"Please wait while other shoppers finish up."



  • I came across this on The Consumerist today. 

     
    WTF?  Surely there had to have been a better way to address a high server load than manually gating clients to only let X number check-out at once.



  • Isn't that just a somewhat more end user-friendly version of, say, "Server Too Busy"? I don't know why everyone assumes this is only presented at checkout when the submission said nothing of the sort.



  • I agree, how is this worse than a "Server busy" message?



  • @NerfTW said:

    I agree, how is this worse than a "Server busy" message?

    TRWTF is the fact that they actually produced a user friendly error message and had auto-refresh for pseudo automation.



    Is it "a user friendly" or "an user friendly"? The former actually sounds correct, but disobeys the a/an rule.



  • @Lingerance said:

    @NerfTW said:
    I agree, how is this worse than a "Server busy" message?

    TRWTF is the fact that they actually produced a user friendly error message and had auto-refresh for pseudo automation.



    Is it "a user friendly" or "an user friendly"? The former actually sounds correct, but disobeys the a/an rule.

    Use the one that sounds correct. Rationalise it because although user starts with a vowel, when spoken, it begins with a consonant sound - it sounds like 'yuser'.



  • I'd imagine the logic behind this one is comparable to that in today's WTF.

    Our servers can't handle the load.  Management won't give us the money to actually fix it...  So let's do something ugly.

     

    Of course, you'd think you could convince management that the money they'd lose to fed up customers abandoning the site would be far greater than the cost of the hardware and/or development time necessary to actually fix the bug.
     



  • But I shop online in order to avoid lines!



  • @m0ffx said:

    Rationalise it because although user starts with a vowel, when spoken, it begins with a consonant sound - it sounds like 'yuser'.

    It's actually a pretty clear rule, and your suggestion is a good way to remember it. If the beginning of the word is vowel sound, then you use an, if the beginning of the word uses a consonant sound, you use a. A handful of consonants and vowels give us trouble here (as user, which you point out sounds like yoozer in American English). The letter H has a few weird ones like 'an honor' versus 'a house.' People differ on 'historic.' Despite the fact that virtually all Americans aspirate the h (as in "hiss tor ic"), some choose the archaic way of writing: "an historic event." I haven't met an American who says "iss tor ic" when speaking, so I don't understand why they sometimes write it that way.

    I only speak American English, so I have been careful not to imply that English, Australians, Canadians, or anyone else does it quite this way. 



  • @pacohope said:

    @m0ffx said:

    Rationalise it because although user starts with a vowel, when spoken, it begins with a consonant sound - it sounds like 'yuser'.

    It's actually a pretty clear rule, and your suggestion is a good way to remember it. If the beginning of the word is vowel sound, then you use an, if the beginning of the word uses a consonant sound, you use a. A handful of consonants and vowels give us trouble here (as user, which you point out sounds like yoozer in American English). The letter H has a few weird ones like 'an honor' versus 'a house.' People differ on 'historic.' Despite the fact that virtually all Americans aspirate the h (as in "hiss tor ic"), some choose the archaic way of writing: "an historic event." I haven't met an American who says "iss tor ic" when speaking, so I don't understand why they sometimes write it that way.

    I only speak American English, so I have been careful not to imply that English, Australians, Canadians, or anyone else does it quite this way. 

    Actually, when I see a highway sign indicating a town's "Historic District," I prefer to pronounce it rhymingly -- pronouncing the first word more like 'history'.  Hiss-trick diss-tricked.



  • @pacohope said:

    @m0ffx said:

    Rationalise it because although user starts with a vowel, when spoken, it begins with a consonant sound - it sounds like 'yuser'.

    It's actually a pretty clear rule, and your suggestion is a good way to remember it. If the beginning of the word is vowel sound, then you use an, if the beginning of the word uses a consonant sound, you use a. A handful of consonants and vowels give us trouble here (as user, which you point out sounds like yoozer in American English). The letter H has a few weird ones like 'an honor' versus 'a house.' People differ on 'historic.' Despite the fact that virtually all Americans aspirate the h (as in "hiss tor ic"), some choose the archaic way of writing: "an historic event." I haven't met an American who says "iss tor ic" when speaking, so I don't understand why they sometimes write it that way.

    The irony is that the rule is actually iron-clad: use 'an' when followed by a vowel and 'a' when followed by a consonant.

    The problem is not with "some vowels and consonants". The problem is that your teacher, when you were five years old, followed the "modern" prescribed program of telling you that the vowels are the letters 'a', 'e', 'i', 'o', and 'u', and that the consonants are the other letters.

    This is, in fact, a complete load of bollocks. Vowels and consonants are sounds, not letters. Everything that you were taught about how words are constructed is probably wrong.

    The phoneme is the fundamental unit of speech (a sound which can't be broken up into component parts). A vowel is any phoneme made by passing air through a fully open configuration of your vocal system, and a consonant is any phoneme made by a configuration which is not fully open (you need a diagram to understand this precisely, but you can get a rough understanding of what it means shortly...).

    The basic English vowels (assuming a fairly clear pronunciation, not a Texan) are:

    • ea in weak
    • i in is
    • i in tip
    • a in about
    • e in bed
    • u in nut
    • u in fur
    • a in fat
    • ar in cart
    • oo in boot
    • oo in hook
    • oa in boat
    • o in hot

    You read that right - there are about a dozen distinct vowels in common usage, not five. There are actually 28 distinct vowels in all well-known spoken languages; some dialects of English will move a few of those around, combine one or two, or even add in a couple of the missing ones. The rest are only found in other languages. English uses an uncommonly large number of them, which is a big part of the reason why many foreigners have trouble pronouncing it (Japanese, for example, has only five vowels). I've only used the letters 'a', 'e', 'i', 'o' and 'u' here because I wanted to use words that have distinct pronunciations in most dialects, but there is no direct connection between those letters and the vowels - the 'y' in rhythm is a vowel, and the second 'a' in abacus is not a vowel.

    Try pronouncing each of them out loud, and pronouncing some other phonemes, and pay attention to how the air moves through your mouth, what you're doing with your tongue, and where exactly the sound is coming from. You should get a rough idea why we describe vowels as an 'open configuration' and how we distinguish between them.

    (In case you were wondering, there are 68 well-known consonants, most of which English uses)

    I do not know what the historical origin is of the gibberish that is commonly taught to young children, but in recent years there has been a push back to teaching them how English actually works, because some researchers did a study and found that, surprisingly enough, they find it easier to learn rules which actually work rather than "simpler" made-up ones which don't. Unfortunately the whole thing has become political and promptly split down partisan lines.



  • @asuffield said:

    I do not know what the historical origin is of the gibberish that is commonly taught to young children, but in recent years there has been a push back to teaching them how English actually works, because some researchers did a study and found that, surprisingly enough, they find it easier to learn rules which actually work rather than "simpler" made-up ones which don't. Unfortunately the whole thing has become political and promptly split down partisan lines.

    Plus, they'd have to change the rules to Wheel of Fortune.
     



  • I remember back in '96 or 97 I was in an audience watching a Cold Fusion demo from the good folks at Allaire, they were telling the audience how robust CF was, and to demonstrate went to some e-com site that was meant to be their biggest. The demo jock typed in the URL and the browser showed a nice CF error- Server has reached the maximum number of connections available for this site- 15. Ahh live theatre.

     

     

     

     



  • @asuffield said:

    the second 'a' in abacus is not a vowel.

     

    How is that not a schwa?  (i.e. how is it not the same as 'a' in about?)

     



  • If they can afford a parade, they can afford a few more servers.



  • @emurphy said:

    @asuffield said:

    the second 'a' in abacus is not a vowel.

    How is that not a schwa?  (i.e. how is it not the same as 'a' in about?)

    Okay, I suppose this bit may be limited to the British dialect (I don't have an American on hand to hear how they pronounce 'abacus'), but when I say it, the 'ba' is pronounced as the 'b' consonant, and there is no vowel after it. It's difficult to find examples of this that are consistent across dialects.


  • Discourse touched me in a no-no place

    @asuffield said:

    @emurphy said:

    @asuffield said:

    the second 'a' in abacus is not a vowel.

    How is that not a schwa?  (i.e. how is it not the same as 'a' in about?)

    Okay, I suppose this bit may be limited to the British dialect

    Or a particular British dialect...
    (I don't have an American on hand to hear how they pronounce 'abacus'), but when I say it, the 'ba' is pronounced as the 'b' consonant, and there is no vowel after it. It's difficult to find examples of this that are consistent across dialects.

    I'm having difficulty auralising[1] that. Ab-a-cus is how I pronounce it and others that I've heard say it.

    You appear to be describing ab-cus as a pronunciation and I've never heard anyone do it that way.

     

    [1] I know that's not a word. Is there a specific audio equivilent of 'visualise' for sounds? Beyond visualise of course. 



  • @PJH said:

    @asuffield said:
    Okay, I suppose this bit may be limited to the British dialect
    Or a particular British dialect...
    (I don't have an American on hand to hear how they pronounce 'abacus'), but when I say it, the 'ba' is pronounced as the 'b' consonant, and there is no vowel after it. It's difficult to find examples of this that are consistent across dialects.

    I'm having difficulty auralising[1] that. Ab-a-cus is how I pronounce it and others that I've heard say it.

    You appear to be describing ab-cus as a pronunciation and I've never heard anyone do it that way.

     

    [1] I know that's not a word. Is there a specific audio equivilent of 'visualise' for sounds? Beyond visualise of course. 

    I'm assuming that he means a-ba-cus, not ab-cus.



  • @RayS said:

    @PJH said:
    @asuffield said:
    Okay, I suppose this bit may be limited to the British dialect
    Or a particular British dialect...
    (I don't have an American on hand to hear how they pronounce 'abacus'), but when I say it, the 'ba' is pronounced as the 'b' consonant, and there is no vowel after it. It's difficult to find examples of this that are consistent across dialects.

    I'm having difficulty auralising[1] that. Ab-a-cus is how I pronounce it and others that I've heard say it.

    You appear to be describing ab-cus as a pronunciation and I've never heard anyone do it that way.

    I'm assuming that he means a-ba-cus, not ab-cus.

    a-b-cus, to be precise - three syllables, the first being a bare vowel, the second a bare consonant, and the third a CVC.
     



  • @asuffield said:

    a-b-cus, to be precise - three syllables, the first being a bare vowel, the second a bare consonant, and the third a CVC.
     


    according to wikipedia (i know, not to be trusted blindly), only sonorant sounds can serve as a syllable nucleus (which is required in all languages), and "b" is not a sonorant sound, thus you cannot have a bare "b" syllable. you're probably either unconsciouslly ignoring a sonorant after it, or it's just the coda of the first syllable of the word.



  • So not, a-b-cus, but a-bu-cus (or a-buh-cus?).

    In which case its still a vowel: "u".
     



  • @PJH said:

    I'm having difficulty auralising[1] that.
    [1] I know that's not a word. Is there a specific audio equivilent of 'visualise' for sounds? Beyond visualise of course.

    Vocalize?



  • @lanzz said:

    @asuffield said:

    a-b-cus, to be precise - three syllables, the first being a bare vowel, the second a bare consonant, and the third a CVC.
     


    according to wikipedia (i know, not to be trusted blindly), only sonorant sounds can serve as a syllable nucleus (which is required in all languages), and "b" is not a sonorant sound, thus you cannot have a bare "b" syllable. you're probably either unconsciouslly ignoring a sonorant after it, or it's just the coda of the first syllable of the word.

    I assure you that I am quite capable of pronouncing a lone consonant without a vowel after it. You can call it two syllables if you really want, but that would be rather silly because it's obviously three (one dactylic foot).



  • Oh God I hate when these threads devolve into the nitty gritty of this language or that language.  There are a lot of things I like delving into the nitty gritty of, but language is not one of them.

    That said, asuffield, I don't see a lone consonant without a vowel after it in the word you promoted.  The y functions as the vowel.


  • Discourse touched me in a no-no place

    @Lingerance said:

    @PJH said:
    I'm having difficulty auralising[1] that.
    [1] I know that's not a word. Is there a specific audio equivilent of 'visualise' for sounds? Beyond visualise of course.

    Vocalize?
    Not quite.


    I was trying to imagine hearing it, rather than saying it.

    The difference between Aural and Oral so to, erm, speak.



  • @shadowman said:

    @asuffield said:

    I do not know what the historical origin is of the gibberish that is commonly taught to young children, but in recent years there has been a push back to teaching them how English actually works, because some researchers did a study and found that, surprisingly enough, they find it easier to learn rules which actually work rather than "simpler" made-up ones which don't. Unfortunately the whole thing has become political and promptly split down partisan lines.

    Plus, they'd have to change the rules to Wheel of Fortune.
     

    "Pat, I'd like to buy a diphthong."

     



  • @asuffield said:

    The basic English vowels (assuming a fairly clear pronunciation, not a Texan) are:

    • ea in weak
    • i in is
    • i in tip
    • ...

    We have an older (1984?) Collins English Dictionary whose guide to the International Phonetic Alphabet on the inside of the cover claims that in the word "busy", the same vowel is used twice. I don't get it. It's "bi-zee" with two distinct vowels. I speak with a mixed neutral English accent, sort of southern, sort of northern.

    And now I see that you differentiate the vowel in "is" and "tip", which puzzles me, as I don't see how that would work. I don't recollect any English accent where they could differ in that way.

    You also appear to consider schwa not a vowel sound, although the dictionary definition of schwa is that of a vowel sound. Take the word "bat" -- you don't say "buh-at", so why is a bare 'b', "buh"? I would consider consonants to be modifiers that cannot be pronounced alone, so the middle syllable of "abacus" is b-schwa, not b, and I think this is how everyone else would also consider it. What's 'f' as a bare vowel? "fff"? "fuh"?



  • @Daniel Beardsmore said:

    And now I see that you differentiate the vowel in "is" and "tip", which puzzles me, as I don't see how that would work. I don't recollect any English accent where they could differ in that way.

    You also appear to consider schwa not a vowel sound, although the dictionary definition of schwa is that of a vowel sound. Take the word "bat" -- you don't say "buh-at", so why is a bare 'b', "buh"? I would consider consonants to be modifiers that cannot be pronounced alone, so the middle syllable of "abacus" is b-schwa, not b, and I think this is how everyone else would also consider it. What's 'f' as a bare vowel? "fff"? "fuh"?

    Okay, so my attempt to find a dialect-neutral set of examples was imperfect. The 'i' in 'tip' should be the near-close near-front unrounded vowel ɪ in all dialects. The 'i' in 'is' should be the close central unrounded vowel ɨ in those dialects which have it, but some dialects may merge these two and not use ɨ at all (I use both).

    The 'a' in 'about' is the schwa everywhere, that's in the list I gave. I suppose you could pronounce 'abacus' with bə in the middle syllable, but that just sounds weird, and I don't. 'f' is not a bare vowel at all, since it's a consonant, and it's the sound made when you put your upper teeth on your lower lip (labiodental), your tongue at the bottom of your mouth (open), and blow without vocalising; since it's a fricative, you can keep doing this until you run out of air, and never generate a vowel.

    In the word 'bat', the first syllable is the voiced bilabial plosive consonant 'b' followed by a vowel (which varies with dialect). This does not mean that every other syllable beginning with the 'b' consonant is the same. You pronounce b by placing your lips together and then blowing them apart while vocalising. If you continue to vocalise after this then you are pronouncing a vowel; if you stop, then there is no vowel (this is normally used in gemination, which is uncommon in English but still present).

    You seem to be having difficulty with the concept of consonants itself. 



  • What's with the loony span tags? Where does CSS class "IPA" come from in TDWTF? I guess you're using the WYSIWYG post editor and copying from another page such that HTML is cross-inserted straight into the post?

    @asuffield said:

    Okay, so my attempt to find a dialect-neutral set of examples was imperfect. The 'i' in 'tip' should be the near-close near-front unrounded vowel ɪ in all dialects. The 'i' in 'is' should be the close central unrounded vowel ɨ in those dialects which have it, but some dialects may merge these two and not use ɨ at all (I use both).

    I seem to form the two sounds identically. I'm not aware of any way that they can sound different to each other -- it would be interesting to hear a recording of it.

    @asuffield said:

    The 'a' in 'about' is the schwa everywhere, that's in the list I gave. I suppose you could pronounce 'abacus' with bə in the middle syllable, but that just sounds weird, and I don't.

    It's almost splitting hairs. The spelling is "ba" but as an unstressed syllable, it comes out close to "bə". You could argue that it's strictly shorter than "bə", but then, what's "butter"? The final syllable is what? The American Heritage Dictionary has "bŭt'ər", i.e. an 'r' after the schwa, which would obviously justify abacus having "bə" (which is how they write it). The sound "ba" in Abacus, if treated as "bə", is identical to about starting "ə-b".

    @asuffield said:

    'f' is not a bare vowel at all, since it's a consonant ...

    I imagine you realised that I meant to type consonant, since "bare consonant" was your term, not mine.



  • The Real WTF. What is it? Is it that this, too, has devolved into a language nitpicking thread? No. Is it that the original poster hotlinked the screenshot which somehow broke it (at least for me)?

    No; I think the real WTF is the session cookies that the consumerist.com server is trying to set. Seeing "SESSID_GANJA" and "GANJAUSERSETTINGS" thrown in your face makes you wonder what the hell the people putting that site together were thinking. Or alternatively, it makes it pretty clear...



  • @Daniel Beardsmore said:

    What's with the loony span tags? Where does CSS class "IPA" come from in TDWTF? I guess you're using the WYSIWYG post editor and copying from another page such that HTML is cross-inserted straight into the post?

    Yes, my keyboard doesn't actually have IPA characters on it, so I paste them from wikipedia's IPA article. Hadn't noticed that it picked up a pointless span in the process.



  • There is no way I'm going to vocalise the second A in abacus. ab'cus.

    Despite the fact that virtually all Americans aspirate the h (as in
    "hiss tor ic"), some choose the archaic way of writing: "an historic
    event." I haven't met an American who says "iss tor ic" when speaking,
    so I don't understand why they sometimes write it that way.

    I write a historic, but pronounce anistoric. But not always. I like to pull stuffagether.

    Also, HHHHHHHHHHerbs. Beause, quoth' Eddie Izzard, there's an H in it.



  • 'an user' is correct, i suppose, even though it doesn't sound like it :/

    the 'a' / 'an' rule is a *sound* rule, not a *spelling* rule. if thww first letter sounds like it starts with a vowel, use 'an'.

    thats why you write "an hour", the 'h' is silent and *sounds* like a vowel :)

    i should know, i've been speaking english handsomely since i was twice ;)


  • Discourse touched me in a no-no place

    @Junkieman said:

    'an user' is correct, i suppose, even though it doesn't sound like it :/

    I take it you haven't read the rest of this thread. 

    the 'a' / 'an' rule is a *sound* rule, not a *spelling* rule. if thww first letter sounds like it starts with a vowel, use 'an'.
    And if you read the rest of the thread, you'll find that user doesn't start with a vowel sound. It starts with a similar sound to that at the start of 'yellow.'

    Would you construct a sentence with the phrase 'an yellow ball?'

    Actually, judging by your first and second sentences taken together, you would actually.... 

    i should know, i've been speaking english handsomely since i was twice ;)
    And it shows.
     




  • @dhromed said:

    Also, HHHHHHHHHHerbs. Beause, quoth' Eddie Izzard, there's an H in it.

    I think you meant to say, "Because there's a failureing H in it."

    @Junkieman said:

    i should know, i've been speaking english handsomely since i was twice ;)

    Vincent Twice, Vincent Twice



  • @m0ffx said:

    Use
    the one that sounds correct. Rationalise it because although user
    starts with a vowel, when spoken, it begins with a consonant sound - it
    sounds like 'yuser'.



    This looks strange to me, do you mean that 'Y' is classified as a consonant in english? I would have written what you probably mean as "djuser".

     
    And now for some trivia: 

    Here in Sweden we have 9 wovels: A, O, U, Å - the "hard" wovels, and E, I, Y, Ä, and Ö - the "soft" wovels. Most of these also have "short" and "long" pronounciations, which are different. Five vowels seem to be too few for a proper language! :)



  • @Lingerance said:


    Is it "a user friendly" or "an user friendly"? The former actually sounds correct, but disobeys the a/an rule.

    Man, look at the crap you started.

     



  • @boh said:

    This looks strange to me, do you mean that 'Y' is classified as a consonant in english?

    See above. Letters are not tied to any particular sound in English. The letter 'y' can represent numerous sounds, including both vowels and consonants.


    Here in Sweden we have 9 wovels: A, O, U, Å - the "hard" wovels, and E, I, Y, Ä, and Ö - the "soft" wovels. Most of these also have "short" and "long" pronounciations, which are different. Five vowels seem to be too few for a proper language! :)

    The long pronounciations are probably different vowels in their own right that just share a letter. I don't have a Swede on hand to find out. The wikipedia article on swedish phonology claims 17 distinct vowels, but I don't know how accurate that is. Strangely enough there are only 18 consonants listed, which is about half what most languages use, but might account for the unusual sound of Swedish.

    Japanese's five vowels is very unusual, but then the structure of the spoken language itself is very unusual - it is far more orderly than normal, having a very simple and rigid structure to its phonology, which makes it much more efficient in terms of the amount of information carried by each sound. Most languages need far more phonemes (and for that matter, much longer phrases) to carry the same information.



  • @asuffield said:

    Japanese's five vowels is very unusual, but then the structure of the spoken language itself is very unusual - it is far more orderly than normal, having a very simple and rigid structure to its phonology, which makes it much more efficient in terms of the amount of information carried by each sound. Most languages need far more phonemes (and for that matter, much longer phrases) to carry the same information.

    Each sound doesn't really carry more information in Japanese; there are fewer sounds, so if anything they carry less information, strictly speaking; and you really need context to figure out what each particular sound is supposed to mean.

    But I guess that's not really what you're saying. What you say is perfectly true for the written language, each unit really does carry more information. Makes things more "interesting", to say the least...



  • @magetoo said:

    @asuffield said:
    Japanese's five vowels is very unusual, but then the structure of the spoken language itself is very unusual - it is far more orderly than normal, having a very simple and rigid structure to its phonology, which makes it much more efficient in terms of the amount of information carried by each sound. Most languages need far more phonemes (and for that matter, much longer phrases) to carry the same information.

    Each sound doesn't really carry more information in Japanese; there are fewer sounds, so if anything they carry less information, strictly speaking; and you really need context to figure out what each particular sound is supposed to mean.

    No, I really do mean that the information content of each sound is higher, because it is more efficient than most other languages in this respect. In a language like English, there is a considerable amount of redundancy; on average, written English prose has one bit of information per character (and consumes about five bits of coding space per character). The ratio for the spoken form is similar, I just don't know the numbers offhand. The redundancy in Japanese is lower, so the information content of each sound is higher. This is partly because English has whole ranges of impossible nonsense sounds like 'kkqzx', while you just can't construct something like that in Japanese.



  • @Daniel Beardsmore said:

    What's with the loony span tags? Where does CSS class "IPA" come from in TDWTF? I guess you're using the WYSIWYG post editor and copying from another page such that HTML is cross-inserted straight into the post?

    @asuffield said:

    Okay, so my attempt to find a dialect-neutral set of examples was imperfect. The 'i' in 'tip' should be the near-close near-front unrounded vowel ɪ in all dialects. The 'i' in 'is' should be the close central unrounded vowel ɨ in those dialects which have it, but some dialects may merge these two and not use ɨ at all (I use both).

    I seem to form the two sounds identically. I'm not aware of any way that they can sound different to each other -- it would be interesting to hear a recording of it.

    You don't happen to be from the West Coast of the US (Washington, Oregon or California)?  Like Texas and the South, the (rather new) West Coast accent has a lot of drawl to it.  I've got one, and it significantly compresses my vowel set and alters which vowels I use where.  This, combined with a nasty tendency to use complex sentence structure, makes me difficult to understand for non-native speakers (it's my own fault, and I've got a lot of patience when trying to make myself understood). 

    For instance, I mispronounce 'abacus', which Merriam Webster says you break a-bə-kəs, but I tend to slur it a little more into the ab-ə-kəs realm.  I also pronounce the 'i' in is and tip the same (close central unrounded version).

    (And, yes, I'm talking about that surfer/stoner accent, y'know, like, it's one of those like ones where y'know you open your mouth, and the dude you're talking to is all, like, wow, that dude smokes, like waaaaaaay too much pot?  It's a similar effect to Jeff Foxworthy's automatic IQ demerit that people give to Southern accents.) 



  • @asuffield said:

    No, I really do mean that the information content of each sound is higher, because it is more efficient than most other languages in this respect. In a language like English, there is a considerable amount of redundancy; on average, written English prose has one bit of information per character (and consumes about five bits of coding space per character). The ratio for the spoken form is similar, I just don't know the numbers offhand. The redundancy in Japanese is lower, so the information content of each sound is higher. This is partly because English has whole ranges of impossible nonsense sounds like 'kkqzx', while you just can't construct something like that in Japanese.

    Still sounds (rimshot) to me like you really mean the written language. (What does "kkqzx" sound like when spoken anyway? But it would be interesting to see comparisons with English using, say, IPA, though.) I certainly believe you when you say the redundancy is lower, and that the fundamental units doesn't allow for impossible-to-pronounce unwords, but you still have the massive amount of homonyms. If the sounds themselves carried so much information, my IME wouldn't offer up 10-20 different suggestions for what word I really mean when typing "foo". The language might be more efficient, but it's not from having the sounds carry more information.

    I suspect there's truth on both sides here, but we're talking about slightly different things.

    Edit: I think this is really interesting, but it's a bit off topic here. Do you know of any decent forums/etc for talking about language?



  • @magetoo said:

    What does "kkqzx" sound like when spoken anyway?

    Three clicks and a buzz. It takes a little practice.

     

    I certainly believe you when you say the redundancy is lower, and that the fundamental units doesn't allow for impossible-to-pronounce unwords, but you still have the massive amount of homonyms. If the sounds themselves carried so much information, my IME wouldn't offer up 10-20 different suggestions for what word I really mean when typing "foo".

    It's lower, it's not hugely lower. The difference is on the order of a few percent.

     

    The language might be more efficient, but it's not from having the sounds carry more information.

    I'm using "information" in the sense of Shannon coding theory, so overall efficiency and per-symbol information content are the same thing (efficiency is defined as the mean information content of a symbol divided by the logarithm of the total number of symbols).

    Do you know of any decent forums/etc for talking about language?

    No, my exposure to linguistics is primarily through programming language theory (which is something of an odd branch - you need to know the basics, but the history of Sumerian isn't particularly relevant).



  • @asuffield said:

    The language might be more efficient, but it's not from having the sounds carry more information.

    I'm using "information" in the sense of Shannon coding theory, so overall efficiency and per-symbol information content are the same thing (efficiency is defined as the mean information content of a symbol divided by the logarithm of the total number of symbols).

    Right. And I'm talking about the theoretical maximum information content of each symbol, I guess. I think the conclusion would be that Japanese makes more efficient use of the symbols it has. Eliminating redundancy, even to the point where it creates ambiguity.

    I think I'm going to be reading a lot about information content in spoken language over the holidays. Time to raid the library. :-)

    Do you know of any decent forums/etc for talking about language?

    No, my exposure to linguistics is primarily through programming language theory (which is something of an odd branch - you need to know the basics, but the history of Sumerian isn't particularly relevant).

    Aha. Well, in the interest of Getting To Know each other; I took some basic linguistics and classical language courses in high school myself, coming straight from your average computer geek background. Now, much later, I've (tried to) read up on basic DSP, taken a few CS courses at university, and just recently jumped on a Japanese language course/program. Computational linguistics would have been the sensible choice, in hindsight...

    Anyway. I'll start a thread in General Discussion when/if I have something interesting to add. (I don't suppose you're active on the xkcd forums, or somewhere like that?)


Log in to reply
 

Looks like your connection to What the Daily WTF? was lost, please wait while we try to reconnect.