Hacker Newsnew | past | comments | ask | show | jobs | submitlogin

The combination of the ambiguous word "is" and the mixed capitalization styles in the title is interesting. When I first read the title, I saw it as:

   ReadingHackerNews.is_going_to_launch? # false
And wondered who the Reading Hacker News team was, and why they decided to scrap their launch. After a couple seconds I saw the forest for the trees and realized it said:

   ReadingHackerNews !== Launching       # true
Now, what's interesting is the interplay between:

   - capitalization conventions in human language
   - capitalization conventions in programming languages
   - the author's specification of [Reading Hacker News]
     in the article, that indicates that he is referring
     to it as an object in itself
   - the requirement for him to use brackets instead of
     merely case to indicate this, because he needed to
     capitalize "Reading", and "Hacker News" to be proper
     english anyway.
   - the way the brackets were left of in the submission
     title
   - the fact that programmers often use things like
     CamelCase in english to communicate concepts
     analogous to those found in programming, like
     object-ness.
   - the existence of a lot of third party HN tools, such
     that one of them could conceivably be called "Reading
     Hacker News"
There is a lot of stuff going on in this title for the brain to process. I wouldn't have noticed it at all except that my brain glitched and produced the wrong answer for a second or so.

It's crazy to think about (1) how many difficult background questions our minds silently get right, and (2) how many of these questions our minds silently get wrong.




"Garden path sentences are less common in spoken communication because the prosodic qualities of speech (such as the stress and the tone of voice) often serve to resolve ambiguities in the written text."

Another random and crazy thought: people have experimented with textual and graphical programming languages -- I wonder what an aural programming language would be like? It's almost certainly impractical, but the idea is intriguing nonetheless.

Maybe slightly more practical would be some kind of aural complement to conventional sight-based programming systems...


Not really, it's just ambiguous...


Reading "Hacker News" is not launching—ambiguity solved.


The convention is to quote article titles and emphasize site titles, so I'd recommend:

"Reading Hacker News is not launching"

Perhaps this is a good reason to use formatting in submission titles?


Come on, "Reading HN != Launching" is clearly the optimal solution here.


That's really only because English is hard to parse, in other languages this is unambiguous.


"Other languages" cover a vast range of possible and required levels of ambiguity. Distilling that to "English is hard to parse" is pretty inaccurate.


Greek, for example, is quite a bit less ambiguous than English. We don't have anything close to the number of homographs in English, and parts of speech are very distinct (a word can almost never be both a verb and a noun).

In English, "right" can mean any number of things, it can be an adjective, a noun, an adverb or a verb. In Greek, that would almost never happen.


I think you might be running into native speakers' bias. (I'll grant you I know nothing about Greek, just going off Wikipedia here, so forgive me.)

Greek is classified as a fusional language, essentially meaning that it's an agglutinating language with a more complex inflectional scheme that's harder for linguists to figure out. Naturally there's going to be less lexical ambiguity than in an isolating analytic language like English, since grammatical information we're used to inferring from context or marking lexically you're encoding as an affix. By contrast, we find your system of affixes complex and difficult.

The important thing to understand here is that a native English speaker treats words the way a native Greek speaker treats affixes. When we read a sentence, we automatically filter the words on the basis of the context, just the same way when you read a Greek sentence, you're automatically filtering affixes on the basis of context. The "ambiguities" you perceive — "right" can be an adjective, noun, or verb — are effectively the same as, say, an ambiguous -o suffix which marks a tense on verbs, but a case on nouns. For native speakers, they simply never notice.

(Of course speakers of all stripes make mistakes and need to clarify ambiguities— no matter how they're encoding information.)

And it goes the other way too. Take a native Californian language like Eastern Pomo, in which verb stems describe relationships of energy and motion which are difficult to describe in English. Copying from notes here: For example, the stem pa-qá-t(-ki-) has the root meaning "to apply pressure", the instrumental prefix implying from kinetic energy, and an intensifier, semelfactive and stative suffix. This stem means both "to pry abalones loose from rocks with an iron bar" and "for high flood water to come down a creek and wash something away." Remember, this is a polysynthetic language— nearly every word has this sort of variance in meaning. We perceive that as horrifically ambiguous, and nigh impossible to parse without concerted effort, but a native Pomoan would never mistake them.

So, you know... Cut English some slack :)


I see what you mean, but I disagree about the bias. I'm pretty handy with English, yet parsing some headlines is a challenge sometimes, like the poster above noted. I have yet to encounter parsing trouble in Greek.

The example you give about the -o affix, while valid at first glance, is actually not, simply because verbs end in omega while neuter nouns end in omicron. Thus, they could never be mistaken.

I'm not saying Greek is easier (approximately zero foreigners can make sense of our affixes, always choosing one and tacking it onto every word when they try to speak Greek), but I am saying it's less ambiguous :) For the record, I had no trouble parsing this title, and, in fact, was unaware of alternate interpretations until the poster above pointed it out.

Fruit flies like a banana!


Whoop, that's carelessness (or extraordinary good fortune?) on my part. I didn't mean to imply an actual Greek suffix, I was using -o as a theoretical example.

However, just glancing again at Wikipedia I see that "In Modern Greek Ω represents the same sound as omicron". So actually this is a perfect example of the native speaker's bias I'm talking about. You can't tell those suffixes apart in speech because of spelling, you can tell them apart because of context— but you're so used to conceptualizing them as fundamentally different, it didn't even occur to you that to a non-native they sound the same.

It's also worth noting also that headlines follow their own peculiar grammar rules which make them unusually prone to forming garden path sentences— lots of native speakers get confused sometimes too. ESL courses cover headlines specifically, since their grammatical structure is so opaque: http://www.esl-lounge.com/student/reading/4r46-headline-engl...

Don't take any of this as a criticism of your broader point— I'm willing to grant there are levels of ambiguity, and English with its highly lexicalized grammar — and especially Headline English with its nearly absent explicit grammar — is definitely up there on the scale. I just wanted to point out that no matter how we analyze them, all languages are natural to parse for native speakers and awkward to parse for non-native speakers— that's not related, or is only vaguely related, to their intrinsic ambiguity.


I definitely agree with your broad point, especially about headline grammar. I disagree on this being mostly a native vs non-native speaker issue, though. I'm sure you will agree that a language where most words can be used as multiple parts of speech is more ambiguous than one where they cannot.


Oh, like I said, I'll totally grant that if we're talking about lexical ambiguity. Words in isolating languages accumulate extra meanings, just like morphemes in inflecting languages do.

But again, that's just because of highly lexicalized grammar; it's practically no different from having an "ambiguous" system of inflection, or bizarre energy-focused polysynthesis, which a native speaker would never mistake. That's where I'm saying someone who grew up with a strongly inflected or polysynthetic language would be biased.

And I hope you don't take this as a criticism of your English— you're obviously perfectly fluent! There's simply a basic level of intuitive understanding of a language which is impossible to get without growing up with it from infancy. That's why in linguistics we don't talk about fluency as much as native vs. non-native speakers. It's where you can really separate the meat from the bone, if you will.


The word "is" is overloaded pretty hard...


Reading it again, it sounds like I meant "in all other languages", when I meant "in some".


This comment thread about language is awesome by the way, enjoyed reading it more than the post!




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: