Re: [tlug] epcEditor

["Stephen J. Turnbull" <stephen@example.com>]

>>>>> "Charles" == Charles Muller <acmuller@example.com> writes:

    Charles> Stephen wrote:

    >> A program that doesn't do input is usually called a "demo",
    >> unless you call it "wallpaper" or "screensaver".

    Charles> Apparently I should have been clearer.

    Charles> Of course you can do input. It is just that it seems like
    Charles> the Japanese IME in Mandrake does not seem to be
    Charles> compatible at this time.

Ie, "of course" you _can't_ do Japanese input.  How about Chinese (any
flavor)?  Korean?  Devanagari?  Thai?  Tibetan?

    Charles> I doubt if one can truly judge the value of a program
    Charles> like this unless one is spending the main portion of his
    Charles> or her time creating and editing multilingual UTF-8 XML
    Charles> documents.

I gather that you do _not_ spend the "main portion of [your] time
creating and editing multilingual UTF-8 XML documents" _on Linux_.
Nor will you for many months, if this is your tool of choice.  (I have
no problem with that, if another platform gets the job done.)

One of the reasons I do XEmacs development is that on my platform of
choice, it's the closest thing to a truly multilingual multipurpose
editor available.  I want that as a goal in itself.  If I could only
write Devanagari and Chinese :-), I could do DTD-directed editing in
validated XML with multilingual UTF-8 content.  _Now_.  Emacs supports
all of the languages mentioned above, and has a validating DTD-
directed xml-mode (derived from code written by James Clark).

True, so far I've only done a tiny bit of X[HT]ML authoring in UTF-8,
proof of concept.  But definitely, lack of a satisfactory XML editor
that handles UTF-8 is not going to be a bottleneck.

    Charles> I, for one, am delighted to see any software company make
    Charles> the effort to provide an application for XML authors
    Charles> that, as distinguished from competing packages like XML
    Charles> Spy and XMetal, installs and works quite well on the
    Charles> Linux platform

That's certainly praiseworthy.  My point is merely that that is
something completely different from "excellent UTF-8 support".
in a new locale!

This _matters_.  My advice is that people with a strong preference for
the Linux platform who need to do multilingual (Asian language) work
not wait for epcEditor.  The technical problems of making multilingual
input work in the POSIX model are pretty tough.  I doubt that the Tcl
people will solve this; it's really not something that a scripting
language needs to do.

What Tcl should do is wait for IIIMF (currently basically vaporware on
Linux, although *surprise* there's an implementation for Emacs) which
handles this.  And the epcEditor people evidently would really like to
ignore the whole issue by relying on Tcl's facilities.  _This is the
right strategy_ for them, as both Jim and I have pointed out.

It also happens to be a strategy that does nothing to address the
needs of multilingual Asian Linux users _now_, or in the near future.
Three to four years, probably, for it to be widely deployed.

Hey, you could learn Emacs in that amount of time.  Add another year
or so, you could learn to _like_ Emacs, even.  ;-)

-- 
Institute of Policy and Planning Sciences     http://turnbull.sk.tsukuba.ac.jp
University of Tsukuba                    Tennodai 1-1-1 Tsukuba 305-8573 JAPAN
              Don't ask how you can "do" free software business;
              ask what your business can "do for" free software.