From nellardo@xxxxxxxxxx.xxxx Thu Dec 30 14:34:05 1999 X-Digest-Num: 324 Message-ID: <44114.324.1765.959273825@eGroups.com> Date: Thu, 30 Dec 1999 17:34:05 -0500 From: Brook From: "And Rosta" > > > From: "Bob LeChevalier (lojbab)" > > > > At 04:09 PM 12/29/99 +0000, And Rosta wrote: > > >My response to the Top-Down idea of IAL or Lojban adoption > > >is to wonder why it should be a good thing for the adopting > > >body? Take the European patent organization: it would be > > >a trivial task to develop a language that shares Lojban's > > >virtues of nonambiguity and other areas of suitability to > > >the formulation of patents but is much simpler and easier > > >to learn; > > > > Really? If it were so easy, why haven't they done so? > > Either because they haven't perceived the need or because some > cost/benefit analysis doesn't justify it. I'd be surprised if a roomful of patent lawyers *didn't* recognize the need for clarity in language. The intelligent ones would realize that much of the ambiguity in patents is due to the vagueness of most human languages. Whether or not they would go from there to the concept of another language, I don't know, but I suspect most of them are familiar enough with the concepts of formal semantics (at least, the ones in software patents would be). And the cost/benefit analysis is what you're *both* talking about. However, at the risk of putting words in peoples' mouths, it seems that: a) lojbab thinks the cost of creating an IAL is high, the cost of switching to one might be high (but is lower the more established the language is) and the benefit of using one might be high, depending on the application. b) And Rosta thinks the cost of creating an IAL is low, the cost of switching to one is high, and the benefit of using one is probably low, especially for patents. Lojbab is suggesting that using an existing IAL (like lojban) reduces the cost dramatically. I think And would agree, but that doesn't seem to matter, as And feels that the cost is low to begin with. We all seem to agree one way or another that the cost of *finding* a suitable IAL is low (whether that involves using lojban (for lojbab) or creating one (for And)). Switching costs are discussed below. Right now, let's get into utility: > > Personally, I don't think you can get much simpler than > > Lojban and still do the job. The primary extraneous feature > > of Lojban not applicable to patents is the attitudinal/evidential > > system. Even audible unambiguity has some value. > > There is nothing relevant that Lojban can do that standard > predicate logic notation can't. In a Polish/Reverse Polish predicate > logic notation you need nothing but predicates, variables, > one or two quantifiers and two or three connectives. This is true in the sense that this is all you need to be Turing-equivalent. Actually, you don't even need that much - the Lambda calculus is Turing-equivalent and all it really has are lambda expressions (functions) (which are just lists of other lambda expressions) and applications of those expressions (i.e., calling a function). But this neglects an important part of *expressiveness* - just because something *is* Turing-equivalent, does not mean it is an effective way to express it. Minimize too much and you make it cumbersome and long-winded to say anything useful. Put too much in, and you've made somethign too complex. Finally, looking at predicate calculus (or *any* calculus, for that matter) is misleading because it ignores the question of *vocabulary* (a point lojbab makes below). Try programming in Prolog with no libraries. Sure, you can write quick-sort in a couple of lines, but what are you going to *do* with it? And how do you map it to the real world? Lojban has that in the form of gismu - predicate calculus does not. > In other > words, setting aside how variables are handled, you could have > a language with only 3 cmavo! I'll admit that that number might > be expanded a bit, e.g. to include numbers, but even an expanded > cmavo inventory would be only a tiny proportion of Lojban's. A perfect example of going too far towards minimalism. Sure, AND, OR, and NOT are all you need to make any other truth function but you better believe people that build ICs for a living don't recreate a one-bit adder from AND, OR, and NOT every time they need one - they don't even use one-bit adders - they pull more useful things from a library - 32-bit add, multiply, etc. and a whole lot of other stuff. > Likewise, the entire syntax could be formulated in a single > sentence. Yeah, and how would you say the classic "Colorless green ideas sleep furiously"? Bet it would be a lot longer. A lot. > > > logicians have been using such languages for decades. > > > > 1) What language have logicians used that could be used for writing a > > patent description? Key here is "description", and description takes > > meaningful content words. Patents include both things and processes, and > > both have to be describable, hence tanru and description sumti both > > requiring content words and both capable of being disambiguated > > semantically to an arbitrary degree of specificity as well as > > grammatically. Here's lojbab making the vocabulary point I referred to above. > Of course the predicate words' senses have to be defined. But in Lojban the > predicate words' senses are not defined -- this task has been left to > 'usage' > to achieve. What? Okay, lojbab is hard at work on a dictionary (in his, I'm sure, copious spare time) but gismu are as well defined as any word you care to name in any language - better, in fact, because they only ever have *one* definition (try looking up "run" or "fork"). > > 2. The language of logic that most people have seen is the predicate > > calculus. Being a reasonably bright sort of guy who struggled to barely > > pass a self-paced college level course in the stuff, I daresay that many > > would call the predicate calculus easy to learn. > > Is that irony? If so, I guess that they problem with predicate calculus is > that there's no fudgeability with it, which nonfudgeability is exactly why > one wants a logical language. Note also that predicate logic is a subset > of Lojban, so if you learn Lojban you learn predicate logic plus a load > of extra stuff. Mmm, I agree with your content, but not your point. Having learned lojban, I would agree someone could then learn predicate calculus pretty easily (though I'd bet a "native" lojban speaker would have a hard time seeing the point of predicate calculus (but then, I believe in Sapir-Worff)). But learning predicate calculus is not the same as learning a language. Kids soak up language without having to be taught. No one does that with predicate calculus. Now, unless you believe that "subject", "verb", and "object" are hardwired into brains, I'd submit that a young child exposed to a fluent lojban speaker could pick it up easily enough (I'm not fluent, but my three-year-old daughter seems to get the hang of lojban easily enough). [...] > Why? Most of the people who invent IALs are total lunatics, and most of the > rest are either ignorant or dim. Please - go say that on Auxlang - I want to watch you get toasted. [lojbab talks about how having an existing body of speakers speeds adoption of a language) > I'm not sure what point you're making. I agree that there are these > obstacles to the adoption of Lojban. And as I've said, I think Lojban > and Esperanto would be poor choices for a patent language, or for a > European IAL. I would agree on use of either for an IAL - IMNSHO, YMMV, etc. But for patents, I think having a common, unambiguous language would be a boon. The biggest barrier is probably patent lawyers, who have spent years learning how to deal with the ambiguity in a natural language. > > (I think Lojban has the > > advantage that it needs a lot smaller number than other conlangs to > > achieve critical mass, because Lojban unlike most conlangs DOES have > > the sort of specialty application like patent law and computer-communications > > that is economically viable with only a small fraction of the world learning > > it. And economic viability is the key to "top down" - a top down > > approach will work when someone with power sees a way to make money using > > the language. > > I very much doubt that this will happen, though it happening is Lojban's > only real hope for achieving critical mass. Yes, well, I never would have thought that getting millions of people to write differently would be viable, either, but the Palm Pilot did it. Lojban needs its "killer app," and the only way to find one is to try lots of them. > But at any rate, I don't see why you should care so much. I recognize that > you've decided that the validation for all the efforts you've invested in > Lojban is the creation of a living language rather than just a language, > but I don't understand why you should make that the validation, especially > when it's so improbable. And the original idea that a loglan-speaking > community would test sapirwhorf, I've always regarded as a bit of blarney > baloney by JC Brown who really wanted to invent a language but was trying to > (a) gain respectability for an ill-respected activity, (b) differentiate > the product from others, (c) attract adherents. flamebait, flamebait, flamebait. At least on a *lojban* list it is. > > > The only hope for > > >Lojban to succeed Top-Downly is that some organization is > > >intelligent enough to see the merits of adopting a logical > > >language, but stupid enough to choose Lojban to do the job. > > > > Gee, thanks. %^) > > What I mean is this. First, the overriding goal of the Lojban project was > always to get a minimally adequate product out into the world. The policy > was "if it's not broken, don't fix it". But if you're an organization that > is so dissatisfied with existing natural languages that you want to > adopt a logical language, you're probably an organization that wants the > language to be as good as is practicable. Um, but you may also recognize that "Better is the enemy of good enough," and see the reduced costs of using something in existence, even if it isn't quite what you were looking for. In other words, if the reduction in benefit from using lojban (as opposed to something tailor-made) leaves the cost benefit ratio positive (including switching costs), then you still make the switch, because doing otherwise takes longer and incurs other costs, making the cost benefit of a custom solution less appealing. Classic "buy vs build." la lojban needs to find the situations where it makes sense to be bought. > Secondly, and more importantly, > Lojban was designed as a compromise between many different goals. It is > probable that an organization adopting a logical language would have > different and fewer goals, and that Lojban would be a relatively poor > solution for these goals. Can you give a concrete example of the needs of an organization? > I suppose that once one organization used Lojban, that would then become > a reason in itself for other organizations to use it too. But I really > can't see it being a sensible decision for any organization to adopt it > otherwise. True, it already exists, so would save labour in concocting > an alternative language, but if you're going to invest so much in getting > your organization to use it, a redesign would probably save you cost > in the long run. Hmmm, maybe, if your organization didn't get killed by the short-run damage. Again, a classic example of network businesses - the first fax machine was useless. The *second* fax machine had some utility, but not much. The 100,000,000th fax machine has a great deal of utility. These kinds of things grow very slowly for long periods of time, then suddenly, they're huge (look at Microsoft). > I'm not hostile to Lojban. If the United Nations decided to choose a > language to be a global general purpose second language, and if I > had a vote, then if the choice had to be made from an existing language > then I would vote for Lojban. And even if there was the option of > designing a new language I would vote for Lojban to avoid the risk > of the designed language being worse than Lojban. Boy, you sure couldn't tell that from the above. > On the other hand, > of course, if the United Nations decided to entrust the task of > designing the language to me, then I would not choose Lojban...;-] Be our guest - design it, let us know about it. > > >(This isn't an attack on Lojban. Lojban is more complex > > >than it needs to be for limited, formal, written applications > > >because it needs also to be usable for the full range of > > >linguistic functions. > > > > What linguistic functions other than attitudinals are not needed > > for patent work? > > Lojban is designed to be general purpose, flexible, nonconstraining, > culturally neutral, etc. etc. The only two of its goals necessary > for patent work are logicality and nonambiguity. Flexibility isn't necessary for describing *inventions*? Non-constraining isn't necessary? Okay, maybe not strictly *necessary* in some sort of Turing-equivalence sense, but so useful as to be necessary for practical purposes. And cultural neutrality seems like a very desirable trait for patent description in the *European Union*! Even more so for global patent uses. > > More importantly, how much simpler could a language optimally > > designed for a limited purpose be than a Lojban subset that simply omits > > those features not needed. After all, a large portion of the > > Loglan/Lojban concept is optionality of features. > > If you pared Lojban down to the smallest adequate portion you'd still > be left with unnecessary stuff (e.g. zo'u, terminators) and what remained Your particular examples of zo'u and terminators seem again, perhaps not strictly *necessary* but so useful to merit inclusion. > would be Lojban only in as much as that unnecessary stuff would remain > and that the vocabulary items would be Lojban. And the vocabulary items > being Lojban would be a positively unnecessary hindrance to efficient > use of the language. It would be much easier for all concerned to use > a posteriori European vocabulary. Em, no, I'd disagree - look to the incredibly broad applications of patents in software that are being put in place, in part because of ambiguity of definitions (see "run" and "fork" again, only talk about it to a computer programmer). [...] > My own language is a general purpose one like Lojban, and has to grapplie > with a similarly disparate set of design goals, and the difficulty is > mainly in the amount of work involved. I could design the basis for > a European patent language in scarcely more than the time it would take to > decide on the phonology. Please, do so. Show it to us. [...] > First, it is not so easy to do better than Lojban if you have the same goals > as Lojban. It is easier to do better than Lojban only if you have a more > restricted set of goals. Sure - an application-specific language *might* be simpler, but I would suggest that patents, by being descriptions of new inventions, are sufficiently broad that this in fact, would not be the case. But I'd love to be proven wrong. > Second, if it is possible to do better than Lojban, > with the same set of goals, this is largely because it is possible to learn > from Lojban's 'mistakes', i.e. it is by standing on Lojban's shoulders that > Lojban can be bettered. Please, do so! :-) > Third, even if it were easy to improve upon Lojban's > design, there remains the matter of the huge amount of labour necessary to > get any language to the level of completion that Lojban has attained. Ehrm, sounds like you just said using lojban is a good thing, but before you were saying it wasn't. > Also, in a certain sense, it has been proved that it is easy to do better > than Lojban, because over the years people have often proposed valid > improvements that were not adopted (on the grounds that completion was > a more important goal than improvement). A cost/benefit tradeoff - did the change improve things enough to destabilize the design? I wasn't there, but I'll bet it didn't. Brook ---------------- Klactovedestene!