Re: [lojban] Trying to follow XXS

["And Rosta" <a.rosta@lycos.co.uk>]

Rob:
> As someone who hasn't been part of the involved jboske discussions, I
> find XXS to be a very pleasing gadri system. In general, I like the idea
> of changing the meaning of {lo PA} and letting the other gadri keep
> something like their original meanings.
>
> (See
>
http://www.lojban.org/tiki/tiki-index.php?page=XXS:%20Extended%20XS%20propos
al
> for what I'm referring to.)
>
> However, some things strike me as bafflingly odd about it. I'd mention
> this on the Wiki, but there's already such a teeming mass of commentary
> around all the proposed gadri systems (and, as far as I can tell, rather
> limited RecentChanges functionality on Tiki) that any new comments would
> be lost in the mess.
>
> * What's all this about xod-collectives? Is there some reason that lVi
> should be this construct instead of the usual fuzzy collective?

'Fuzzy collectives' are handled by lV. That leaves lVi semantically vacant.
At the same time, there is no easy way to say "The things that are each
broda are jointly but not separately brode", so one suggestion is to use
lVi for this purpose. (Personally, though, I think it would be nicer to
generalize the notion of 'set' ever so slightly, so that lV'i would cover
this.)

> * Why is alternative B there? If you want to make a distributive
> into a collective, you should just use {lu'o}, not rearrange the
> meanings of several cmavo to make way for your new word that does this.
> I'm no conservative, but that's an absurd amount of tinkering.

There is no easy way to make a distributive into a collective. We weren't
able to find a coherent interpretation for lu'o that would allow it to
do this job.

Alternative B is not that great, not least because the dist>coll problem
arises only with o-gadri, but I think it is a problem that demands some
solution. To put it in context, the only way (in XS) to turn a nonspecific
distributive into a collective is to first of all stick the collective
in a prenex and then form the distributive by PA+RI (RI an anaphor targeting
the collective). It would be nice to have a gadri that does all that
invlsibly.

> * What's with the change to MOI tacked on? It seems like a good idea if
> I understand it correctly (let {memimoi} be just {mimoi}), but what does
> it have to do with gadri? Shouldn't it be a separate issue?

IIRC, the change is from {me mi moi} to {mo'e mi moi} or {vei mo'e mi moi}
or something like that. If {mi moi} were allowed, that'd be great, though.

The tenuous logic of mentioning MOI on the page is as follows:
* PA LE is always short for PA da poi BRODA LE, with three possible values
for BRODA, the value being glorked.
* It is suggested that for the unglorkative equivalent there should be
three new ME, so that the unglorky equivalent of PA LE wd be PA ME LE
* Having new members of ME would raise the nonsensical question of what
they mean in the ME+MOI construction, but the problem lies with the
monstrous abortion that is the ME+MOI construction.

But all the stuff about getting rid of ME+MOI is just wishful thinking --
it would only become a live issue if we were trying to tidy Lojban up.

--And.