On Tue, 30 Apr 1996, Brian Pape wrote: > As I said, you've evidentally never heard a foreign language... You might > possibly be able to determine word seperation with a language that is > similar to english (although from your post, you would probably have a > tough time with a regional U.S. accent, let alone another language). In > reality, it is virtually impossible to distinguish one word from another > in a language that you are unfamiliar with, unless the speaker conciously > seperates one word from another to accomodate you. As food for thought > (if you're capable of intelligent thought?) find someone who speaks a > foreign language that you don't know, have them speak to you in a > conversational pace for 30 seconds, and then try to determine how many > words they said, along with any meaning you can derive (which you alluded > to in your previous off-base post). First off, you cannot assume you and I are in anyway similar with our linguistic abilities. If I can decipher word speration and you can't, that's one of the many differences between us; because you lack a particular skill doesn't mean that someone else cannot possess the ability AND, as I said, it's not a foolproof mechanism since some of the more intricate languages (in comparision with the languages I know) have too many similar sounding words and a consistent pace to actively determine when they complete one word and begin another. That's why I said, I can do it about 75% of the time, depending upon language, pace of the speaker, etc. Secondly, if my post was off base then all of your replies to this thread have been. You are leaving flame-bait all over the place. If your reply was not meant to be a flame, then why toss in insults? NEXTLY, your point was that it was impossible to determine pauses between words when someone is speaking at a conversational (ordinary) pace in a language that I don't know. This isn't true, it's possible, and while it's not foolproof you were posting an argument that was off-base and seemingly for the soul-purpose of starting an argument. > That was the point of this entire (useless) thread, comparing muds > languages to ``real'' languages. No it wasn't. The point of this entire (agreeable useless) thread was, to my knowledge, implementing languages to the most realistic but sensible extent on a MUD. While this involved discussion of real languages, it was not the point of the entire thread. My original point was that it is possible to figure out that the Spanish word "radio" sounds quite similar and is spelled exactly the same as it's English counterpart and therefore, someone that does not speak Spanish at all can logically assume the meaning (although the presumedmeaning may not be correct). I didn't get into language speeds or what not, someone else did and you decide to come along and drop in unfounded insults. > At any rate, I have no idea what you are trying to prove... I'm not trying to prove anything, YOU WERE TRYING TO PROVE SOMETHING, I was offering a counterpoint. You took it upon yourself to insult me in your postings. So, what are you trying to prove? It's completely pointless to get into discussions of speed and how languages are pronounced because, while it may not be impossible to implement, it's not sensible to write a whole new language just for the purpose of your MUD complete with grammar and pronouncation rules. Let me remind you, I made a statement, which you tried to disprove, thus, you were trying to prove your point (prove that what I said was false). And, I, sir, have no idea what YOU are trying to prove. > On my muds I > simply generate a random character string, with repeated random character > seperation to indicate that words are being spoken, but the receiver can't > understand them. When learning another language, we begin to randomly > integrate the garbled output with the actual text the character is > speaking, until they fully understand it. We however allow all characters > to speak common 100% as a default skill (actually we do have a few races > that have problems with this, but they are a special case), and give each > race their own language. Depending on intelligence and other attributes, > characters can learn multiple languages. Ah-hah, an intelligent suggestion. A very good suggestion. You could have simply got to this without the prelude, and I implore you to do so more often, > > I think the 'hello there' == 'hxxxo txhxxx' method is not particularly > advantageous for a muds. It's not the best of sollutions, but it is one of the easiest if you want to go on a semi-realistic pathway. > > The 'hello there' == 'dfjks dfh dla' method is very simplistic, but a > bit more 'realistic'. I don't see that as more realistic as randomly replacing letters with "x"s [I do like the idea of randomly replacing letters with other letters, BTW]. I say they're about at the same level. > > The 'hello there' == 'lother' method (which requires a smart algorithm, > along with some basic phonetic/phoneme databases for each language) can > be the most interesting, as it lets you approximate the sounds of > different languages to another language. Depending on the values you > choose, certain languages will be fairly puzzleable, while others will > essentially come out as garbage. This is a method that I have partially > written for my muds, and seems to generate a fair amount of interest. Sounds fairly interesting, although it also sounds like it's rather time consuming if you want quite a few languages.
This archive was generated by hypermail 2b30 : 12/18/00 PST