Naked Science Forum
General Science => General Science => Topic started by: remotemass on 07/03/2022 07:59:30
-
Maybe Deepmind would be ideal to design a new version of English optimized for the Semantic Web and for Google's aims of making the world more understandable.
What features should a modern new and simplified version of English should have?
So far I came up with the following features that that new version of English, let's call it "GoogleEnglish", should have:
1) Have only 9999 words.
2) The list of those 9999 words would be made comparing the rank of how frequent each word shows up in English and in the different most used oral languages.
3) Deepmind would analyze the frequency of words in English and a few other very popular similar languages with a somehow similar origin, alphabet, and also the morphology and phonology of those words to come up with the ideal non-ambiguous version of the new 9999 GoogleEnglish words.
4) There would be no ambiguities. "By" and "Buy", "Be" and "Bit", "To" and "Two" would be completely sorted out so that there were no ambiguities and words were all said exactly as they were written. Some languages, like Turkish, are particularly good with that in avoiding any ambiguities. The phonology of Turkish is the pronunciation of the Turkish language. There are no ambiguities. Words are read exactly as expected from their written form.
5) The phonology of the graphism of the alphabet would be made ideal, with as little modification from our nowadays English, as possible, but sorting out redundancies like the letter "C" and "K" being used pretty much with the same sound causing redundancy and undecidability/ambiguity.
6) Words with the same meaning would be avoided and for instance, you would distinguish between different types of vehicles using other words to specify them, like, "car" could be "Four Wheels Motor Vehicle" and bike "Two wheels Pedals Vehicle". For that matter, there should be a huge @Deepmind optimization so that for the size of words and the different commons expressions they would sound in a nice way and so that it would be quite easy and convenient mentioning them. A whole linguistics endeavor should be undertaken for such optimization.
7) Punctuation and accentuation should be minimal. Basically, you would use a full stop to separate sentences that should tend to be short. And that would be pretty much it.
8.) Since the average number of visible stars to the human naked eye is just about 9999 (a bit less, actually, about 9096), the visible stars would be mapped to the new GoogleEnglish dictionary (listed according to the rank of how frequent each word was expected to show up, on the web), according to how close that star is to us. So the Sun star, our closest star would be mapped to our first GoogleEnglsih word, and the most distant star visible to us with a naked eye would be named our last GoogleEnglish word. Also, then, to mention a cube you would only need to say the first 2 digits followed by five #GoogleEnglish words. (22 digits = 2 + (5 * 4)). See: CubicPostcode.com to better understand what do I mean by "cube". #CubicPostcode_com
9) GoogleEnglish should be designed so that it would probably not need to be modified any soon.
(https://i.imgur.com/8ROP4SB.jpg)
-
A language that followed those rules wouldn't be English.
Why have you mislabeled the pictures which clearly illustrate "shape", "card game", "tumble-dry only", clock", "prayer" and "safety"?
The list of those 9999 words would be made comparing the rank of how frequent each word shows up in English and in the different most used oral languages.
The commonest word in spoken English is "er".
So the Sun star, our closest star would be mapped to our first GoogleEnglsih word
So you plan to rename the Sun "Er".
Obviously, the most widely spoken language might not be English. Perhaps you should propose messing up Spanish.
I'm still not not sure if that would be any more sensible than your unworkable "CubicPostcode",
-
Right, it would be called GoogleEnglish. The idea is that it would be resembling English a lot but much better since it would be optimized taking into account the frequency, morphology, and phonetics of words in English and also in other popular and similar languages with the same kind of origins and alphabeth. The idea is that anyone fluent in English would be able to get fluent in it in a matter of days. The image is just to show how flashcards (that would be only 9999) could help bridge the transition and foster very quick learning, namely for new learners of such a language.
Please don't take me wrong or be too offended but I am pretty sure your grumpy cynicism could possibly be monetized in a very successful way ;-)
Yep. Our Star "The Sun" would probably be called simply "The", and the last visible star to the average naked eye: "End".
-
There are about 170,000 words in English.
Do you really think we don't need 94% of them?
alphabeth
Is that a new word?
Would it be on the list of 9999?
Please don't take me wrong or be too offended but I am pretty sure your grumpy cynicism could possibly be monetized in a very successful way ;-)
I am, as it happens, essentially paid to explain how things go wrong.
I bought my house on the strength of that monetization.
In your new English would monetization mean turning things into the painting style of the French Impressionists, or would that require a capital M?
Anyway, It seems the project already began.
https://xkcd.com/1133/
-
Knowing 4,000 to 10,000 words makes people advanced language users while knowing more than 10,000 words puts them at the fluent or native-speaker levels.
Yes, English has the vastest vocabulary corpus of all languages which makes it sweet for literature and poetry.
But for efficient communication along the lines of simplicity 9999 would certainly be enough. Many words are just slight variations of the same concept and are poorly designed to focus on the meaning specificities that the semantic web needs to address.
Sorry, for the typos that my spell corrector may not highlight.
I am more into cubism than into impressionism. But I like Monet, indeed ;-)
-
One reason English has become the dominant language for science, medicine and business is because it is not a "regulated" language like French or Latin. Words can be borrowed from friendly cultures or constructed from mixed roots or acronyms as convenient, without having to be submitted to a committee of linguofascists (how's that for a good one?) and those that enter common parlance eventually make their way into the Oxford Dictionary (or Webster's if you prefer American spellings) to more or less define their meaning until the next edition.
Basic vocabularies already exist and have some mandatory authority. In order to fly internationally, a pilot needs to learn about 1000 English words. Pretty much the same for a ship's captain. If the military intercept your passage, they must hail and direct you with a vocabulary of fewer than 100 English-ish words, 36 of which are the phonetic alphabet and numbers. That pretty well defines a safe minimum.
At the other end of the scale, basic nurse training involves learning about 3000 words that aren't used outside the medical context, and specialists can add another 5 - 8000 to the "fluent-native" vocabulary. The great joy of computers is that there is no realistic limit to their vocabulary, and the great joy of science and medicine is that even when the principal language of a conference is not English, practically everyone understands the technical terms which, for simplicity, are just added to English as the subject develops.
Attempts to reform and systematise English grammar, accidence, pronunciation and spelling have all failed in the past because the language grows and evolves faster than any committee can keep up. Part of the robustness of English is that the sense of most words is c*rr**d in the c*ns*n*nts rather than **e *o*e**, so accent doesn't obscure meaning as much as in, say, French.
-
Hey, that's my first daughter's name. Don't make fun of it.
Next thing you know you'll make fun of Betabeth as well.
They should meet the Mexican firefighter's sons, Jose and HoseB
-
If I were going to rationalise language I'd start with the alphabet first, with one letter for each sound, and one sound for each letter.
There are about 170,000 words in English.
You can multiply that by 347.
"59 million words of the OED second edition" (https://en.wikipedia.org/wiki/Oxford_English_Dictionary#:~:text=Diagram%20of%20the%20types%20of%20English%20vocabulary%20included,them%2C%20and%20540%20megabytes%20to%20store%20them%20electronically.)
-
You can multiply that by 347.
"59 million words of the OED second edition"
I could, but that total includes the definitions etc, so it's a bit irrelevant.
On the other hand, they do say:
"As of 30 November 2005, the Oxford English Dictionary contained approximately 301,100 main entries."
So, maybe I should multiply by about 2.
If I were going to rationalise language I'd start with the alphabet first, with one letter for each sound, and one sound for each letter.
OK.
https://en.wikipedia.org/wiki/International_Phonetic_Alphabet
That's stage 1 done.
Now get people to use it.
...
I will wait.
-
If what I am proposing here did not have its allures then the language Esperanto that was designed from scratch and that has only 600 root words that add up to about 2500 common words wouldn't have appealed so much to so many generations of idealists and dreamers. What I am proposing here is using modern computation systems like Deepmind and a whole linguistics endeavor of dedicated experts to bring about something like Esperanto that resembles a lot English and is super easy to learn to English speakers, creating a language that tends to simplify and uniformize the thought processes in an ideal way for an efficient universal communication designed for the semantic web and everlasting future. You could still use scientific and technical terms in italic, I suppose. But at least it would make you think if that is really what you wanted to use or would rather keep simple, clear, very understandable, and widely universal.
"Take a strange language like Englilsh, for example, that has well over 500,000 words - most of which are useless for the average speaker and well unknown to the average speaker. Many of the words in English are technical terms, dialect, slang or obsolete. Then there are the overlaps as like breakfast, lunch, dinner, supper, etc which are all types of meals but which have to be learned separately, the first concerning a morning meal, the second a meal around noon and the third an evening meal whereas the fourth may be substituted for the third according to the dialect spoken. Do you mean morphemes? Or do you mean compound words (like textbook)? If textbook counts as one word, does then lernolibro in Esperanto count as one word? Or do you mean just text and book with the possibility of creating textbook though not text citation? Counting words is a little like counting clouds.",
Quoted from a post at:
https://lernu.net/forumo/temo/18161 (https://lernu.net/forumo/temo/18161)
-
Hi.
The main argument against standardising languages and using things like Esperanto has always been the destruction of culture and art.
Different regions have different languages and this helps to maintain some unified culture in the region while putting some nominal barrier up against a different region. I'm not saying this is a good thing, it isn't always a good thing, it's just a thing but a thing which produces and maintains a diversity of cultures. So there are some aspects which are considered to be good.
For example, I have a friend who loves France. However, they wouldn't find France all that interesting if all of their culture started shifting to become almost exactly the same as, say for example, New York in America.
There is also the effect on art when you broaden art to include such things as poetry and creative writing. The writings of great poets probably degenerate into utter rubbish if you translate it into Esperanto.
Obviously there is some value in developing a language that is ideal for a modern world with the internet but you have to ask if the price is worth paying.
Best Wishes.
-
If I were going to rationalise language I'd start with the alphabet first, with one letter for each sound, and one sound for each letter.
Try German, Spanish or Italian, then. Rational and consistent spelling and pronunciation, and more popular than Esperanto. Or Welsh, perhaps.
The strength of English is its wanton origins and utter disregard for "purity". The weakness is Caxton's historic spelling.
Apropos phonetic alphabets. the International Civil Aviation Organisation has been deliberating for many years over the letter N. All the others have one or two syllables (OK, Uniform is two and a half) but nobody has come up with an acceptable substitute for November.
-
Speaking of phonetic alphabets...
A is for Aisle
B is for Bdellin.
C is for Cnidaria.
"What features should a modern new and simplified version of English should have?"
First of all it should have purpose, and I haven't seen that yet.
Once Google translate can handle more than 9999 words of English, your idea is pretty much pointless.
It's also based on a false premise.
Then there are the overlaps as like breakfast, lunch, dinner, supper, etc which are all types of meals but which have to be learned separately,
Because they are different.
If I ask a woman in the pub if she would care to join me for lunch, it doesn't have the same meaning as if I ask her if she would like to join me for breakfast.
something like Esperanto that resembles a lot English and is super easy to learn to English speakers
This may surprise you, but that group can already communicate without using a "new" language.
-
Counting words is a little like counting clouds.
Then don't do it.
1) Have only 9999 words.
-
Apocryphal, but a good one. There used to be an advertising hoarding above the cliffs at Dover
"Ovaltine - pronounced success"
that proving that Britain cannot be understood, let alone defeated.
Or G B Shaw's "ghoti". gh (as in laugh) + o (as in women) + ti (as in nation) = "fish".
But here's a true one. The first automatic railroad crossings had a sign "do not cross while lights are flashing". In Yorkshire dialect, "while" means "until". And there was indeed a disaster.
If it ain't busted, don't fix it. English is organic and illogical, but it ain't busted.
-
I enjoy "The history of English" podcast. It illustrates the many and varied ways that the English language has been distorted and corrupted over the centuries. Every episode I discover more reasons for its illogical and confused current state.
https://historyofenglishpodcast.com/
I feel sorry for the people around the world who are trying to learn English as adults. Learning it as a class in school would be a lot more successful. Even easier, countries with a smallish population of native speakers (eg Norway or Holland) don't often translate English films into the local language, but do add subtitles. Young children who can't yet read are then exposed to the sounds and idioms of English without a direct explanation; that trains their brain to identify the sounds and patterns of the language, which would greatly help with formal classes when they attend school.
Up Goer Five
...is a translation of "Saturn 5 rocket", from a book called "Thing Explainer", which took on the arduous task of explaining modern technology like space rockets and atomic bombs using a vocabulary of the most common 1,000 words in English.
- He The author of this book (Randall Munroe) did cheat a bit, as he didn't consider adding an "s" for a plural, or "ed" for past tense as separate words.
- And he didn't count "er" as the most common word
-
- And he didn't count "er" as the most common word
Which "he" do you mean?
I pointed out that
The commonest word in spoken English is "er".
though I'm happy to accept that "Um" might win the race.
Neither is a glorious name for our local star.
https://en.wikipedia.org/wiki/Filler_(linguistics)#In_English
-
the many and varied ways that the English language has been distorted and corrupted over the centuries.
Pedantically, this presumes there ever was a "pure" language identifiable as English. Before the Roman invasion I guess most of the inhabitants of the big flat bit of these islands spoke something like modern Gaelic or Welsh. The Norse/Saxon/Norman invasions added layers and Chaucer points out that even by the 14th century there were so many distinct dialects that folk from 50 miles away couldn't understand one another.
Uncritical evolution and accretion, certainly, but not distortion and corruption - there was no basic structure to distort.
-
It illustrates the many and varied ways that the English language has been distorted and corrupted over the centuries.
Do you mean " the many and varied ways that the English language has been distorted and corrupted improved over the centuries."
-
improved
One aspect of the language is its spelling. And English spelling is a real hodge-podge.
For example Greek and Middle English had a perfectly usable f/φ letter, but I seem to recall that Latin didn't.
So Greek words that came into English via Latin are often spelled with "ph" instead of "f".
- And some scholars actually encouraged complications like this, as it showed the etymology of the word.
-
And some scholars actually encouraged complications like this, as it showed the etymology of the word.
And who are you to say that they were wrong?
-
And English spelling is a real hodge-podge.
This is partly ascribed to Caxton's attempts to capture the spoken word and assemble some consensus from written papers, and partly the result of English promiscuity in taking words from other languages. It is possibl too capchur the sounds ov Inglish bei adopting Germanik speling rools but in dooing soe yoo woud looz the etimologi ov Nourman wirdz. And yoo woud probabli wont to yooz an umlaut okaeznli.