fu'ivla

[Nick NICHOLAS <nicholas@uci.edu>]

cu'i la'o ny. Arnt ny.

> This whole business with "wholesale borrowing" worries me.  It may be OK
in
> cases such as Linnean names, element names, country names and language
> names, which are "namelike", and can be based on the place structures of
> danlu, ratni, gugde and bangu.  But how about the other "common
> international science words"?  Words such as "hypothesis",
"visualization",
> "correlation coefficient" and "synchrotron" may be similar across
> languages, but that doesn't necessarily mean that borrowing them into
> Lojban would be a Good Thing.  How should the place structure of such
words
> be derived, if not individually?

Arnt is quite correct, and I think that has been reflected in practice;
unless I've missed something between '94 and '01 (which is not
impossible), there hasn't been any movement to introduce loanwords for
those *precisely* because their predicate nature is much more important.

Where this leaves us with computer terminology, I'm not sure (though this
is an issue that explicitly came up as we were planning the dictionary at
Lojban Central.) But
Lojbanists are disinclined to use fu'ivla for anything other than concrete
realia, for good reason.

-- 
Momenton senpretende paseman mi retenis kaj  #  NICK NICHOLAS.
   kultis kvazaux                           #   TLG, UCI, USA.
      senhorlogxan elizeon                 #   www.opoudjis.net
         (Dume:                           #    nicholas@uci.edu
[Victor Sadler, _Memkritiko_ 90]         #