Message-Id: <199311191954.AA05504@eli.CS.YALE.EDU>
Date:         Fri, 19 Nov 1993 18:41:04 +0000
Reply-To: ucleaar <ucleaar@UCL.AC.UK>
Sender: Lojban list <LOJBAN%CUVMB.bitnet@YaleVM.YCC.YALE.EDU>
From: ucleaar <ucleaar@UCL.AC.UK>
Subject:      Re: proposal for new member of SE
To: Erik Rauch <rauch>
In-Reply-To:  (Your message of Thu, 18 Nov 93 11:02:50 GMT.)
Status: RO

> > Could we not instead have an additional SE (with rafsi) that means:
> >  exchange 1st & last places
> >Used recursively, this would allow a different way of getting at
> >places beyond the 5th.
> >For example, if this new member of SE were "xehe", rafsi "xez",
> >then the lujvo for "x1 is the vehicle of an event of going"
> >would be "xeznunkla".
>
> .i'enai .oi
>
> This is rampant ad-hocery. I also don't see how it could be
> used recursively.

Nor do I now. I got confused. What I was thinking of was a
SE that would, rather, make the last place the first & shunt all
the others along by 1.

If this is ad hoc, it is not adversely so. It would not, I think,
change the grammar, as "xehe" cd be included in SE.

In fact, more additions to SE could be very useful. For example,
often the place structure 41235 is more fitting to one's
purposes than 42315 (which ve gives), but getting 41235 is very
complicated. I once worked out all the SE combos necessary for
all possible permutations of 5 places, & in a few cases 5
SEs were needed. OK, if you want something really scrambled
like 53241 then FA is more helpful, but 51234 is a very natural
sequence & one wishes it were easier to achieve.

Ideally, there would be two types of SE - one that exchanges
two places & one that promotes a place to first but leaves
the sequence of the remainder unchanged. In addition, an extra
SE shd be added for accessing the last place.

> I see nothing wrong with using zei.

It's unzipfean. That's something *very* wrong with having to use zei.

----
And                      KO JBOBANPEHO