Because digit numbers are not foreign to Chinese. As simple as that.
The majority of Chinese don't speak any western languages. A domain name of letters is no different from a random combination of letters to them, even if the name is composed of real words in English, Spanish, or French, etc. They are completely foreign. While a combination of digits is not, like phone numbers, zip codes, bus numbers, train numbers, etc. They deal with them everyday.
My parents don't speak any other languages other than Mandarin. [To be accurate, my mother spoke some Mongolian, and my father spoke a little Russian, when they were young. But they couldn't speak any of them now.] They have no problem in recognize and pronounce the 26 letter in English. After I have been living in American for so many years, they still can't spell my home address. They remember my phone numbers and my house numbers, though, but not the street names. Think about it. It is amazing. Because I find it is much easier to remember street names than phone numbers. Phone numbers are simply a random combination of digits, except the area code.
That makes complete sense. The number system is quite universal and global, writing systems are not, or at least not to the same extent.
I would be lost if you asked me to type in a Cyrillic URL (and Cyrillic is quite similar to Latin script, all things considered!), I wouldn’t even know where to start if you asked me to type something from an even more different writing system in.
I suppose English is on its way to gain some of the same universality as Arabic numbers, but it’s still a far way away from that.
Thank you for sharing that, that makes complete sense.
My parents also don't speak English. They speak Romanian, so when they see English they read it as if it is Romanian. Like if they see "house" they might pronounce it as "hoe-oo-seh" (just say every letter, and pronounce with as if it is from a Romanian alphabet). I guess I never thought what it might be like if it was a completely unpronounceable alphabet like Greek for example. At that point a string of digits would work better.
One reason why Chinese have less difficulty remembering and using these numbers than the western fellows might be the difference in pronunciation. In Chinese, all single digit number has the same structure: a consonant plus a vowel. For example, 7, in Chinese is pronounced as [ch-i:], while in English is pronounced as [ˈsɛvən]. Also, every digit takes the same time to pronounce. For example, a string "123456" is [yi:, er, san, si:, wu, liu]. That, in my experience, makes a long string of number easier to read out. And easier to read out means easier to memorize.
A side node: In China every kids in their elementary school if not kindergarten can recite the "table of multiplication". That is, they can remember the answer from 1 * 1 to 9 * 9. I doubt how many westerners can do so, due to the language difference.
This is accurate. I was educated in a bilingual environment since young and remember strings of numbers (credit card numbers, telephone numbers, identity cards) much easier in Mandarin than in English.
For example, this is a small part of the multiplication table:
English:
six times five equals thirty-five
six times six equals thirty-six
six times seven equals fourty-three
six times eight equals fourty-eight
vs
六五 三十五 (liu wu, san shi wu)
六六 三十六 (liu liu, san shi liu)
六七 四十二 (liu qi, si shi er)
六八 五十六 (liu ba, wu shi liu)
---
one times one equals one
two times two equals four
three times three equals nine
four times four equals sixteen
five times five equals twenty-five
vs
一一 得一 (yi yi, de yi) (de == "gets")
二二 得四 (er er, de si)
三三 得九 (san san, de jiu)
四四 十六 (si si, shi liu)
五五 二十五 (wu wu, er shi wu)
I left out the tones for simplicity.
After a while, the mandarin multiplication table becomes like a poem to recite, while the English counterpart doesn't quite have the same ring to it
Of course, we would only need to memorize (10-x) items for the x table, except 1x anything, which is trivial
so, know 2x2, 2x3, 2x4, 2x5, 2x6, 2x7, 2x8, 2x9
from 3, you can begin at 3x3, then 3x4, etc
to be more advanced you only need to know up to 5, and do the rest with math... 97? easy peasy. (5+4)(5+3) = 55 + 53 + 45 + 43 -> 25 + 15 + 20 + 12 = 72
Well, that's an example of the time-memory tradeoff, right? It takes a lot longer to calculate 9·7 that way than it does if you've just memorized it. The real question is where to draw the line with respect to diminishing returns on memorizing products for larger and larger numbers.
> where to draw the line with respect to diminishing returns on memorizing products for larger and larger numbers
Everything from 2 * 2 to 9 * 9 where the second number is greater or equal to the first, i.e. 36 combinations. But it would also be good to memorize sums from 3 + 2 to 9 + 8 where the second number is less than the first, i.e. another 28 combinations. So we'd need to memorize 64 combinations from 2,2 to 9,9 where we'd know whether the result is addition or multiplication from the ordering of the numbers.
all 9s..in multiplication take second muliplicand subtract one and that is the first digit of the answer..now take that first multiplicand and subtract that first digit of the answer and that is the second digit of the answer
you wrote 9 * 7 then proceeded to calculate 9 * 8. I find the 9s easy to remember. simply take the other number, reduce it by 1, then add the complementary that together sum to 9. for 9 * 7, this is 63,
This is Malcolm Gladwell nonsense and not backed up by the fact that Chinese-Americans who learn English first (e.g. Nobel laureate Steven Chu) do just as well as their Chinese counterparts.
But we're talking about an entire country in which most people don't speak English at all. If I have to tell my 55 year old aunt to go to www.hotmail.com then I'd have to spend half an hour describing what English letters to type.
I can imagine it's something like explaining how to change some obscure setting through a GUI interface over the phone, but significantly worse.
On one hand you have: "OK, do you see a button in the lower left? Click that. Now press "Options", then "Configure". What? You don't see those? Which button did you click? ..."
On the other hand you have: "OK, do you see the button that looks like a squiggly line? It's in the top-leftish corner, between the button that has a circle with a small line through it and a button that looks like a comb. Press that one three times..."
It's not about being good at math, just at memorizing numbers. I'm Chinese American and was considered to be "good-at-math" in grade school, but damned if I can remember long sequences of numbers. Maybe I should try thinking about them in Chinese.
Most people I know have several bank account and credit card numbers memorized. It's the same for me, and I also remember my phone number from when I was 4; but not the area code!
I think there is a cultural bias in the U.S. against remembering numbers, but that people are generally able to do so. Memorization tends to happen naturally through use: it's easier for me to remember a bank account number after making deposits over a few months than by purposefully memorizing it, for instance.
Memorizing times tables used to be common in UK schools. They stopped because it doesn't appear to help children understand what multiplication actually is.
It is probably going to come back because of our current minister of education's liking of memorization.
I find certain higher level concepts become much easier to grok when you've drilled/memorized the layer below. When I tutored calculus I found that most people rely struggled with the algebra, for example factoring and binomial expansion.
I think there is actually value in memorization. Even in the age of google.
My kids' school went away from drilling for times tables when my oldest two were in elementary, in favor of a new math program that would "help them understand what multiplication is". The result is they struggled with math ever after because the mechanics of solving more complex problems was constantly interrupted by not having the needed basics.
They since changed to a more balanced approach: some memorization, some conceptual.
Personally, I've come to the point that I think they should fully embrace technology and make math completely about the conceptual while letting tools generally solve the mechanical. There is no reason a seventh graders needs to be challenged in algebra because they don't remember their times tables.
Personally, I find them to be a distraction from the much more important issue of changing the cultural outlook on education, particularly regarding parental involvement and supervision. Trying to optimize pedagogy is analogous to curing a disease rather than preventing its occurrence in the first place.
I'm surprised there's no mention here of punycode and IDNs. Is there just not enough widespread support for IDNs (the article mentions a need for plugins, but I thought most browsers already support IDNs), or are numbers really that much easier to type and remember than short, meaningful domains in native characters?
As I understand, punycode is translated by the browser to a standard ASCII string, so there's no need for special support in the DNS system or other infrastructure other than a simple translation in the browser's URL handling. Seems like a pretty straightforward/simple solution.
While it doesn't mention punycode directly, it has paragraph referring to IDN:
Why don’t Chinese web addresses just use Mandarin characters? Because that’s a pain, too. The Internet Corporation for Assigned Names and Numbers (ICANN), which sets the rules for web addresses globally, has periodically hyped the expansion of domain names to include non-Latinate scripts, but Chinese web sites have yet to take full advantage. Some devices require a special plug-in to type in Chinese URLs, and even then it takes longer to type or write out characters than to input a few digits. Plus, for web sites that want to expand internationally but don’t want to alienate foreign audiences with unfamiliar characters, numbers are a decent compromise
Sounds like the core problem is just "Chinese web sites have yet to take full advantage".
I'm really wondering how many (or what percentage of) devices actually need "a special plug-in". Seems like it would be a very small - and shrinking - fraction, but I could be wrong.
If the user's device is (presumably) already set to use Mandarin (or any specific language), then I assume that would be the default input method, so why does that take significantly longer to use than numbers? I'm genuinely curious, never having experienced that use case myself as a native english speaker.
As for websites wanting to expand internationally and avoiding unfamiliar characters, that doesn't really make sense to me - if they want to expand internationally, I don't think the numeric domains are going to help much (they'd be just as cryptic to me as Mandarin characters), so why wouldn't they just register alternate domains in the target regions/languages as many already do?
>they'd be just as cryptic to me as Mandarin characters
Yes the meaning would be just as cryptic, but the act of typing and probably memorization would be much easier. As an exercise, try downloading an input method for a foreign character set and typing out a string that you see rendered in an image.
When typing Chinese characters, each character takes multiple key strokes. If each character is a numeral, it's one keystroke per character. It's like an acronym.
Interesting. From what I can tell though, the lack of support from those TLDs seems to be deliberate rather than due to any technical challenge (the domain you register ends up just looking like standard alphanumeric ASCII with some dashes). I may be wrong, but it probably requires more effort for a TLD to disallow IDNs as they'd have to implement a test to try and detect a punycode string within the ASCII.
The one technical challenge I suppose could be a concern would be optionally implementing some sort of test to detect domain registration attempts that use similar-looking characters to those in existing legitimate domains (e.g., registrations from scammers attempting phishing attacks with domains that look similar to other domains but use different extended characters).
Still, it feels like the lack of support from these TLDs is more a matter of policy than a technical challenge.
> The one technical challenge I suppose could be a concern would be optionally implementing some sort of test to detect domain registration attempts that use similar-looking characters to those in existing legitimate domains (e.g., registrations from scammers attempting phishing attacks with domains that look similar to other domains but use different extended characters).
Another thing to mention, for elder generations like my father, even if you can use Chinese character as URL, they don't know how to input it, because there's IME needed, nothing like native Americans learn how to input ASCII. Obviously, ASCII are printed on most keyboards.
ICQ is kind of an example of the numbers-website phenomenon in English: the letters themselves are meaningless, but when pronounced sound like "I seek you" (compare to the Alibaba example in the linked article).
One of the more popular OpenDomains I have is 17m.com - auite a few people ask to use it for a group buying site. This article helped me see the importance of numbers, but when I looked it up on google translate but could not understand what '17m' was supposed to mean. Can anyone tell me?
I actually created an email alias for myself that looks like 5553423123@hotmail.com (555-... is my real phone number). My logic is: whenever I need to say my email over the phone, it's much less error prone to tell a string of numbers followed by the common domain name than spelling my real (custom) name and domain name.
The phone has more to do in other ways with why digit urls in China than the article mentions. Most Chinese internet users who are visiting websites use their phone, and the 12-key digits pad is a lot easily to type into than the 30-key letters pad. The people who use netbars or home computers in China are playing online games, not visiting websites.
the point of using number in the domain is, a lot of Chinese people don't know letters, they don't know English words and they can't get "PinYin"(using letters to represent Chinese words) right. How do you make these people type urls in the browser? well, luckily, they all know numbers!
Are there really computer-literate Mandarin speakers in China who don't know pinyin? How do they input text? The only serious non-pinyin IMEs I'm aware of are the bopomofo ones for traditional characters in Taiwan.
Mandarin speakers in some parts of China speak their Mandarin differently, so can't remember the exact pinyin. E.g. around Wuhan people can't remember whether a word ends with -ang or -an, similarly with -eng and -en. As a foreigner, I can't remember nu and nv, similarly lu and lv. Beijingers have trouble with -r endings.
many older people like my dad, don't know pinyin well. he doesn't input texts, but he browses web sites. He uses a default home pages full of links to start exploring the web.
But why is this phenomenon restricted to China? I haven't observed it with, for example, Japanese sites, even though they too use a non-Latin alphabet.
There's a few things things that contribute to this:
Market powers in Japan chose early on to advertise their websites not through URLs but through search keywords.
They will literally ask you in advertisement to enter certain phrases in a search engine to let you find their web property.
Japanese internet usage is traditionally based around cellphones, which traditionally were centered around portals. This is changing with smartphones.
Japanese numbers don't lend themselves as easily as Chinese to these double meanings. Japanese words are quite long compared to Chinese ('wo', is Chinese for 'I', the Japanese equivelant is 'watashi')
> Japanese numbers don't lend themselves as easily as Chinese to these double meanings. Japanese words are quite long compared to Chinese ('wo', is Chinese for 'I', the Japanese equivelant is 'watashi')
I think this is false. Japanese is quite rich for this. It's not worth debating whether Chinese lends itself more to this than Japanese, but as a Chinese speaker (cantonese and mandarin) who speaks Japanese, I find that Japanese is more versatile than Chinese.
Here's why:
1. all the digits have multiple pronunciations.
2. despite your point that many japanese words are monosyllabic, japanese are notorious for shortening their words and maintaining the meaning and also there are the chinese pronunciations ("on reading" vs "kun reading") so there are often monosyllabic ways to pronounce it.
3. additionally, for a given word, there are often tons of different (seemingly sensible) ways to pronounce it. This is confusing to everyone, including japanese nationals. they basically are just really good at figuring out meanings. I often have no idea what the correct pronunciation of written text is, but I know exactly what it means. This is true for japanese nationals as well. here's a great video of a game where japanese people try to guess how to pronounce kanji: https://www.youtube.com/watch?v=sY63F5wIPzc
Ultimately, the language's ability to be able to be used in the described way depends on the ability to map the pronunciations of digits to other meaningful words. Having a smaller set of sounds leads to more overlap.
Japanese and Chinese both have the property of having single syllable digit names tons of single syllable words in the language to map to. (even though arguably in Japanese "ichi" [1] or "nana" [7] are 2 syllables, "i"" or "na" would be acceptable -- also, in Japanese "shi" and "yon" are both acceptable single syllable ways to say the number 4)
Basically either is ripe for double entendre (not sexual, though possibly that too... for example, in cantonese, you can pretty much say any number to mean penis, though usually with the wrong tone)
(edit: I just wrote a longer detailed explanation of this with examples of neumonics for square roots one comment level above this.)
> They will literally ask you in advertisement to enter certain phrases in a search engine to let you find their web property.
I have noticed this on occasion, and found it very surprising. They're basically handing control over access to their site to the search engine. That seems like a very bad idea.
Well, in the case of Facebook, the audience is actually spending their time on your Facebook page. That's different from if they're just clicking through a search engine to reach your own website.
Which is kind of the point. Unless people know which URL to type (Disney, New York Times) in other words unless you have an offline brand, you don't have an online brand - you have a relationship with a supplier of customers.
For the phenomenon of number string in email address, that's because they don't use QQ, an IM service from Tencent. For other numbers in domains, I think it's because number in Chinese are easier to pronounce, and easier to remember.
Japanese does have the phenomenon of being able to ascribe meaning to digits though, even though it's not common to do so in URLs.
In school in japan, kids often learn tricks for memorizing multiplication tables and square roots this way, similar to how US high school students have a song for the roots of a quadratic (-b /- root(b^2-4ac... to the tune of Pop Goes The Weasel).
For example:
sqrt 2 = 1.41421356 (night after night, I watch/take care of someone)
sqrt 5 == 2.236069 (a parrot sings at Mount Fuji)
sqrt 6 == 2.44949 (cook the meat well)
sqrt 7 == 2.64575 (there are no bugs in the cabbage)
sqrt 8 == 2.828 (smile)
Breaking down the first one as an example, sqrt(2) : 一夜一夜に人見頃 (hitoyohitoyonihitomigoro)
hito (一) = 1, hitotsu -- the counter term for nights, like "yi ge" in chinese
yo (夜) = 4, this is the chinese reading of "night/evening". this character is pronounced "ye`" in mandarin. in japanese, in common speech you would say "yoru" for night.
[then repeat 一夜 -- actually sounds poetic and makes it mean "nightly" or "every night"]
ni (に) = 2, the same sound also is a particle indicating place or time
hito (人) = 1, pronounced the same as 1, but the character 人 is a homophone for person. person could also be pronounce jin (chinese reading, close to "ren" in mandarin) which is how it would be pronounced if combined with other chinese characters, like "nihonjin".
mi (見) = 3, in japanese, if you recite the numbers from 1 to 10 you would say "san" for 3, but to count items, you would say "mittsu". for example, the company "mitsubishi" has the number 3 (三) as the first character of it's name and it is pronounced "mi" like in this context. mi could mean a ton of other things too, like "ear" (耳) or "body" or "flesh" (身). This digit has a lot of flexibility.
goro (頃) = 56, 5==go, 6==roku. goro is unambiguously understood to be 56. and "koro" or "goro" describes the duration during which something is occuring -- basically "time".
To me this is very similar to the neumonic that US elementary school kids use to learn the planets: "My very educated mother just served us nine pies".
And just like you can use an alternate sentence like "My Very Energetic Mother Jumps Skateboards Under Nana's Patio", Japanese syllables are rich enough that you could construct totally different phrases to walk through those digits.
Japanese is arguably richer than Chinese for this because there are multiple pronunciations for each digit. 2 can be "fu" (like in Fuji above) or "ni" (a preposition like "in" in root 7).
> -b /- root(b^2-4ac to the tune of Pop Goes the Weasel
Woah. In the UK the most we got was SOHCAHTOA: "Smiles of happiness come after having tankards of ale", or "Some old hag cracked all her teeth on apples".
Or chose the heights of new buildings. Tokyo Skytree was built 634 meters high, because that number can be read "Musashi", the name for the ancient province of what is roughly present day Tokyo.
This article even mentions Hindi-Arabic numbers, which is an interesting thing to me as a student of the Arabic language
My teacher explained to us once, the numbers we use are called Arabic numbers, but Arabs don't actually use them... they use Hindi numbers. ١٢٣٤٥٦٧٨٩٠
Which is twice as funny as Arabs not actually using Arabic numbers, because (he said) especially Muslims are very prejudiced against Hindi people and Hindu religion. He would say things like "silly Hindu", then have a big belly laugh and get back to serious classwork.
This phenomenon is also visible in Korea, except the way to navigate to popular websites is to search Naver, Daum, or Google with the Korean name of the website and then hit the top link. I found it very strange at first but completely understandable given they typical just know the name of websites via their names translated into Korean letters.
This is not a good comparison because Cyrillic (at least, the slavic version) has almost a one-to-one relation to its latin counterparts. Russian Cyrillic is a little more complex, with letter modifiers (like "soft sign" and "hard sign") while in other flavors like Serbian Cyrillics, those characteristics were built into the language.
Hotmail.com directly transliterated to Cyrillic would be: хотмаил.цом (h=х, i=и, l=л, c=ц and the rest are the same).
The point is that there is a direct transliteration available between cyrillic and latin, and it's also not so culture-dependent as 617.cn is. You could trivially automate that process too.
> In 2012, the United States refused to sign an international telecommunications treaty, supported by both Russia and China, that would shift the Internet away from its current U.S.-centric form of governance.
As the linked article expounds upon, this statement is strictly true but does not give a very good picture of the situation. It was rejected on free speech grounds, and not just by the US.
Great article - I bookmarked it - when our startup will go online I will buy associated number domain and redirect all Chinese traffic to it. I think that's a great idea!
Until reading this article I assumed every 163.com email I saw was just using one of those temporary email inboxes, for fear of spam emails or due to privacy concerns. I, too, dismissed those emails and considered them as useless as seeing blah@blah.com. Now I know they're actual accounts, just based in China.
True, but most so do most email clients, and this still would be breaking her rule of "refus[ing] to communicate with anyone whose email address consisted of a string of numbers".
I often see many rants on HN about US government and how it is becoming a police state. However I think the fact that internet is US centric is better than the internet where countries like Russia or China have have greater say.
Yao (pin yin first tone) is a common way to say "one" in northern china when reciting multiple digits and basically the only way Beijing locals say 1 in a phone number or a string of numbers. for going to a market and asking for 1 of something, "yi" is the preferred way.
Pretty uncommon among southern Chinese people and Taiwanese (many American mandarin speakers are of Taiwanese descent).
chinese speakers like yao often because it sounds LESS like other numbers and is harder to miss.
yao1 (1) and liu4 (6) should be pretty distinct. but if the way you pronounce them is similar for whatever reason, you can always use yi and chinese listeners won't flinch. in fact, taiwanese americans will be confused by yao, but zero people will be confused by yi so it's arguably better to use that if you don't know the background of your counterparty.
Perhaps I'm mishearing but it sounds to me more like they are pronouncing it "yo" (English) or "you" (Chinese) here in Beijing, which does sound more like 6. yi/qi are slightly similar but I never have the same issue. I've given up on trying to be correct and use yi most of the time.
if there's question about pronunciation, i find that putonghua (mandarin) educational material is very good and there are lots of references with clear pictures of what your mouth and tongue should be doing when pronouncing that make it just about impossible to make the wrong sounds (it won't fix your tones though).
it's difficult to describe and people pronounce english words differently so this isn't definitive, but
yao1 is like "yao ming", like if you stub your toe and say "owww!" in pain, or "yowzers!" -- those vowel sounds are all the same for me, so if they are for you too, that's probably the standard sound for "yao" in chinese.
liu4 is much more closed mouth. rather than opening your mouth wide, your mouth will be closed as if you are pronouncing the english name of the letter "e". the vowel sounds are roughly like taking the words "he" and "who", stripping the "h" sounds, and ramming the vowel sounds together. this (for me) sounds similar to the sound a child makes when he/she sees something disgusting and says "Ewwww!" or the english word "few" (the opposite of many). hopefully those 3 also sound the same to you.
It's the tone. yao sounds very different from liu because of the tone. yi also sounds different from liu because of the tone. yi does not sound very different from qi, and yi is hard to disambiguate with multiple yi in a row.
you use yao because it's easier to understand
qi yao yao yao ba san
than
qi yi yi yi ba san
It's mostly the qi yi yi / yi yi yi that gets hard to disambiguate.
What is likely happening is that you don't yet have the ear for the tone.
The majority of Chinese don't speak any western languages. A domain name of letters is no different from a random combination of letters to them, even if the name is composed of real words in English, Spanish, or French, etc. They are completely foreign. While a combination of digits is not, like phone numbers, zip codes, bus numbers, train numbers, etc. They deal with them everyday.
My parents don't speak any other languages other than Mandarin. [To be accurate, my mother spoke some Mongolian, and my father spoke a little Russian, when they were young. But they couldn't speak any of them now.] They have no problem in recognize and pronounce the 26 letter in English. After I have been living in American for so many years, they still can't spell my home address. They remember my phone numbers and my house numbers, though, but not the street names. Think about it. It is amazing. Because I find it is much easier to remember street names than phone numbers. Phone numbers are simply a random combination of digits, except the area code.