From cbmvax!uunet!CUVMB.BITNET!LOJBAN Fri Jun 26 14:48:19 1992 Return-Path: Received: by snark.thyrsus.com (/\==/\ Smail3.1.21.1 #21.19) id ; Fri, 26 Jun 92 14:48 EDT Received: by cbmvax.cbm.commodore.com (5.57/UUCP-Project/Commodore 2/8/91) id AA11165; Fri, 26 Jun 92 04:59:53 EDT Received: from pucc.Princeton.EDU by relay2.UU.NET with SMTP (5.61/UUNET-internet-primary) id AA16973; Fri, 26 Jun 92 00:49:51 -0400 Message-Id: <9206260449.AA16973@relay2.UU.NET> Received: from PUCC.PRINCETON.EDU by pucc.Princeton.EDU (IBM VM SMTP V2R2) with BSMTP id 1484; Fri, 26 Jun 92 00:18:31 EDT Received: by PUCC (Mailer R2.08 ptf027) id 6002; Fri, 26 Jun 92 00:18:03 EDT Date: Fri, 26 Jun 1992 13:45:28 +1000 Reply-To: cbmvax!uunet!MULLIAN.EE.MU.OZ.AU!nsn Sender: Lojban list From: cbmvax!uunet!MULLIAN.EE.MU.OZ.AU!nsn Subject: Phone Game proceedings: Gleem X-To: lojban@cuvmb.cc.columbia.edu X-Cc: nsn@ee.mu.oz.au To: John Cowan , Eric Raymond , Eric Tiedemann Status: RO X-Status: Sorry I've been so late in publishing this; been having exams. The procedure with the Phone Game is that a phrase is passed around a circle of net.lojbanis, being translated in and out of Lojban, to see how far the meaning of the phrase wanders.) (Note: the order of responses made was: Lojbab, Ivan, Mark, Ivan2, Sylvia, Colin. Responses with ']' were made during the game.) Colin was started off with the following phrase, which he himself selected (though he intended it to be subjected to some other wight): "9 out of 10 of the housewives we asked prefer Gleem to their ordinary washing powder." He came up with: sofi'upano le zdasazri poi mi terpreti foke'a cu maurzau lai glim. leike'a fadni terlumpu'o and comments: ]"9 out of 10" was amazingly hard. "piso lei" is obvious, but somehow ]doesn't seem right for this meaning. I was trying for something like "so ]leropano" (which I think actually means "9 of all 10", rather than "9 of ]each 10" as it should"), or "so leroboi panomoi" (yes, a non-elidable ]"boi"!), but then I realised that "9 of each 10" is not actually true - ]the "9 out of 10" is aggregate. I take it this means that, if you pick out any ten women (in fact, all decuplets of 'em), it is not true that precisely nine of them will use Gleem? Well spotted. Ivan>Well, it _can't_ be true in the first place. Are engineers taught Ivan>this kind of thing? :-) mark>Huh? Where would you fit in the {panomoi}? That's a selbri, so it'll form mark>a tanru. Would you have {so le ro zdasazri panomoi}? ("9 of all mark>house-operator 10tuples"--I can't imagine you'd want {panomoi zdasazri}, mark>that makes even less sense) I think that's a bad plan. --- Ivan>It is wrong. If {re le} means `two of the', not `twice as many', Colin>I didn't say "piso le" I said "piso lei". Ivan>then {piso le} means (in this case) `.9 of a house operator', not `.9 Ivan>of the entire quantity of house operators'. {pimu le pano cukta} is Ivan>half a book, not five books. mark>Yes, but. Note that he has {piso lei}, not {piso le}, thus meaning ".9 of mark>the specific mass of house operators", which may therefore be more than mark>one. Ivan2>He has {sofi'upano le}, which is the same as {piso le}, in his fyky Ivan2>sentence. mark>In fact, if you recall the quantifier discussion of earlier, you'll mark>remember that {le} has a default "internal quantifier" of {su'opa}, while mark>{lei} defaults to {ro}. So {piso lei} isn't so bad as all that. Ivan2>Then {re lei} makes no sense, as you can't select quantity 2Q out of Q. Ivan2>(If {lei cukta} refers to 10 books, {relei cukta} would be 20 of them.) --- mark>Hmmm. Now I'm less sure. Even with {lei}, had I seen {re lei}, I'd have mark>thought "two [house operators] out of the mass of...", and thus by Ivan's mark>argument, I'm forced to agree that {piso lei} should be "nine-tenths [of a mark>house operator] out of the mass of...". Ack. Then what good is {lei}? --- Ivan>{lero gerku} `both dogs'. {leropano zdasazri} `all ten house operators'. Ivan>{so leropano zdasazri} `nine of the ten house operators'. There were Ivan>exactly ten of them, and nine preferred Gleem. mark>I think {so le ropano} is probably good: there are exactly 10 mark>house-operators under consideration here, and 9 preferred it. Maybe {so mark>le'e ropano} would be even better. I suspect so. Ivan2>There aren't exactly 10 house-operators, that's the catch. We may Ivan2>have surveyed 1000, and found that 900 preferred Glimm. ]I settled on the fraction - I think we are too ready to use decimals an ]not ready enough to use fractions - they have a different kind of ]precision about them. In fact, of course, the chances are that in this ]case a decimal might be a truer claim, but that is not the claim in the ]English." Ivan>Why? As I understand it, since (mathematically) 9/10 = .9, Ivan>{sofi'upano} means exactly the same thing(s) as {piso}. Nowhere does Ivan>it say that one type of fraction may be less precise than the other. mark>Gotta stick with Ivan here. If fractions aren't good, neither are mark>decimals; they're just alternate representations of different things. Colin>NO NO NO NO NO NO!!!! Colin>This is elementary metrology. In arithmetic, 9/10 = .9, Colin>but in physics, or applied maths, or the real world, they are not the Colin>same, because they imply different standards of accuracy. As I said in Colin>an earlier mail, all measurements (and hence all numbers used as Colin>quantifiers) have an express or implied accuracy. If Lojban is not to Colin>reproduce the scientific implications of using decimal and vulgar Colin>fractions, then we must state explicitly what accuracy is to be applied, Colin>and will need to use my "accuracy" operator a lot more. [The remainder of the discussion on accuracy was conducted on Lojban-list]. ]"housewife" was hard as well. It looks odd, but makes sense. Ivan>Not to me. I interpret {zdasazri} as `chief butler'. You know, Ivan>someone you have hired to take care for your domestic economy. mark>I'm not so sure it's the greatest either, but on the flip side, I can't mark>think of a good alternative. Colin>This is our (your and my) argument over precision again. I agree that Colin>"zdasazri" can mean "chief butler" - I did not see it necessary to Colin>distinguish the kind of house-operator. The {ke'a} there is wrong; it has no referent. It should be {vo'a}. Sylvia: Nine out of ten housekeepers who were questioned preferred Gleem to their ordinary wash-powder. Don't you hate it when these games work? :) "housewife" has been succesfully desexed into "housekeeper", Nick notes. Okeydokey. The questioners are no longer "us". Ooh, Sylvia, licence! :) Lojbab would at this point demand I specify the goals of the game so as we can know whether Sylvia did a Good or Bad Thing. Someone relieve me of the moral dilemma. :) Sylvia>I agree. I considered the 'we asked' construction, but decided that Sylvia>the passive voice 'who were asked' sounded more typical of this kind Sylvia>of English. "zdasazri" suggested to me both 'housewife' and 'house Sylvia>operator'. I don't remember how I decided to use 'house keeper'. As quick as someone reasonably quick, Nora relays: fi vei so pa'i pano le zdaku'i poi ve preti cu zmapluka fa la glim. le cabna lumpu'o pe pa'aku Now let me see this. In {vei so su'i pano}, you take 9, you take 10, and the answer is what you get when you add 'em, and that's how many housekeepers you get. So you take 9, you take 10, you take their ratio... and is the result something you can put in front of "housekeepers" and have make sense? If it comes out of Fairfax :) Ivan>Well, {vei so fe'i pano (which is the same as {piso} or {sofi'upano} Ivan>le zdaku'i} is what you'll get if you choose one of the housemaids, Ivan>take a scimitar and hack off one-tenth of her (about 6kg). Hopefully Ivan>{vei so pa'i pano} is npot the same thing. Seriously, I'm a bit worried about this. The syntax of {pa'i}, a VUhU rather than a GOhA, imply that a ratio is a number. It's just that I think of a ratio as a predicate, a relation. In either case, the meaning of a ratio as a quantifier is problematic. I look up my World Book Dictionary: "1. the relation between two numbers of quantities expresses as a quotient [...] 2 a quotient expressing this relation". The former is a predicate; the latter, the number 0.9 . The former looks wrong, but is defaulted out of by syntax. The second still makes me uncomfortable. (Lojbab comments: >The intent in putting pa'i into VUhU was specifically for the "9 out of 10" >case. It allows you to express a ratio as a quantity, without necessarily >claiming that there were exactly 10 housekeepers asked, and that 9 out of them >preferred Gleem. The fraction 9/10 might work as well, but then why does >English not say "9/10 of all housewives". So we put in a quantifier >equivalent construction that mimics the distinction - semantically identifying >it as a ratio by the choice of cmavo. > >Is this good? Who knows? "housekeeper" is now {zdaku'i} - we are getting closer to housemaid and further from housewife. Nora's lujvo always interest me :) . lumpu'o is less finicky than terlumpu'o. The place structure of zmapluka - x1, more than does x2, pleases x3? I'm a tad uncomfy with this, but so be it. Why the {pe pa'aku}? The ordinary (now current, in a bit of behind-the-scenes interpretation) washer-powder is an "also" with respect to Gleem - how? Mark takes us back from maids to housewives: Nine out of ten homemakers surveyed preferred Gleam to their current washing powder. ]Needless to say, I can't know the spelling of the name; one has to guess. ]There is (or maybe just was) an American brand of toothpaste called Gleem, ]so who can tell? Ivan>He's worrying about the spelling! Why, what do you know about the Ivan>pronounciation in the first place? Why is it "Gleam" or "Gleem" Ivan>[gli:m], not "Glimm" [glIm]? Long live la'o, I say. mark>Heh. Because I "happen to know" that the sentence began its life in mark>English, as something vaguely idiomatic. I can picture a washing-powder mark>called "Gleam", but not "Glim", in an English-speaking market. mark>Translation: I cheated. I used external, culturally-dependent information. Ivan>"Nine out of ten" may mean lots of things. Say, among the homemakers Ivan>there were ten whose names began with an "A", and nine of them Ivan>preferred Gleam. ]This one was very easy, more straightforward that the usual. 'Course, ]there's some idiom going on here; "Nine out of ten" doesn't mean the same ]as "Ninety percent", but the Lojban shouldn't make the distinction, though ]I read it in, seeing as how I and Nora are both English speakers. This is ]similar to my use of {citno nanmu} in the first one. I don't think the distinction need be made either, but if it can be, Nora's manner is more effective than Colin's. Ivan, finally: lu'a pino lei lazyzda sazri poi ve preti cu jinvi leka xagmau be le ta'urlumpu'o poi cata'e pilno ke'a kei la'o gy. Gleam gy. Yes, Ivan, {pino} should be {piso} :) I suppose {lazyzda sazri} is indeed an appropriate translation of "homemaker", given "home"'s familial connotations. I see Ivan uses {lu'a} to convert the mass of homemakers into individual people; personally I don't think this any more necessary than, say, {pa'i}, but it's at least as elegant. Ivan specifies the washing powder to be for clothes, which is quite proper, and maybe better for the dictionary. mark>More than maybe. "Washing powder" could mean abrasive sand you use to mark>clean off the bottom of your boat, or any of a million other things. I'd mark>interpret it as a powdered type of soap, but with no judgement as to what mark>it cleans. Hell, we don't call the stuff you use to wash clothes "washing mark>powder" here in 'Merica much; commercials all refer to it as "laundry mark>detergent". Ivan2>True. But the road to Lojban will go through `washing powder' anyway. --- Ivan>Lest [the washing powder] is understood as powderised soap or shampoo Ivan>for people. mark>Nothing wrong with non-specified "washing powder", just realize that you mark>can't count on people to understand that it's for clothing. I note that Ivan has a clefted x2 of {jinvi}: x1 thinks x2 of x3. This is what appears in the '88 list. In the '90 list, this is uncleft: x1 thinks x2. This is, of course, a simple transformation; and the intended meaning is pretty obvious, in any case. I'm not sure about Ivan's phrase. On the one hand, it's kinda stiff. On the other, it gives lots of useful information which the others have neglected. In the interests of a chaotic game, I welcome this dichotomy :) So then: we started with: "9 out of 10 of the housewives we asked prefer Gleem to their ordinary washing powder." - a fairly stereotyped phrase, readily reconstructed, I suppose - - we went through all sorts of "9 out of 10" expressions, which was instructive - and ended up with: Those who are 0.9 of the family home operators, and who were asked, think of "Gleam" that it's better than the clothes-washing powder that [they] currently usually use. Cool. --- 'Dera me xhama t"e larm"e, T Nick Nicholas, EE & CS, Melbourne Uni Dera mbas blerimit | Mail: nsn@munagin.ee.mu.oz.au Me xhama t"e larm"e! | Omiloume ellhnika/Esperanto parolata/ Lumtunia nuk ka ngjyra tjera.' | {mika'e tavla baula lojban.je'uru'e} - Martin Camaj, _Nj"e Shp'i e Vetme_ | (Better .sig suggestions welcome)