WikiDiscuss

pc:
> Well, skip concept and go directly to translating English (where concept
> talk usually ends up anyhow). I think then that there are many common
> English words we get at only through oblique places.

Hard to tell. The x3 of {tirxu} for example won't do for the common
word "stripe". I think the intent was that those oblique places
would cover many English words, but the effect was the opposite,
a reduction in the coverage of semantic space. Also, places of the
"under conditions x" and "by standard x" sort are, at least for me,
just a hindrance in the learning of place structures.

> Come to that, how do we
> add on the places that are there now if we lop them off. Is there, for
> example, a word for "out of material x" that is independent of {zbasu}
> ({marji} is no help here)?

Why is {marji} of no help? In fact, if {zbasu} didn't exist, I would
use {majgau} for "x1 makes x2 out of material x3". Of course, {marji}
is bloated too with its x3, so that really comes out as
"x1 makes x2 out of material x3 in shape/form x4". Good thing
nobody thought of adding that place to {zbasu}.

> And, of course, the fact that {klama} is one
> concept does not mean that it is not composed of several — even five --
> concepts, into which it may be broken down.

That's the problem: it can't really be broken down into simpler concepts,
unless through clumsy means like {zi'o}. With {klama} you get the
full package or nothing.

> Harrdly more restricted from a practical point of view; there is only one
> case that ever comes up when the materials place is inappropriate.

There was a similar example with {finti} the other day.

> And
> practicality, rather than raw semantics, was a guiding principle in gismu
> creation.

I'm not arguing for raw semantics though, and I don't even think
{zbasu} is one of the worst offenders, three places is still
manageable. But I think that bloatedness in general goes against
practicality.

> That's not to say that the constructors didn't err sometimes — in
> both directions (though, admittedlly, mainly in havimg too many places).

Right, nobody expects them not to have erred sometimes. It would be nice
if there was something we could do about it though.

> Probably not, since there is neither origin nor destination nor vehicle,
> only route.

Right, and there is nothing for going without origin, destination
and vehicle. {litru} comes close, but it has a vehicle. So what
does the Earth do? {muvdu} is also too heavy.

> So {klama} is not very good for "go" in a lot of cases (not
> news, since it already is said to cover "come" as well).

{klama} covers come and go, but only a fairly restricted sense of
come and go. A one place predicate "x1 comes/goes" coupled with
the wide variety of spatial tenses Lojban offers would cover much
more ground. As it is, the spatial tenses are wasted because there
are almost no predicates that will support them. If you try to use
spatial tenses with {klama} you mostly get nonsense.

mu'o mi'e xorxes

__
Do you Yahoo!?
Yahoo! Mail - 50x more storage than other providers!
http://promotions.yahoo.com/new_mail