Peter T. Daniels wrote:

>
> >
> > What hardware keyboards do you expect them to have access to?
> >
> > What computer software do you expect them to have access to?
>
> Is there are reason that these more limited syllabaries can't use the
> standard 47-key keyboard with NO ROMAN LETTERS ON IT WHATSOEVER, with
> keyboard drivers that produce the right characters on screen, WITHOUT
> GOING THROUGH A TRANSLITERATION-TO-ROMAN STEP?
>
> Vai, obviously, can't use a 47-key keyboard.

Seems like any language can be typed using any physical keyboard, given
a scheme for interpreting the keystrokes which is likely to be
implemented in software. After all, you could type in Morse code with
just one key!

So, I think Michael's original plan seems like a good one - build
something that he thinks is likely to work, put it out there in the
wild, and see what happens. If the people can't use it, or won't use it,
or feel like their use of it is not as effective as it might be, then
someone ill notice that and use the information that emerges from the
experience to build an improvement. I hardly think that it needs to be
perfect on the first try. After all, I don't see anyone saying the plan
Michael proposes won't work in generating the characters, only that it
might be difficult for Vai people to use and, indirectly I am hearing
that the market acceptance will be low.

I did see Michael say several times that he welcome alternatives -
whether it is at the physical kb layer, the software layer, the teaching
layer, whatever anyone can bring to the table. For now, his choices are
made, and while others are arguing, he is building. He seems willing to
do the practical thing and "see what happens" and then revise from there.

Personally, I think in general that is a good approach for new markets.
while of course the Vai are pretty new to computers now, that is not
always going to be the case.

Someone mentioned Michale as being a "cultural imperialist" for
proposing and acting on his plan and not giving up in the face of the
opposition here. But what I wonder about is that the implications of the
arguments of the others seems to be that the Vai are not capable of
learning how to use new tools, or that if they are, they will only be
able to do it once and so the first time better be perfect and optimal
somehow.

I find that subtext troubling. My own background is on the engineering
and product development side, like Michael. It is not my experience that
any product, computer or not, is ever perfected in the first go. If it
is not useful, then it drops from the market. If it is useful, then user
feedback is used to generate new and improved versions. Ongoing product
iteration That is life in the product development world.

I mentioned earlier about how Vai might take this under their wings, let
me elaborate a bit under this context. There is free and open software
available that runs on standard computers and can be modified at will
for additional language related features (and anything else one can
think of). We know that as Linux, but there are also others, such as the
BSD variants of Unix.

While it is hardly likely that the first group of Vai users are going to
suddenly become expert enough computer programmers to take their destiny
entirely into their hands, I do think it is fair that among the first
group of users there will be at least one and maybe a group who will
recognize any shortcomings in the first system, based on practical
experience, and align themselves with people who are interested and able
to build their suggestions. Right now, that is looking like Michael and
his colleagues. But others could take up the mantle too, and each could
work independently of other groups if they wanted to. Ultimately the Vai
will decide what is best based on market considerations and other
usability factors that seem hard to describe right now without actual
experience.

As for market factors, one thing that has been touched on in this
discussion is the practicality of using existing Latin Keyboards, but
hasn't been really delved into. That is, there is a real manufacturing
cost, that is not trivial, to build specialized keyboards, especially in
limited numbers.

On the face of it, given a non-standard keyboard, a limited market, and
an extremely poor market, it seems unlikely any commercial manufacturer
is going to create a non-standard keyboard such as some here might
desire (although they haven't specified the specifics of it yet other
then it is non-Latin based). I can see that a NGO *might* be persuaded
to put up some money to design and even produce a limited amount of such
keyboards. Surely they would physically work, given the accompanying
software layer.

And they might even turn out to be more usable then Michael's plan. Who
knows? But are they optimal in the bigger picture, or would the be just
a local optimum? IOW, who would produce more when they are needed -
keyboards break down, new ones are needed for new users, etc. At some
point, the NGO route is not going to keep funding the growth and market
forces will take over. Maybe it is reasonable that the Vai would be able
to afford the extra cost at that point, I don't know, seems unlikely,
but conceivable to me.

So, back to the "cultural imperialist" remark: One approach seems to
lead towards a way to allow the Vai to get started and then participate
in the wider world while having a collaborative opportunity at self
determination of what kind of keyboards they will like. The other seems
to offer a paternalistic approach of "we know best" that eventually
leads to a dead end.

Now, there are probably some in-between options too, so don't nobody get
your britches all bunched up :) But my point is that calling names is
not practical, but recognizing the underlying reason for choosing a
rhetorical approach - Michael's engineering, "put it out there and see"
approach , and other's academic "let's design the perfect optimum
independent of actual market forces" are not mutually incompatible. The
first creates a channel for getting product which can be evaluated into
the market, and the second is part of the process of analyzing the usage
and market feedback for possible iterations.

Pure Academic discussions won;t help the Vai speaker "on the street".
But a method to put options in front of him/her, including funding, in
repeated iterations for improvement are likely best for the end user in
the long run. Both sides have a role to play in that wider process, and
I think instead of both sides saying "I know what's best" and sniping
about that, it would be a better use of limited resources to understand
the bigger product design and integration process and focus where one
can best add value.

Best,

Barry