That's the key question. What tasks are you doing regularly that your past failures to learn high school math are stopping your from?
I use some form or another of "math" regularly, but I'll tell you one thing: most of high-school math isn't very useful for me. I've never needed calculus, and barely ever needed geometry. Algebra is ocassionally useful, but the very basic bits of it are good enough (I remember that there is such a thing as the quadratic equation and factorization of polynomials, but I've never really needed to use them).
On the other hand, graph theory, mathematical logic, lambda calculus, probability and statistics have been very useful, and I suspect abstract algebra would also be so if I understood it. But guess what? None of those are regularly taught in high school. (Hell, mathematical logic isn't even regularly taught in university math departments.)
Don't just assume you need high school math. Make some effort to figure out what kind of math would be useful, and go with that. If you're into programming, you may want to try a discrete mathematics textbook.
It's unconventional ("we and pogue" would be more idiomatic), but I don't think it's ungrammatical; note that this is a subject, not an object (hence "we", not "us"). Am I missing something?
Yes, you're missing something. You're not testing your hypothesis against data from actual usage.
Neither have I, but I'm going to guess:
There will be a preference for the order that places the pronoun second, but plenty of examples of either order.
Coordinated object pronouns (like in Pogue and us) will be far more frequent in subjects than coordinated subject pronouns (Pogue and we). This probably means that whatever folk theory you have about when to use subject pronouns and when to use object pronouns is false.
Pogue and we, like that, with a subject pronoun, is hypercorrection. It's a construction that exists only because some people, who have fundamental misunderstandings about grammar, formulate rules that are clearly contradicted by the actual usage data, and then bully others into writing according to those rules.
It's grammatically correct, but it's very awkward. The grouping of the collective "we" on an equal footing with "Pogue" strains the mental picture of "we". This grouping, intimating a close association, is such that Pogue would naturally be assumed to be part of the "we" in question, so puzzlement ensues when he is not.
On what basis do you claim it's "grammatically correct"? You don't make an argument, and I suspect you know jack about grammar. (Do you happen to have at the very least a B.A. in Linguistics, or equivalent experience?)
We is not a collective, under standard grammatical terminology, which usualy uses the term to refer to singular nouns that denote groups. We just a plural.
Why would we and Pogue be coordinated within a noun phrase, instead of using we in a way that includes, um, us and Pogue? (Note how I just had to coordinate us and Pogue! Could I have said just us there?)
Why, because they're different discourse referents, which have been introduced into the text separately; but more importantly, because the text opposes Pogue to us (where "us" is the author of the text and the readers it is addressed to). That's the whole point of the piece--the claim that there is a conflict between Pogue's interests and ours.
"Strains the mental picture"? WTH are you talking about?
You go use a computer somewhere else, of course. Do you seriously think this is a serious problem that should cause us to sacrifice the ability for them to use their own script in the 99.99% of the times they are able to?
Travelers deal with much bigger problems than this all the time, in any case.
Unless I'm mistaken, all keyboards can do the basic ASCII characters.
And guess what, all keyboards can do any characters. The keyboard is just a lot of switches and a cable that tells the computer which switches are pressed; the mapping between keys and characters is established by keyboard layouts, which are configurable in all major OSes. The differences between keyboards for one script and another are, for the most part, the symbols printed on the keys.
Having a keyboard layout that doesn't match your keyboard is certainly not optimal, but it's not terrible; back in the days where Spanish language keyboards were still not widely imported into my country, and all the keyboards were in English, plenty of people learned to type Spanish layout on an American English keyboard. In the worst case, just buy some damn stickers to put on the keys.
How are you going to enter these URLs on a standard keyboard from a different region?
By typing them, of course. After all, the standard keyboards from a different region are designed for that script!
Seriously, dude, it's not like your OS doesn't support international keyboard layouts, or like you can't buy international keyboards, or even just sticker overlays for your existing keyboard. If you care to learn a language with a different script, you can most certainly use it in the computer that you have today.
And you're making another fallacy (I think; I may just misunderstand you a bit). You assume that "first character typed = leftmost character" and that the "natural" order for a computer to display them is from the left. And of course neither is true.
(Yes, I know. I was playing dumb. I'm arguing the same points as you are in a few threads below...)
I have to wonder what's to stop someone from going out and registering www.ébay.com? I mean, it seams like this would be pretty easily abused by scammers and phishers.
Have the registrars and/or administrators of DNS zones enforce policies that decide which domain names are allowed in their zones, and also, which are equivalent to each other. That way, the.com zone administrators can just forbid registration of 'ébay', and in addition, enforce that queries for 'ébay' must resolve the same way as queries for 'ebay' do.
We already do this: you can't register 'EBAY.COM' separately from 'ebay.com', and if you type the former, you get the same answer as the latter.
Oh, no. Wait. I just thought of something bad. You know, when I actually get to this site, it's probably going to be really hard to understand what's written on the page. Funny squiggles and such. I suppose there's really just no reason for me to go to such a page, if I can't read it anyway, so why even bother?
That one's easy. I'd read it just for the pictures.
How are they going to use them in a net cafe on the other side of the world?
Oh great. So you are proposing that we do not improve their browsing experience in the case that holds more than 99% of the time, because of what happens in the relatively rare number of cases when people travel abroad?
They'll just figure out how to turn on an appropriate input method or keyboard layout in the computers in the net cafe, and use that. Sure, this is somewhat awkward, because the layout often doesn't match the symbols on the keycaps, so they'll have to hunt often to figure out which key enters what character. But dude, when people travel, they already have to figure out how to do all sorts of things differently to cope with an unfamiliar environment; this is just one more detail.
I doubt that. Your computer is already able to handle those characters. Have you bothered to check out your OS's multilingual text processing features? I don't know what's the state of Linux distributions nowadays in this regard, but Windows and OS X certainly come with character palette utilities, and a selection of alternate keyboard layouts and input methods tailored for other languages.
And if it's really that important to you to be able to type text in a particular language, well, just buy a keyboard for it.
By using them in the DNS system, a part of the net is effectively being segregated.
How often do your read sites in Chinese, Arabic, Japanese, Russian, Korean or Hebrew? Seriously? If you do, well, then, shouldn't you figure out how to enter text in the corresponding scripts?
Are you under some sort of misconception that relevant sites with internationalized domain names would not show up in your search results? That you wouldn't be able to email them? That your computer wouldn't be able to resolve the host names' IP addresses? That links to them would not work?
As for accusations of "cultural imperialism" - can I just point out that English speaking people developed the Internet at their own time and expense (and a lot of tax-payers money) - so they are entitled to have it in English if they want [...]
The USA is not, and has never been, a monolingual country. English enjoys a privileged status, but there isn't a time in the history of the USA where there hasn't been a lively non-English speaking cultural scene in many parts of the country, simultaneously. (And they pay their taxes, too.)
So just have DNS zones set up equivalence rules between domain names, so that the difference between, for example, E, e, É, é, È, è, Ê and ê is ignored for DNS requests within zones where it is appropriate to do so.
The fact that multilingual domain names would require some policy decisions in the parts of registrars doesn't have nearly as much importance to the task of providing a mechanism for providing such names.
If it's going to use characters not present on normal keyboards, what's the point?
To use characters present in abnormal keyboards. You know, the kind of keyboards that people have in countries where these sites' audiences would be.
I am impressed by the fact that you managed to frame the question in exactly the relevant terms, what characters are present in your keyboard, and not the facile, easy way out that those crazy foreigners do, when they insist in framing this in terms of presenting addresses in their own native language and script.
You're making the assumption that the computer should display characters in the order in which they were typed. You need to lose that assumption. A good technical solution would be simply to make the computer able to display the characters in the opposite order to which they were typed. That way, languages that are written backwards can be displayed correctly even when their users type the text in the correct order.
Thanks for pointing this out, and for bringing up the verb "to be." This is, by default, the oldest verb in any language (except perhaps Russian, which they tell me doesn't have it), and therefore the most irregular.
What do you mean by "oldest"?
You'd have an argument there if you'd said "most frequent" instead of "oldest." Other people have pointed out how you messed up Russian (and dear God, can't you look up stuff before you spout off?).
Based on this, I have formulated the theory that "to be" is irregular in every language (that has it). In good scientific methodology, I am seeking out evidence to the contrary. Can anyone provide any?
You know, idly making hypotheses about stuff you don't know anything about, and then demanding that others do the hard work of testing them for you, hardly qualifies as "good scientific methodology."
The question is ill-defined anyway, since you give us no criterion for deciding which words in other languages count as "to be." Do both ser and estar in Spanish qualify? Ok, that one might not be a problem, because they are both "irregular" (but more on that below). Does Cape Verdean Creole é (as in Mi é bu amigu 'I am your friend') count as a verb? That one doesn't vary with person or number, but its grammar is otherwise very odd compared to other verbs (long story).
An even more profound question: how do we decide if a verb is "irregular"? For example, in the traditional grammars of Romance languages, most verbs are classified as belonging into one of a handful of conjugation groups. In Spanish, for example, these are the verbs that end in -ar, -er and -ir; in French, -er, -ir and -re. Verbs that can't be correctly conjugated just by applying the rules appropriate to one of these classes are called "irregular."
Now the problem is that these definitions of "irregular" are language-specific; the definition of "irregular verb" for Spanish is no use for Turkish. Therefore, the results that you obtain for one language may not be comparable to those you obtain for another.
I read your comment carefully, and it still strikes me as irrelevant; the study is about the rate of language change, not about mechanisms.
At any rate, the idea that child language is the primary motor of language change is hopelessly flawed. I'd have to look it up, but IIRC there's a lot of sociolinguistic research that casts doubt on it. Language change seems not to be driven by children, but by adolescents and young adults, who've long. (And the really big mechanism is phonological change, not morphological assimilation, anyway.)
Re:Yes, well...however...there are other methods.
on
The Evolution of Language
·
· Score: 3, Interesting
Languages are 'derived', sure - they evolve as derivations of other languages and/or common usage that pushes some words into popularity while others fall into history.
Linguistics 101 lesson: a language is not a bag of words. Any generalization about language that treats it as if it is some bag of words (e.g., in this case, that language change consists of new words entering the bag, while other words fall out of it) shows a profound ignorance of the fundamental ideas of linguistics. A language is a grammar; people invent and adopt new words spontaneously all the time, but not, say, morphological paradigms, case agreement, or new forms of valence-changing rules like the passive or the causative alternation. (Yes, I'm using words that most people who read this won't understand, but that's the point--if you don't understand terms like that, your "insights" into laguage aren't very valuable.)
The Korean language that has been in use for the last four hundred years is the only 'human' invented language on the planet. At one time, when the country was unified by one King, it became clear that the multiplicity of dialects in use around the country were barriers prohibiting trade, mobility, communication, learning from each other, etc. The top thinkers were gathered and ordered to design a language that was simple to learn and speak...read and write. Once this was done, the King simply decreed that all citizens adopt it, shedding their separate dialects.
That sounds like a combination of myth and hyperbole about a perfectly ordinary language standardization process (e.g., the kind that happened in Spain during the reign of Alfonso X of Castile, and again after the publication of Antonio de Nebrija's grammar). I don't know what Korean king you're talking about here; my first thought was Sejong the Great, but the timeline is wrong (he lived about 600 years ago, not 400). At any rate, his great contribution was an orthography (Hangeul) that wasn't adopted until much later.
This isn't really news. We linguists have known this for a long time, as the article mentions, and we've known why: a child learning a language tends to regularize irregular forms.
Are you aware of any historical linguistics research that makes quantitative hypotheses about the relationship between word frequency and morphological regularization? I don't know if there are any (and I wouldn't be surprised either way), but whether this study is "news" depends on the answer to that question, not on the all-too-well known fact that children learn regular inflectional paradigms before they learn irregular ones.
...that Rudolf Flesch doesn't know what the hell he's talking about.
Since Chinese is a spoken language rather than a written language (The writing is mostly pictorial representing whole concepts),
Um, Chinese is no more of a spoken language than English is, and no less of a written language either. Chinese script isn't "pictorial," either; it's logographic, with characters representing words. For mnemonic reasons, characters are related to others in a set of ways that I will not explain. (Because I don't in fact understand it well enough to explain it. See, it is possible to refrain from speaking about what one does not know.)
it wasn't frozen in place with a bunch of affixes (suffixes, prefixes, etc.) or genders and all that other stuff that makes English hard to learn.
1) How did they measure the frequency of use of these verbs? Especially now when English is being used as a second language for almost every literate person?
TFA is short on details, but they must've used a historical corpus (that's linguistese for "a database of texts").
This of course raises the question of what kind of language the corpus is representative of, and what kind of language is is not representative of. The bodies of text we have for Old and Middle English are far less representative than what we have for contemporary English; e.g., we don't have a lot of transcribed recordings of phone conversations between family members from back in Chaucer's day.
Of course, it's kind of hard to criticise this study without looking at it. The only thing that strikes me so far: none of the authors seems to be a linguist.
5. Person B strips out the BSD license, which he's allowed to do because he's not bound by the terms of the BSD license.
That's the step where it goes wrong. Person B is still bound by the terms of the BSD license.
Let's simplify it a bit, and assume just one file. Person A obtains a file consisting solely of one person's work, licensed under the BSD terms. Person A makes significant changes to it, and releases a derived work to Person B, under the GPL license. Person B now needs two licenses to modify and distribute what Person A gave them:
The license to the original work that Person A modified.
The license to the derived work that results from Person B's modifications.
The key point is that Person A can only enforce copyright on his GPL'ed work to the extent that the alleged infringing material is not based on the original, BSD-licensed file. For example, if B takes a function from A's GPLed file and puts it into a non-GPL program, A can only complain if that function didn't appear in the original BSD-licensed work.
You're assuming that if you somehow just get a bunch of smart engineers on the job, the solution will happen. Nope, it doesn't work that way in real life.
Slashdot Mirror
That's the key question. What tasks are you doing regularly that your past failures to learn high school math are stopping your from?
I use some form or another of "math" regularly, but I'll tell you one thing: most of high-school math isn't very useful for me. I've never needed calculus, and barely ever needed geometry. Algebra is ocassionally useful, but the very basic bits of it are good enough (I remember that there is such a thing as the quadratic equation and factorization of polynomials, but I've never really needed to use them).
On the other hand, graph theory, mathematical logic, lambda calculus, probability and statistics have been very useful, and I suspect abstract algebra would also be so if I understood it. But guess what? None of those are regularly taught in high school. (Hell, mathematical logic isn't even regularly taught in university math departments.)
Don't just assume you need high school math. Make some effort to figure out what kind of math would be useful, and go with that. If you're into programming, you may want to try a discrete mathematics textbook.
Are you in a position to be able to tell?
Yes, you're missing something. You're not testing your hypothesis against data from actual usage.
Neither have I, but I'm going to guess:
Why, because they're different discourse referents, which have been introduced into the text separately; but more importantly, because the text opposes Pogue to us (where "us" is the author of the text and the readers it is addressed to). That's the whole point of the piece--the claim that there is a conflict between Pogue's interests and ours.
You go use a computer somewhere else, of course. Do you seriously think this is a serious problem that should cause us to sacrifice the ability for them to use their own script in the 99.99% of the times they are able to?
Travelers deal with much bigger problems than this all the time, in any case.
And guess what, all keyboards can do any characters. The keyboard is just a lot of switches and a cable that tells the computer which switches are pressed; the mapping between keys and characters is established by keyboard layouts, which are configurable in all major OSes. The differences between keyboards for one script and another are, for the most part, the symbols printed on the keys.
Having a keyboard layout that doesn't match your keyboard is certainly not optimal, but it's not terrible; back in the days where Spanish language keyboards were still not widely imported into my country, and all the keyboards were in English, plenty of people learned to type Spanish layout on an American English keyboard. In the worst case, just buy some damn stickers to put on the keys.
By typing them, of course. After all, the standard keyboards from a different region are designed for that script!
Seriously, dude, it's not like your OS doesn't support international keyboard layouts, or like you can't buy international keyboards, or even just sticker overlays for your existing keyboard. If you care to learn a language with a different script, you can most certainly use it in the computer that you have today.
(Yes, I know. I was playing dumb. I'm arguing the same points as you are in a few threads below...)
Have the registrars and/or administrators of DNS zones enforce policies that decide which domain names are allowed in their zones, and also, which are equivalent to each other. That way, the .com zone administrators can just forbid registration of 'ébay', and in addition, enforce that queries for 'ébay' must resolve the same way as queries for 'ebay' do.
We already do this: you can't register 'EBAY.COM' separately from 'ebay.com', and if you type the former, you get the same answer as the latter.
That one's easy. I'd read it just for the pictures.
Comparing the situation with aviation is a very bad analogy. Computers don't crash.
Oh great. So you are proposing that we do not improve their browsing experience in the case that holds more than 99% of the time, because of what happens in the relatively rare number of cases when people travel abroad?
They'll just figure out how to turn on an appropriate input method or keyboard layout in the computers in the net cafe, and use that. Sure, this is somewhat awkward, because the layout often doesn't match the symbols on the keycaps, so they'll have to hunt often to figure out which key enters what character. But dude, when people travel, they already have to figure out how to do all sorts of things differently to cope with an unfamiliar environment; this is just one more detail.
I doubt that. Your computer is already able to handle those characters. Have you bothered to check out your OS's multilingual text processing features? I don't know what's the state of Linux distributions nowadays in this regard, but Windows and OS X certainly come with character palette utilities, and a selection of alternate keyboard layouts and input methods tailored for other languages.
And if it's really that important to you to be able to type text in a particular language, well, just buy a keyboard for it.
How often do your read sites in Chinese, Arabic, Japanese, Russian, Korean or Hebrew? Seriously? If you do, well, then, shouldn't you figure out how to enter text in the corresponding scripts?
Are you under some sort of misconception that relevant sites with internationalized domain names would not show up in your search results? That you wouldn't be able to email them? That your computer wouldn't be able to resolve the host names' IP addresses? That links to them would not work?
The USA is not, and has never been, a monolingual country. English enjoys a privileged status, but there isn't a time in the history of the USA where there hasn't been a lively non-English speaking cultural scene in many parts of the country, simultaneously. (And they pay their taxes, too.)
So just have DNS zones set up equivalence rules between domain names, so that the difference between, for example, E, e, É, é, È, è, Ê and ê is ignored for DNS requests within zones where it is appropriate to do so.
The fact that multilingual domain names would require some policy decisions in the parts of registrars doesn't have nearly as much importance to the task of providing a mechanism for providing such names.
To use characters present in abnormal keyboards. You know, the kind of keyboards that people have in countries where these sites' audiences would be.
I am impressed by the fact that you managed to frame the question in exactly the relevant terms, what characters are present in your keyboard, and not the facile, easy way out that those crazy foreigners do, when they insist in framing this in terms of presenting addresses in their own native language and script.
You're making the assumption that the computer should display characters in the order in which they were typed. You need to lose that assumption. A good technical solution would be simply to make the computer able to display the characters in the opposite order to which they were typed. That way, languages that are written backwards can be displayed correctly even when their users type the text in the correct order.
What do you mean by "oldest"?
You'd have an argument there if you'd said "most frequent" instead of "oldest." Other people have pointed out how you messed up Russian (and dear God, can't you look up stuff before you spout off?).
You know, idly making hypotheses about stuff you don't know anything about, and then demanding that others do the hard work of testing them for you, hardly qualifies as "good scientific methodology."
The question is ill-defined anyway, since you give us no criterion for deciding which words in other languages count as "to be." Do both ser and estar in Spanish qualify? Ok, that one might not be a problem, because they are both "irregular" (but more on that below). Does Cape Verdean Creole é (as in Mi é bu amigu 'I am your friend') count as a verb? That one doesn't vary with person or number, but its grammar is otherwise very odd compared to other verbs (long story).
An even more profound question: how do we decide if a verb is "irregular"? For example, in the traditional grammars of Romance languages, most verbs are classified as belonging into one of a handful of conjugation groups. In Spanish, for example, these are the verbs that end in -ar, -er and -ir; in French, -er, -ir and -re. Verbs that can't be correctly conjugated just by applying the rules appropriate to one of these classes are called "irregular."
Now the problem is that these definitions of "irregular" are language-specific; the definition of "irregular verb" for Spanish is no use for Turkish. Therefore, the results that you obtain for one language may not be comparable to those you obtain for another.
I read your comment carefully, and it still strikes me as irrelevant; the study is about the rate of language change, not about mechanisms.
At any rate, the idea that child language is the primary motor of language change is hopelessly flawed. I'd have to look it up, but IIRC there's a lot of sociolinguistic research that casts doubt on it. Language change seems not to be driven by children, but by adolescents and young adults, who've long. (And the really big mechanism is phonological change, not morphological assimilation, anyway.)
Linguistics 101 lesson: a language is not a bag of words. Any generalization about language that treats it as if it is some bag of words (e.g., in this case, that language change consists of new words entering the bag, while other words fall out of it) shows a profound ignorance of the fundamental ideas of linguistics. A language is a grammar; people invent and adopt new words spontaneously all the time, but not, say, morphological paradigms, case agreement, or new forms of valence-changing rules like the passive or the causative alternation. (Yes, I'm using words that most people who read this won't understand, but that's the point--if you don't understand terms like that, your "insights" into laguage aren't very valuable.)
That sounds like a combination of myth and hyperbole about a perfectly ordinary language standardization process (e.g., the kind that happened in Spain during the reign of Alfonso X of Castile, and again after the publication of Antonio de Nebrija's grammar). I don't know what Korean king you're talking about here; my first thought was Sejong the Great, but the timeline is wrong (he lived about 600 years ago, not 400). At any rate, his great contribution was an orthography (Hangeul) that wasn't adopted until much later.
Are you aware of any historical linguistics research that makes quantitative hypotheses about the relationship between word frequency and morphological regularization? I don't know if there are any (and I wouldn't be surprised either way), but whether this study is "news" depends on the answer to that question, not on the all-too-well known fact that children learn regular inflectional paradigms before they learn irregular ones.
...that Rudolf Flesch doesn't know what the hell he's talking about.
TFA is short on details, but they must've used a historical corpus (that's linguistese for "a database of texts").
This of course raises the question of what kind of language the corpus is representative of, and what kind of language is is not representative of. The bodies of text we have for Old and Middle English are far less representative than what we have for contemporary English; e.g., we don't have a lot of transcribed recordings of phone conversations between family members from back in Chaucer's day.
Of course, it's kind of hard to criticise this study without looking at it. The only thing that strikes me so far: none of the authors seems to be a linguist.
That's the step where it goes wrong. Person B is still bound by the terms of the BSD license.
Let's simplify it a bit, and assume just one file. Person A obtains a file consisting solely of one person's work, licensed under the BSD terms. Person A makes significant changes to it, and releases a derived work to Person B, under the GPL license. Person B now needs two licenses to modify and distribute what Person A gave them:
The key point is that Person A can only enforce copyright on his GPL'ed work to the extent that the alleged infringing material is not based on the original, BSD-licensed file. For example, if B takes a function from A's GPLed file and puts it into a non-GPL program, A can only complain if that function didn't appear in the original BSD-licensed work.
Yes, Google does have funds. But guess what? Microsoft has considerably more.
You're assuming that if you somehow just get a bunch of smart engineers on the job, the solution will happen. Nope, it doesn't work that way in real life.