Hello Guido, et al, 

Here are a few quick comments (belatedly) to this
earlier posting in the still-running (!) $100 computer/laptop thread. I
regret that it is late and also just before GKD began the DOTCOM
discussion, so it didn't permit immediate follow up.

First, responding to Don's and Edward's mentions of MS, I agree that
although MS has made the public commitment to localize its software in
many places, it would be hard to imagine it being economical for them to
do for most languages. Also, I have heard comments from some observers
that these promised localizations have been slow. All of which of course
contributes to making OSS localization an attractive and dynamic area.

More in the text below...

On Thu, 17 Mar 2005, Guido Sohne wrote:

> I noticed that Ghana came up and I wanted to make a few comments since I
> am based there.
> On 3/14/05, Edward Cherlin wrote:
> > On Tuesday, 8 March 2005, Don Slater wrote:
> > 
> >> If Windows XP were sold at the price it usually commands in pirate
> >> markets, it would be perfectly OK.
> > 
> > Not really. There is no practical way to get Windows into local
> > languages. The only way Microsoft allows this, apart from doing the
> > development itself (Don't hold your breath) is for a government to take
> > out a license, contract out the development work, and then hand the
> > results back to Microsoft to sell. This is not realistic for more than a
> > few major languages.
> There is not a huge demand for local language applications right now. I
> am not for example, aware of a local language newspaper, though from
> time to time, one sees local languages being quoted in the press,
> however, these are expressed in an "English" encoding, since the
> characters required are absent from most fonts.

The News in Ghana (apparently connected with the Herald in Accra?) has
or at least did have until recently news articles online translated in
Twi, Ga and Ewe. See the "Ghanaian languages and ICT" board at
<http://www.quicktopic.com/16/H/9xffAXi7whnv> for links (in messages 7 &
8 - easiest to see if you click "All messages" and scroll down). These
are in PDF format, with what appear to be the appropriate orthographies.

I'm not that familiar with the typesetting software used by newspapers,
but most are not yet Unicode aware, which would facilitate multilingual
usage. BTW, it looks like the PDFs in News in Ghana were done with a
non-Unicode font.

(Parenthetically, it may be of interest that the important and
widely-used book publishing software, Quark, has finally made its
product Unicode-aware.)

> It could be that there is not a huge demand because the capability is
> not well integrated with the operating environment, but I would place
> more weight on what is seen in the press. Radio is an entirely different
> matter where the demand is very strong for local content. Local language
> usage here is usually an oral affair outside of the academic
> environment.

I'm no expert on the press but I suspect that its mass nature makes
going for the common-denominator an economic imperative. There is an
interesting example of a private Setswana language paper in Botswana -
it has made a go of it since the vast majority of the population speaks
the language, but even then it required a lot of smart work (see
<http://groups.yahoo.com/group/AfricanLanguages/message/222> for an
address by the editor).

That said, I wonder if it is not a mistake to expect or look for "huge
demand," but in any event it seems to me to be beside the point. I've
heard this turn of phrase elsewhere before and in that context (not your
use of it) it sounded like a strawman - no huge demand, ergo no need to
bother with even the basics of some free fonts and keyboard layouts.
Kenneth Keniston's observation a few years ago re localization in India
seems still apt in much of Africa and is the flip side if you will of
the "no huge demand" argument: "It can be argued that, given the fusion
of language, wealth and power in India, there is simply no market (and
perhaps no need) for software in any language other than English. Asked
about localization to Indian languages, international software firms
sometimes reply, 'But everyone speaks English in India,' by which of
course they mean that the present market consists of people who speak

IOW, on the one side there's "no huge demand" for indigenous language
software, and on the other, "everyone [in the market] speaks English [or

Demand is dynamic, and in many cases latent or potential. The trick, it
seems to me, is to anticipate meeting points down the line as things
evolve. The analogy of a pass in soccer is apt, at the risk of being
pedantic. Very often you pass to where someone will be; if you wait
until he gets there, the opportunity is lost. On the other hand if he
doesn't get there, the fact of having the ball ahead creates a new
dynamic that hopefully others can take advantage of when the ball and
they meet. Admittedly this analogy only goes so far, but in the case of
software localization - and before that even the basic (fonts, keyboard
layouts) capacities to handle African languages that new systems are
already i18n'd to facilitate - waiting for demand to act seems to be
guaranteeing missed opportunities, many of which we can't see yet.

In the case of African languages and ICT, my hunch (as an outsider I
admit) is that as the technology becomes ever more available,
localization in at least the major languages of the continent will (1)
be of interest to many who will in turn at least try them out, (2)
facilitate use of the technology by some, (3) make the technology more
approachable by others, and (4) open up possibilities for all of them
and societies at large that would not exist otherwise. These outcomes
are not unique to Africa - I think such benefits accrue from language
localization in any part of the world.

Your observations concerning radio are right on and similar to what I've
observed in Mali and Niger. But the demand for local content on radio is
not irrelevant to other ICT localization.

My impression (again, as an outsider) concerning the oral-written divide
you (and others) allude to is that it is an artificial and unsustainable
one. It is maintained by educational systems that focus only or mainly
on the former colonial languages and is reflected in anomalies such as
the traditional letter writer in much of Africa who hears in one
language, translates into English or French, and writes a letter that
may have to be back-translated on the receiving end (an example I
brought up in another context on this list a while back). And the end
results of maintaining such a distinction where people speak their first
languages with most ease but must use a second language for written
communication is in many cases imperfect transmission of knowledge and
declining use of indigenous languages (in some contexts or for some
individuals, English or French may be the most efficient languages to
use for writing, but this is not inherent to the languages nor should it
be taken as a given in all contexts). The tragedy here is that good
bilingual (and even trilingual!) models for education exist, are being
used, and have been proven to be beneficial in other world regions, but
Africa, as one of the most multilingual continents, has not benefited
from their use (this is another issue which I won't pursue here).

Yes, African cultures are often categorized as oral, which has some
important aspects, but orality is not a stagnant state, nor is it
necessarily compromised by written usage. How to maintain the strengths
of traditional oral expression while also taking advantage of the
potentials of ICT (which in some measure requires use of a written form)
is a challenge not unique to Africa - it has also been a topic in
indigenous cultures of the Americas for instance. How to move forward
with orality in the information society? This is something that could
use more South-South communication, involve innovative use of ICT, and
be an area in which Africa can contribute a lot (along with multilingual

[ . . . ]

Another aspect of what I see as a false division between oral and
written is what happens when you want to link radio and other ICTs in
various ways. Either the language that is written dominates, or there is
an ongoing need for intermediaries for translation (with all the
potential pitfalls that that has - this "adhoc translation in the field"
option [my term] is already standard practice in extension work, but
deserves more attention as a hobble to effective transfer of knowledge,
understanding and adaptation of it).

Conceding that the languages of "oral cultures" will not be used in
written form in ICT or even education (not that you make this argument,
but some do), is to put the writing on the wall for those languages
(writing which will be in English or French [with apologies to William
Ecenbarger, from whose article on Iceland I borrow the turn of phrase]).
And this arguably has longer term structural consequences.

There may be "always something new coming out of Africa" as the old
Roman saying has it, but from the point of view of knowledge for
development, if everything new coming *into* Africa is transmitted only
in the former colonial languages, its impact will continue to be
limited. Perhaps there will also be a cost to the vitality of cultures,
which are transmitted primarily in indigenous languages, by perpetuating
a linguistic divide between the traditional and the introduced. By
moving the "locus" of translation activity upstream, as it were, closer
to the production of knowledge, there is the chance for better quality
translation, making it available to more people (even when the absolute
number is not huge in the larger scheme of things), and even
facilitating knowledge generation. Localization of web content is part
of this, as is localization of software, which both facilitates use of
indigenous languages and carries the implication that use of these
languages is appropriate and relevant.

None of this is intended to displace English or French or Portuguese as
languages of wider communication (LWCs), but makes a more effective
response to the multilingual realities.

> >  
> >> I tend to get worried (particularly as an ethnographer) when I
> > 
> > So you should appreciate the value of local language support.
> See:
> "Indigenous Knowledge is a Red Herring"
> for my alternate viewpoint on the issue of local language with respect
> to the situation in Ghana.

Thanks for this link and the article, which I found quite interesting.
That said, while you highlight one aspect - the importance of exchanging
knowledge - there other aspects, those of processing, creating and
referencing knowledge that in general is more easily done in the first
language. In a bilingual or multilingual setting it may be logical to
use different languages in different ways, rather than having to use
either the lingua franca (English) for everything or limit oneself to
working only in the indigenous language. Use of the indigenous languages
then does not have to be a barrier or a divider, but a facilitator in a
different way, and the choices re language(s) can involve "both-and"

Beyond that, I think that discussions and debates on indigenous
knowledge have productively moved to where "in situ"
preservation/development is now considered important and that that in
turn necessarily implies consideration of use of indigenous languages.
In addition to your skepticism on the matter there are on another side
theoretical issues of whether recording IK in its original language
makes it a different thing (I'm not putting this well I admit).
Nevertheless, the evolution of study of IK as something done by
outsiders with an eye towards possible application in development
projects (at one time this notion was something of a revelation) to
something that is more actively acknowledged to belong to others
("intellectual property").

The main thing is that we are not talking about static knowledge, but
ways of thinking that have reservoirs of useful information - or
language and knowledge. One could try to translate everything to a LWC
and work in that, but I think you lose something in terms of content and
also process without creating more unity. The ideal would be to strive
for the best of both worlds, which means a productive role for both
indigenous languages and LWCs.

There is a lot that could be said on this topic but I'll leave it here
except to say that one does recognize both the prerogative of peoples to
decide what they want to do with their maternal languages (in ICT,
education, etc.) and the fact that within a society or community there
may be a range of preferences in this regard. Structurally the processes
of globalization as well as other factors (e.g., the notion that African
languages are inherently inferior to LWCs of European origin) have
tended to disadvantage those who would want to use their first
languages, and even to discourage the subject. I think it is important
for those of us in positions of influence in the propagation, use, and
adaptation of ICT in Africa, the rest of the global South, and in
indigenous communities elsewhere, to help keep a space open for "local
languages" and linguistic diversity as a matter of justice and also
practical benefits for each and all.

Again this is complementary and not contrary to the role of LWCs.
Translating between indigenous languages and LWCs for IK and new
knowledge transmission and creation will have costs of course, but these
are justified by the benefits (and small compared to some of the huge
amounts of money spent globally on really wasteful things), and beyond
that is something we look to improvements in machine translation to
facilitate over the coming years.

Sorry this has been longwinded and rambling (it was written at several
sittings). Thanks to all and especially Guido for prompting these
thoughts and I look forward to any feedback.


Don Osborn, Ph.D.         [EMAIL PROTECTED]
*Bisharat! A language, technology & development initiative
*Bisharat! Initiative langues - technologie - developpement

***GKD is solely supported by EDC, a Non-Profit Organization***
To post a message, send it to: <[EMAIL PROTECTED]>
To subscribe or unsubscribe, send a message to:
<[EMAIL PROTECTED]>. In the 1st line of the message type:
subscribe gkd OR type: unsubscribe gkd
Archives of previous GKD messages can be found at:

Reply via email to