On Fri, Jun 15, 2012 at 11:49 AM, Bob LeChevalie= r, President and Founder - LLG <lojbab@lojban.org> wrote:

Jacob Errington wrote:

Well, even that wording isn't really so bad at all. I mean -- what is a= pipe? It's essentially a "transient container" open on both = ends, isn't it?

A container is implicitly defined by its function (to contain its contents)= . =A0tubnu and certain other words are shapes or forms - the function may f= ollow from the form, but the essential concept is the form =A0and not the f= unction.

You'll find that the wording of a lot of gismu make reference to the fo= rm/function dichotomy - something we thought was of fundamental semantic im= portance back when we defined the place structures.

I noticed that in some of the definitions, comparing {kabri}, {t= ansi}, {tanxe} to {kurfa}, {cukla}, {slanu}.

I remember rather li= king how versatile they were because of the distinction, like using {slanu = kabri} for most cups or mugs, while using the same root to build lujvo like= {ctikabri}. It really creates a different class of words that are surprisi= ngly specific in definition compared to cup/bowl/etc. (as in, do you call a= bowl a bowl because it is used for soup or because it is rounded?)

So, defining {tubnu} becomes a matter of whether we wan= t it to be a shape/form definition or functional. If it's a shape/form = definition, I see the functional version being something like {benji tubnu}= for pipe. If it's a functional definition, I see {kevna slanu} being t= he shape/form equivalent. Just looking at those options, I'm kind of in= favor of using tubnu functionally, as a container, while using {kevna} in = general to express shape/forms that have interiors. (Are there any other sh= ape/form words that have two materials? I don't recall having seen one.= )

Additionally, I feel that the current choice of psuedo-= glosses is a functional set, not a shape/form set. It's left ambiguous = as to whether it has to be a circular cylinder or if it could have any shap= e of cross-section, so long as it "carries" or "contains&quo= t; things like a tube/pipe would. For that matter, it's ambiguous wheth= er {slanu} has the math or intuitive definition of cylinder. (The intuitive= definition says it should have a circular cross-section, some math definit= ions merely require the cross-section to be uniform.) This is relevant for = discussing things like {slanu kabri} (above) and {damtu'u} (below). In = the former, I'm trying to use {slanu} to emphasize the circular cross-s= ection shape, but in {damtu'u}, the shape merely has to carry fluid, le= nding to a functional definition.

Also note that we have the lujvo {jabytu'u}, {damtu= 'u}, {moltu'u}, {tu'urbirka} as dependencies. (Anyone see any o= thers?) They all seem to lend themselves more to the functional definition,= though I'm not sure about {jabytu'u}.

All in all, at the start of this email I was pretty ind= ifferent, but looking around the definitions, I really want {tubnu} to have= a more well-defined functional definition, taking dependencies into accoun= t, and I want {slanu} to be clear as to it's cross-section. I almost th= ink it's enough to warrant another sumti place for the shape of cross-s= ection, defaulting to circular, similar in nature to the x3 spot of {kurfa}= . What do others think?