>Agglutinative languages, like Finnish or Hungarian, where word order is largely irrelevant and conjugations, not word order, determine meaning.
I'd argue that word order is just as relevant (and indeed constrained by meaning) in Hungarian or Finnish as it is in, say, English, though the details of how each language's syntax works are very different. Also, there are plenty of languages not usually characterized as "agglutinative"(e.g. Russian) in which affixes rather than word order are the primary determinant of the role that a word plays.
"Agglutinative" is really just a way of saying that meaning-bearing morphemes in the language tend to be more closely bound together: where in English, for example, we'd express "in our houses" with 3 words, whereas in Hungarian, these would form a single word "házainkban" (ház[a] "house", -i- plural, -nk "our," -ban "in"). We know they're separate words in English because other stuff can come between them ("in each of our big houses"); not so in Hungarian. (So arguably agglutinativity is more about constrained morpheme order than it is about free word order!)
>Topic/subject languages, like Japanese, where previously defined context can radically alter meaning. "Watashi wa hamburger desu" can mean "I'm a hamburger", "I'll have a hamburger", "If you ask me, it's a hamburger", etc.
Of course context can radically alter meaning in English too; think of the waiter who says "which one of you is the hamburger?" ("Me, I'm the hamburger, and he's the Caesar salad.")
Almost all language classifications exist along a spectrum, or perhaps it'd be better to talk about a high-dimensional space. Different languages rely more or less on different strategies, but there are almost no linguistic features for which analogues can't be found in most languages.
> I'd argue that word order is just as relevant (and indeed constrained by meaning) in Hungarian or Finnish as it is in, say, English, though the details of how each language's syntax works are very different.
I agree; GP didn’t choose the best examples for this. But there are languages like Yimas, Tiwi and Dyirbal for which word order truly is insignificant, so the point still stands.
Latin would be an example that more people might be familiar with.
There's a difference between "free ordering of constituent phrases", where you may rearrange subject noun phrases, object noun phrases, adverbial phrases and the verb into any order depending on emphasis etc., and completely free word order - like in the languages you mentioned - where "the little man walked through the big door" could end up as "man door through walked big little the the" or something like that, completely breaking apart what we'd consider "constituent phrases" in a language like English, where "the little man" etc. always move together in a sentence.
You’re completely right — indeed, Latin would have been a much better example than the ones I chose. (But I had been reading about Yimas a few hours before posting and it was on my mind.)
> There's a difference between "free ordering of constituent phrases", where you may rearrange subject noun phrases, object noun phrases, adverbial phrases and the verb into any order depending on emphasis etc., and completely free word order - like in the languages you mentioned - where "the little man walked through the big door" could end up as "man door through walked big little the the" or something like that, completely breaking apart what we'd consider "constituent phrases" in a language like English, where "the little man" etc. always move together in a sentence.
But how common is it to have the former without the latter? Most languages I read about have either both or neither.
> But how common is it to have the former without the latter? Most languages I read about have either both or neither.
Many language seem to have at least a somewhat flexible ordering of constituents. It may often depend on emphasis, information structure, etc. German for example can be quite flexible in that regard, and the effects this has on information structure are rather subtle (however, the verb always has to go in the second position, so it's not totally free). There are probably languages where it's even more free, but I don't have any examples off the top of my head. By contrast, there are languages where word order is very rigid (even in English, while you can rearrange constituents up to a point, it's not nearly as common as in German, and I would argue that it can often sound slightly archaic)
archaic might have been a bad choice of words, although it probably depends on the exact sentence. I do think it sounds more formal, though, I think I'd be less likely to hear your first sentence in everyday conversation. Then again, I'm not a native speaker of English.
My native-speaker intuition tells me the opposite — I’m probably more likely to use such constructions in colloquial speech than in formal written text. My second sentence does seem more formal to me though.
The problem with "they" is that it is in fact a very bad translation for "ő" in a sentence like "ő mosogat." We typically find gender-neutral "they" in English when there is an indefinite antecedent like "someone" or "anyone" - "if anyone objects, they should talk to me." In contrast, in Hungarian, including the third-person pronoun "ő" means the speaker has a very definite subject in mind and in fact wishes to emphasize that (a neutral "he/she is washing up" would be just "mosogat", without any explicit pronoun; "ő mosogat" almost means "HE [not someone else] is washing up." (Or "SHE" of course.) The only case where I can imagine translating "ő" as "they" in this kind of context is when I know that the person in question explicitly prefers the pronoun "they," e.g. because they identify as non-binary.
> The problem with "they" is that it is in fact a very bad translation for "ő" in a sentence like "ő mosogat." We typically find gender-neutral "they" in English when there is an indefinite antecedent like "someone" or "anyone" - "if anyone objects, they should talk to me."
There is a school of thought that says we should change how we use the English language. This school thinks that the proper way to speak is to assume non-binary until informed otherwise and that appealing to "what sounds right" or "the rules of grammar" is unacceptable linguistic conservativism.
I am not making that argument. I'm just raising it because it seems relevant here. I would point out that, even if you disagree with this argument, it seems reasonable to think that people should be able to choose how they use pronouns, whether their usage flies in the face of linguistic convention/rules of grammar or goes along with them.
I absolutely agree that people should be able to choose how they use pronouns (or in fact any component of language). Communication is a mutual endeavour, and I will always make my best good-faith effort to understand what an interlocutor is trying to convey (assuming I want to talk to them in the first place :), even if they choose to use language differently from the way I would. In fact (within reason) I will even adapt my own usage in some cases - e.g. I will do my best to use the pronoun "they" (or "xe" or "foobar" or whatever) when referring to someone who wishes to be referred to in that way, even though (despite knowing several languages with gender-neutral pronouns) I find that takes non-insignificant mental effort when I speak English.
That said, precisely because communication is a mutual endeavour, and language is a coöperative framework, there is no guarantee that -- just because you choose to use "they" in this way -- others will follow. I wouldn't be terribly surprised if, within a generation, "they" became an unmarked definite gender-neutral pronoun in English (today it is arguably unmarked in the case of an indefinite antecedent, but extremely marked otherwise). But I wouldn't be terribly surprised if it didn't, either. Language change is an emergent process and not something that can easily be forced in either conservative or radical directions.
I completely agree with you. I'm not arguing for anything to be forced, only for individual users to be able to choose the pronoun rules that a translation program uses.
I was amused to find that "ő orvos. ő sebész. ő agysebész. ő kardiológus. ő pszichiater. ő radiológus. ő bőrgyógyász. ő nőgyógyász. ő szülész. ő fogorvos." was translated as "she is a doctor. she is a surgeon. he is a brain surgeon. she is a cardiologist. she is a psychiatrist. he is a radiologist. she is a dermatologist. she is a gynecologist. she is a midwife. she is a dentist." The (small) majority of physicians in Hungary are in fact female -- I don't know how that breaks down by specialty though. If you are to believe Google Translate, only brain surgeons and radiologists are male by default! (By the way, "szülész" to me means "obstetrician," not "midwife.")
Pronounced, originally, to rhyme with "honey" and "money". The obvious vulgar homophony explains why the word is obsolete today and, when it is used (as, for example, in the toponym "Coney Island"), is invariably pronounced to rhyme with "phony" and "bony." Mind you, the pun would already have been perceived in the original Latin (cuniculus/cunnus) and indeed this pair may ultimately be related; cuniculus (whence also conejo, coniglio, coelho, Kanin[chen], etc.) meant not just "rabbit" but also "rabbit hole."
>any semantic similarity to the K2 GUI system is probably coincidental.
Hardly coincidental! I very much had the K gui in mind when I originally architected it (and I'm not sure I'd describe it as a "reasonably conventional web application"...)
Oh you may be right. I couldn't see the pattern right away though. Is it memory resident databases, and it is a pun on them not really being considered "databases"...?
iambic keying is clever, and fun, but of very little practical value.
True. Also, amateur radio is clever, and fun, but of very little practical value. I've been a ham for almost 30 years. I like to compare it to sailing. If your goal is to get some place fast, you should choose just about any other conveyance besides a sailboat. Messing about with radios is loads of fun, but practicality is rarely a major consideration.
In a situation like this, the key is to get a contingency in the house purchase contract -- and have the work done before the closing.
We bought a house that didn't have cable or DSL in a somewhat rural area -- the sellers were using 3G, satellite Internet, and dialup (!). They assured us that the only reason they hadn't had cable installed was the high cost they'd been quoted for running cable on the property (understandable as the house is almost 1/4 mile back from the street), and the cable company concurred that they'd be able to do the work, "no problem," and quoted a fee that didn't seem all that bad in the end (we took it into account when negotiating the price of the house).
Nevertheless, we made the sellers agree to a contingency - cable co. would successfully complete the work (on our dime, of course) and hook us up before the closing, and the total amount wouldn't exceed what we had been quoted. If the contingency wasn't met, the sale would fall through and we'd get our deposit back.
That contingency clause relieved me of a lot of stress over the next few months. Everything turned out OK in the end, but it was touch-and-go for a while.
Cable co. accepted the order with no problem, did a survey I guess, called us and said they wouldn't be able to do it. They were persuaded (not sure how -- agent took care of it) to reconsider, but then came back with a quote that was at least five times higher than before. Cue lots of emails and phone calls. It was helpful having an agent who lived in the community. In the end a local contractor dug the trench for us at a very reasonable price, cable co. came down on their estimate, and the work actually was completed at just under the original quote. The dirt was literally being shoveled back over the trench on the day of the closing.
As I recall it was actually a few days later that the cable guy came to do the inside wiring and "flip the switch," so strictly speaking the contingency (cable installed and working) wasn't satisfied by the closing, but we assumed we'd be OK at that point, and there were in fact no further hitches. We moved in and have enjoyed a (mostly) reliable, fast connection ever since. But we would absolutely have walked away, even though it was our "dream house," had we not been able to get the cable pulled. The contingency clause gave us the assurance that we could do that with no penalty. And, had it turned out that the work could be done but only at 5x the price, we could potentially have used the clause to negotiate a concession from the seller to cover the costs.
In this instance, the buyer was 100% convinced that the property had the service of interest and was told it did by the local monopoly in charge of providing such service. There was no work to be done before the closing because Comcast won't hook up Internet to a property for someone who isn't the property owner or resident.
What you're describing would be a bit like taking out a contingency on the house actually having running water after you physically walked through the house and checked the taps and toilets, just on the off chance that the water is actually not coming from the local muni water source, but instead from an on-property aquifer that a previous owner had tapped and painstakingly routed into the house's internal plumbing, and immediately after closing the sale that aquifer runs dry.
Real people don't go to that level of contract detail.
Well, at least he will know to go to that level of detail next time! Once burned, twice shy.
I guess my point is that being "100% convinced that the property had the [feature] of interest" based solely on a phone call is pretty risky when you're talking about a purchase as large as a house and a feature as important (to him) as ability to get broadband internet. I totally understand and sympathize with his plight (I was screwed in basically the way, actually, many years ago, though it was a rental rather than a purchase, and DSL rather than cable, and it did work out in the end in that case too). And the degree to which he was jerked around by Comcast here is almost unreal. I could feel my blood boiling when I read the story. But in the end, I don't think he is going to have any actual legal recourse. He said this was his first home purchase; I imagine next time he will insist on seeing the connection in action next time, or getting that contingency clause in if not!
In any case, your analogy doesn't really fit the situation. The seller didn't have cable, or, presumably, any fixed broadband internet service. (That in itself is a bit of a red flag; I'd think most households who can get internet service at a reasonable price these days already have it.) The situation you describe -- well, it's hard to imagine it happening, because the seller would clearly have been legally obligated to disclose that the water was from a well and not municipal, and in any case the buyer would, one hopes, have hired a home inspector who would have noted this (and probably detected issues with the well as w... umm, in addition). But it doesn't sound like the seller hid the fact that there was no internet service currently at the house, which means there's no liability there.
So this is more like buying a house with no running water at all, with the knowledge that this is the case, and then getting a verbal promise from the local water company that they'd be able to hook you up. If you were to buy a house under conditions like this, I maintain that you probably would want a contingency clause on installation of the water.
And for what it's worth our cable company (not Comcast, but another big one) had absolutely no problem starting the work (once they agreed to do it) for us before we were the owner or resident. If they had, we would simply have made the contingency something like "seller is obligated to have internet installed, for which the buyer will pay, prior to closing."
(1) We are looking for someone (v. 'Infrastructure Engineer') who'd be excited to take on the challenge of helping to run, and ultimately running, a rapidly expanding cluster of hundreds of high-performance servers at several datacenters. The environment is pretty unconventional (99.4% proprietary software, for example, and we prefer to use an "exotic" language - K - even for infrastructure purposes); I'd say it's much more comparable to academic/scientific clusters than to your typical web application company. So that kind of background wouldn't hurt! At the same time, though, you need to know Windows, 'cause we don't use Linux yet, and you need to know Linux, 'cause we will sooner or later, and you need to be really au courant on the standard datacenter stuff (networking, firewalls, security, backup and replication, racking hardware, receiving -- and making -- urgent phonecalls at inconvenient times, etc.). As you can imagine, this is a bit of a hard job to fill... you need to be highly experienced (because we need your experience to support the serious growth we're in the middle of) and yet have an extremely flexible mindset (since we do things in such an atypical way). But if you're the right person to fill it, the rewards will be substantial. Be the guy in charge of hundreds of some of the hardest-working servers out there: 1010data is the fastest analytical database on the planet, and our customers are pounding the cluster 24/7...
(2) We are also looking for a 'Web Application Developer'. But again, the dry job posting belies the fact we need something a bit unconventional. What we really mean by this is a hacker who just happens to really love hacking in JavaScript. This is, I sense, a rare combination. But it does exist (we have verified examples at 1010data). We are developing cutting-edge browser-based interfaces to aforementioned fastest analytical database on the planet and since JS is the Language of the Browser... well, that's probably why you, JavaScript Hacker, chose JS. Right? Oh, you say it's because it's kind of an awesome language in its own right? OK, well, whatever the reason: if you hack JS and want to develop cutting-edge browser-based interfaces for manipulating and visualizing large datasets... please, please apply for this job. You're going to love it at 1010data.
(3) We are looking for a 'Systems Developer'. We're not 100% sure how to define this, to be honest, but to paraphrase Justice Stewart, we'll know you when we see you. You need to know a lot about Windows internals, but ideally also Unix/Linux, since one of the major projects you'll be involved in will be a gradual environment shift. You'll be diagnosing performance issues. You'll be trying to wring more speed from our already very efficient cluster. You'll be writing code (bonus! in an exotic language!) to move data around, to do logging and performance reporting, and who knows what else. You're going to be the guy we all go to when it comes to the low-level arcana, so you're very familiar with the Way Things Work. You know who you are. Let us know too.
If you think any of the above is you... then write to jobs@1010data.com and mention that you saw Adam's post at HN.
1010data, by the way, is a fantastic place to work. We've got a whole floor in a grand old midtown building populated with a small but growing bunch of very dedicated, very smart, very happy people. We're growing fast, so there's a lot of energy, and you'll be working hard, but what you do will matter. No one is doing superfluous work at 1010. Your stuff will be used. You get all the startup excitement, but without the startup risk - 1010's a well-established company; we've been around since before the turn of the century. Which, these days, is almost as long as it sounds!
A number of employees have one or more kids. Off the top of my head, they live in New Jersey (commute on NJ Transit), Long Island (LIRR), upstate (Metro North), i.e. the traditional NY suburbs, as well as in Manhattan and Brooklyn; that is, pretty much the whole spectrum is represented. In general it's safe to say that living in Manhattan is always going to involve compromises. Manhattan apartments are small and expensive. There's a strong tendency for families to move out to the suburbs, where there's greenery, space, and good schools. This is helped by the relatively good commuting infrastructure; it takes some of the NJ guys less time to get to the office than it takes me to get in on the subway from our house in Brooklyn.
I'm a New Yorker, but I'm not a real estate maven, and it'd be hard for me to suggest specific neighborhoods without knowing the specifics of your situation. Renting or buying? One- or two-income household? What do you do, what does your spouse do? And so on. But there are lots of options, and most of them are represented by at least one person at the company, so if you come and interview you can ask around :-)
I'd argue that word order is just as relevant (and indeed constrained by meaning) in Hungarian or Finnish as it is in, say, English, though the details of how each language's syntax works are very different. Also, there are plenty of languages not usually characterized as "agglutinative"(e.g. Russian) in which affixes rather than word order are the primary determinant of the role that a word plays.
"Agglutinative" is really just a way of saying that meaning-bearing morphemes in the language tend to be more closely bound together: where in English, for example, we'd express "in our houses" with 3 words, whereas in Hungarian, these would form a single word "házainkban" (ház[a] "house", -i- plural, -nk "our," -ban "in"). We know they're separate words in English because other stuff can come between them ("in each of our big houses"); not so in Hungarian. (So arguably agglutinativity is more about constrained morpheme order than it is about free word order!)
>Topic/subject languages, like Japanese, where previously defined context can radically alter meaning. "Watashi wa hamburger desu" can mean "I'm a hamburger", "I'll have a hamburger", "If you ask me, it's a hamburger", etc.
Of course context can radically alter meaning in English too; think of the waiter who says "which one of you is the hamburger?" ("Me, I'm the hamburger, and he's the Caesar salad.")
Almost all language classifications exist along a spectrum, or perhaps it'd be better to talk about a high-dimensional space. Different languages rely more or less on different strategies, but there are almost no linguistic features for which analogues can't be found in most languages.