesly, probably more than can be fit into a huge ass book. Now imagine not only summarizing that of an entire year but also for entire human history.

40 min later 6998142 Anonymous
>>6998130 You seem to believe that if things are stored in one encyclopedia for good, we don't need other books on the same subject. Nothing could be more false: the way things are ordered and explained in a book is essential in determining wether the book is useful. Furthermore, there are different kinds of books for different purposes (the vulgarisation to entice children, the long-winded explanatory for beginners, the book focused on proofs, the book focused on storing results...). That particualrly applies for scientific and technical books, where the point is to be able to teach students with various way to tackle problems. And that's not to mention there are many situation when two experts couldn't agree on what is the best presentation. Although such an encyclopedia would certainly help centralizing knowledge, it seems you're approaching the issue with a rather simplistic view of scientific writing. Note that something like what you described as already tried in mathematics, with only relative succes. Check out Bourbaki.

49 min later 6998156 Anonymous
>>6998127 >mathematicians alone publish around 20,000 theorems a year And there are around 150,000 mathematcians. If you have 50 books on higher-maths topics, and you get 10 volunteers to write on each, you will have 10 people working through 400 theorems that are from the very field that those people are experts on, which means: They would work through those theorems irregardless of the encyclopedia! So there is almost no additional effort, apart from communicating with the colleagues and writing the stuff down - and there are plenty of people who will gladly do this. And should the situation arise that you have more theorems for a subject than are manageable, you can just increase the number of partitions.The only assumption here is that there is always less knowledge produced that can be consumed - a matter of course in our current world.

52 min later 6998158 Anonymous
>>6998142 >You seem to believe that if things are stored in one encyclopedia for good, we don't need other books on the same subject. I said the exact opposite actually. Read the OP again. >>6998137 see >>6998156 There are no theoretical limits on this endeavor.

1 hours later 6998179 Anonymous
>>6998083 Just take the philosophy route: "YU CANT KNO NUFFIN". Would only take up part of a sticky note.

1 hours later 6998206 Anonymous
>>6998083 There already is. It's called wikipedia. Don't like how it's laid out? Change it yourself

1 hours later 6998217 Anonymous (1418929106754.gif 255x181 1018kB)
>>6998179 Had a giggle, then I cried

1 hours later 6998229 Anonymous
>>6998130 Just to further elaborate on the following thought and it's important for interdisciplinary and autodidactic learning: >Apart from university curricula, it could also be used in exams, thus making the grading system universal Let's say you studied MechEng, but you liked chemistry, not all of it, just organic chemistry. You could read a book of the encyclopedia, or a book of equal knowedge content, and then take an standardized exam without ever enrolling as a chem major or visitng a lecture. Because the exam would be recognized, you could list it among your qualifications when looking for a job as a MechEng for a chemistry company f.e. In the same manner you could gain a partial, recognized (!) MIT-level eduaction without ever paying a cent. I say partial, because for the full thing you'd need exercises, projects and internships too, but that would come with time.

1 hours later 6998231 Anonymous
>>6998158 I was talking about >>6998130 and the idea that an encyclopedia would shorten bibliographies by removing irrelevant books. Why that would happen to some extent, most of bibliographies are long because it's interesting to look at various take in a given subject (at least that's the case in maths). >There are no theoretical limits on this endeavor. But it's a practical endeavor. Again, the main problem is picking and coordinating the experts, and then agreeing on how to edit and present the whole thing. Your post is simply confirming that you're handwaving the main difficulty by saying this is "could be done" in 2 years. It's not only a matter of money, time and ego, but also of cultural differences and (importantly) knowing how to organize a vast body of indirectly related knowledge. You're going to have to put forth assumption about how the thing should be made at some point. But which one, and how to get everyone to agree ? You would also need people working full time on it in order to keep it updated, and not so many academic are willing to threaten their publishing count like that. >If you have 50 books on higher-maths topics, and you get 10 volunteers to write on each, you will have 10 people working through 400 theorems that are from the very field that those people are experts on, which means: They would work through those theorems irregardless of the encyclopedia! That's regardless, and you're making a confusion between writing a theorem within the body of an encyclpedy and publishing new papers and results about that theorem. Any week spent working on an encyclopedia is a week not spent adding new results to your résumé, and for many that matters a lot. And that's not even dealing with the problem of how to organize the whole thing in the first place. I never said this was impossible, but it would certainly take more than 2 years and shitloads of money. TL;DR: You say this is possible, but you don't really say how. Think implementation.

2 hours later 6998298 Anonymous
>>6998231 >how to organize a vast body of indirectly related knowledge. All I'm talking about is a library, and that is quite simple to organize. I'm not saying we should build a knowledge base (in the AI sense of things) for all knowledge of mankind - that is going to be a huge task someday in the future. What I propose is simple: Spread the word and find volunteers and established authors on the many subjects. From there on you can decentralize everything, the maths people do their stuff regardless of the physicists and vice versa. Once this is completed, and the physicists want to reference some maths books, they will need to delete the redundant suff, rework it a little bit, and et voilà, the huge encyclopedia is done. You don't need to compare every book with every other, just the ones that are adjacent, because of common sense that you won't include an introduction to trigonometry in a QM books, especially if you're explicitly told not to. And to keep it updated, you just need the 10 people to put their heads together once a year and you're golden. Remember, the paritions don't ever change, they only get refined. >making a confusion between writing a theorem within the body of an encyclpedy and publishing new papers and results about that theorem. I'm not. I was just saying that even if all theorems ended up in the encyclopedia, it's feasible. While it's true that publishing new papers and writing for the encyclopedia are two different tasks, they do have intersections. Also, I don't expect Terrence Tao to write books, I only expect those people who would write books anyway (or some other volunteers) to write them together. So you see, it's simple, cheap, and awesome, and the only thing holding us back are our egos. Bourbaki was 80 years ago. Today we have the Internet and a quantification of knowledge is more relevant than ever before and this would be the first step towards it - think AI supercomputers.

2 hours later 6998334 Anonymous
>>6998083 >where the union of the knowledge of any two books would be empty Uhmmm... I think you mean intersection...

6 hours later 6998575 Anonymous
bamp

13 hours later 6999007 Anonymous
>>6998083 Good job, Anon. You just thought up what the World Wide Web was supposed to be.

21 hours later 6999787 Anonymous (23 KB.jpg 398x500 24kB)
>>6998122

21 hours later 6999809 Anonymous
>>6998127 >The body of results of medical research in China is believed to be already bigger than that of the Anglosphere, yet we have little access to it. and all of it shit, because smart chinese doctors and engineers gtfo of china as soon as they can afford to

21 hours later 6999871 Anonymous (1394685896621.png 441x421 180kB)
Sort of related, but are there any examples of textbooks that blend clever analogies/metaphors/intuitive things with more rigorous, harder stuff? Pic related. What a godly book.

23 hours later 7000217 Anonymous
As soon as the book is finished the half life of facts theorem will take effect, a lot of the information will therefore be incorrect after several years

26 hours later 7000485 Anonymous
>>6998083 Are you talking about wikipedia?

50 hours later 7002510 Anonymous
bamp

50 hours later 7002516 Anonymous
http://en.wikipedia.org/wiki/%C3%89 l%C3%A9ments_de_math%C3%A9matique

68 hours later 7004125 Anonymous
>>6998083 >most people in the field agree on to be the most concise and rigorous like Baby Rudin /Sci/ is not "most people In the field"

68 hours later 7004138 Anonymous
>>6998087 >for free lol

74 hours later 7004751 Anonymous
>>7004125 Almost every mathematician I've talked to IRL agrees.

3.841 0.126