From: Don Osborn (dzo@bisharat.net)
Date: Mon Apr 30 2007 - 19:18:30 CST
Hi Bill, all, Some interesting exchanges and info in this thread, thanks.
WRT the issues Bill brought up, I think another significant dimension to
remember is that many systems (hence keyboards) where MINELs/1 are spoken
will be used by more than one user. In a multilingual society (many
languages, and most individuals with skills in more than one) this makes it
hard to predict what needs a keyboard will be asked to fill in the course of
its lifetime, except that it will likely involve characters beyond what you
can print on or affix to the keys.
The typical pattern in many countries is indeed for both multiple users and
multiple languages, and in many cases the latter require either more Latin
characters than will fit on one keyboard (one per key), or more than one
entirely different script. It is possible to print more than one symbol on
each key (as is the case with Arabic keyboards or in Nigeria with the
Konyin/2 keyboard) or even to paste on stickers, but as has been pointed
out, that can become problematic in terms of visibility.
Part of my thinking is that the visual finality of the standard keyboard -
that is, what you see on the keys of say a QWERTY layout - disfavors use of
anything that is not seen on the keys. Not so long ago I saw an email from a
French speaker in French apologizing for no accents because he was using a
QWERTY. In a recent exchange on H-Hausa/3 (also cc'd to
A12n-collaboration/4), the issue of what one sees on the "surface" of the
keyboard was specifically brought up in relation to non-use of the full
Hausa orthography in Hausa-language blogs.
On a collective level, such as in public points of access, the dominance of
these standard keyboards arguably leads to an "LCD" (lowest common
denominator) dynamic - it's the standard, what everyone uses, and you make
do with it. Only a few people overcome this.
Another issue is that of user profiles (a bit broader concept than user
skills). A lot of users are relatively new to computers with all that that
might mean in terms of concepts about how the keyboard works, non
touch-typing, inserting characters, switching layouts, etc. In the typical
pay per time scenario of a cybercafé, there is also the disincentive to
figure out how to insert characters even if the concept is clear. The LCD
dynamic rules.
The ability to create keyboard layouts for one or multiple languages
certainly has a lot of potential benefits, but I think these run into
problems, beginning with the physical keyboard and user profile issues
mentioned above.
The possibility of visual onscreen keyboard views in some ways seems to have
the potential to address a number of these issues: it's visible and it can
reflect any number of layouts. A significant usability issue has been raised
concerning the trade-off between ease of visibility and economy of space
used on the screen. Nevertheless, it would seem a good short-term way (at
least) to facilitate use of diverse orthographies with alternate layouts.
Part of the reason I look forward to a cheaper dynamic/interactive/LED
keyboard product (such as Optimus/5) is that it is in theory the only way to
assure ease of input in any script or orthography. In effect it levels the
playing field like nothing else can - this is psychological as well as
practical. But I would hasten to reiterate that I see the main benefit of it
as practical. If the cost were much less, it would be by far the most ideal
technology for non-dominant languages with extended or non-Latin scripts,
and the multilingual contexts in which these tend to be used. So yes, I do
think it is important and has tremendous implications for use of diverse
languages (only the LED keyboard really trumps the LCD one?). That said,
however, I would certainly not suggest waiting for that brighter tomorrow -
other interim solutions as you suggest should be pursued.
(Another, separate, issue with keyboard layout creation is their
promulgation. IOW, public points of access may not have such installed. In
fact, in Africa at least, my guess is that telecenters and cybercafés rarely
do, despite the oft-cited proliferation of layouts, though I know of no
studies. This is an awareness and distribution issue and gets into issues of
localization policy on the country level, decisions [not] made in ICT4D&E
projects, and perhaps even commercial marketing strategies. If it is still a
problem at such time as dynamic LED keyboards are more accessible, then
people like me will not have been doing our work.)
There are other issues (in what I call the "localization ecology") which
bring me to your question of Hausaphone users in Niger. In most of Africa
the educational systems have focused in many cases exclusively on
instruction in the inherited European languages (there is a body of
literature on the costs of that decision, which oddly enough emerges only in
discussions of education and not broader development challenges - that's
beyond the scope of this discussion except to say that it is leading to some
changes). So in many African contexts there is the situation that would be
an anomaly in most other regions of people more familiar with reading and
writing in a second/additional language (in Niger, French) than in their
maternal language or main national lingua franca. So my impression and
limited observation is that users in Niger whatever their linguistic
background (but on the average among the more privileged) use French on
computers. That's a current state of affairs and will certainly change, but
in an ongoing multilingual use framework.
Nevertheless, when someone decides they want to compose in say Hausa (and
even if they are not schooled in it, have some familiarity with the
orthography), they are confronted with the keyboard issue - and behind that
I might add, the persistent notion that some locally developed "special
font" is necessary (and that that is not installed). That's probably as true
in northern Nigeria as it is in Niger, and that's concerning one of Africa's
most spoken languages with a pretty dynamic local publishing tradition (the
latter mostly from Nigeria).
Part of what has me thinking a lot about this is that if we see this kind of
problem with Hausa - many speakers, written tradition, relatively simple set
of additional characters (3 in Nigeria, 4 in Niger) and no necessity for
diacritics, but still issues re input - then it doesn't bode well for other
languages in the region.
Hopefully this makes some sense - it's a bit longer and more rambling than
what I originally intended...
Don
1. MINEL is my coinage, FWIW: Minority (or maternal - some people object to
"minority" even when it is intended merely to describe a demographic
relationship), Indigenous, National (i.e., not official - the term and
distinction are significant in many African countries), Ethnic (or perhaps,
endangered), Local languages. The idea is to cover a range of more or less
overlapping categories that are not dominant or international languages. One
has no pretense that this could be a well defined category, but may be of
use in some discussions.
2. http://www.konyin.com/
3. http://www.h-net.org/~hausa/
4. http://lists.kabissa.org/mailman/listinfo/a12n-collaboration
5. I'm wondering if Art Lebedev Studio might be looking to sell the concept,
once they get it going.
> -----Original Message-----
> From: unicode-bounce@unicode.org [mailto:unicode-bounce@unicode.org] On
> Behalf Of William J Poser
> Sent: Monday, April 30, 2007 2:58 PM
> To: unicode@unicode.org
> Subject: Re: Optimus keyboard
>
> Marnen Laibow-Koser wrote:
> >I believe you're missing the point here. I do a lot of multilingual
> >typing on Mac OS X. It's easy enough to switch keyboard mappings,
> >but that's really not the problem. The problem is that once
> >switched, it's annoying to squint at Keyboard Viewer to see what
> >character is on which key. And Mac OS users are fortunate in that
> >Keyboard Viewer is easily accessible: I *still* haven't found a
> >reliable equivalent in Kubuntu or Windows XP.
>
> I agree that a keyboard like this would be very handy for those
> who switch writing systems frequently. However, it is not clear to
> me that this is relevant to the original question, which was
> whether the lack of such a keyboard is the sticking point for
> widespread use of minority languages. A great many minority language
> users do not need to switch back and forth constantly - if they are
> going to use their minority language, they just need a keyboard
> that lets them enter it efficiently. For such users the switching
> issue is not important - what is important, beyond being able to
> obtain the keyboard mapping they need, is LEARNING the minority
> language
> mapping, for which keycap covers and on-screen keymaps may be quite
> adequate. Once they learn it, they will touch type, so the issues
> raised about keycap covers falling off and becoming dirty and so forth
> are not really relevant.
>
> It is true that SOME minority language users will also have to write
> frequently in a larger language. If they need to switch back and
> forth frequently between languages with incompatible keyboard
> mappings, something like the Optimus would be valuable. The question
> is,
> how large a proportion fall into this category, and how large a
> proportion fall into the category of those who would largely or
> entirely
> write in their minority language if they could.
>
> As I said before, I'm not unappreciative of this keyboard. I'd like
> to have one. What I question is whether it is the key to enabling
> significantly greater use of minority languages. The fact that for
> certain classes of users who write in several languages or who
> frequently mix languages within documents this kind of keyboard wold
> be very useful does not necessarily make it a key to writing in
> minority languages.
>
> I suspect that, unfortunately, the relevant data does not exist.
> What I would like to see is evidence that, e.g., speakers of Hausa
> in Niger would, if they could, write (almost) entirely in Hausa,
> or would, nonetheless, write mostly in French, or would write
> 50% of the time in French and 50% in Hausa. Does anyone know of such
> data?
>
> Bill
>
This archive was generated by hypermail 2.1.5 : Mon Apr 30 2007 - 19:20:41 CST