Localisation and the digital divide
Download
Report
Transcript Localisation and the digital divide
and the
Development and the Information Society
Economic divides
Language divides
Cultural divides
Sustainable development
21st September 2004
localisation and the digital divide
1
national development and WSIS
• contrast in per capita income.
– GDP 1997 - DCs $19,285; world $3,610; LDCs $245
– poverty line $1 a day - Zambia 84.6% (GDP $300)
• Countries need to develop
– To be like the West? Linear process?
– Join the information society
• World Summit on the Information Society (WSIS)
– “information and communication technologies (ICT) are
among the most important contributors to growth and
sustainable development”
– “challenge is to make ICT available and affordable”
21st September 2004
localisation and the digital divide
2
Economic divide
Zambia 0.19% internet, 0.67% PCs (US 35%, 59%)
• Communications networks limited
– good backbone between centres, missing ‘last mile’
• Cost of computers too high
– Bangalore SIMPUTER – $ 200, Berkeley PCtvt – $ 250
– SOLO – rugged but costly ($1000)
– recycled computers
• Maybe founded on wrong model of ownership
– Telecentres and kiosks, Share email addresses
• Software and Intellectual Property Rights
21st September 2004
localisation and the digital divide
3
Language divide
• The normal objective of localisation
– Well understood industrialised process
• identify text – extract – translate - replace
• BUT
– 6000 plus languages
• most unwritten
• writing systems need encoding with inevitable compromises
• adult literacy may be as low as 15% (50% globally, Nepal 35%!)
– lots of politics
• dominant languages suppress others, eg. Nepal
• sometimes perverse, eg Pakistan
– endangered languages, can we save them?
21st September 2004
localisation and the digital divide
4
writing systems
Roman system simplified
by technology
quality of fonts important
there is a fine tradition of
calligraphy in South Asia
and Islam
writing systems created for
one use
by scholars and researchers
need local standardisation
by R.K.Joshi, Mumbai
21st September 2004
localisation and the digital divide
5
Eg. the languages of South Asia
Pushtu
scholarship
Sanskrit
Tibetan
Farsi
Tibeto-Burmese
Kasmiri
Indo-European
Urdu
Punjabi
Nepali
Hindi
Indo-European
Sindhi
Dzongka
Assamese
Bangla
Gujurati
Over 500 languages
17 official in India
Oriya
Marathi
Telugu
Kannada
Dravidian
large diaspora
TibetoBurmese
Malayalam
Tamil
English pervasive
- only 5%
Sinhala
21st September 2004
localisation and the digital divide
6
Nepal language politics
• Gorkhali (later Nepali) official language since ~1750
– mother tongue of ruling elite
– 1950-1990 ‘one nation, one culture, one language’
• around 100 other languages
– Nepali now almost universally spoken
• and in neighbouring regions
– 5 other languages with mature written traditions
– 60 completely unwritten
• 1990 enabled other mother-tongue education
• 2003 mandating mother-tongue teaching
– but where is the money for this?
21st September 2004
localisation and the digital divide
7
Pakistan language politics
• Urdu official language since 1947/8
– mother tongue to 7.57%
– privileged by law and finance
• closely related to Hindi
– just different scripts
– Urdu-Hindi is second only to Chinese
• Punjabi seen as a liability
– mother tongue to 44.5%
• 58 other indigenous languages
• English language of elite
– and globalisation
21st September 2004
localisation and the digital divide
8
the content divide
• India 1.5 M websites in English
• 20,000 in Indian languages
“The new technology has also made us slaves of a
new type of ‘information colonialism’. One which
encourages us to think that the centre of the
world is somewhere in Europe or the US.”
Frederick Noronha in www.i4donline.net June 2004
21st September 2004
localisation and the digital divide
9
the literacy divide
• an illiterate person is ‘text-blind’ but can see its
layout and see the graphics.
– enable receipt of written information through Text-toSpeech generation
– enable contribution of written information through
speech recognition
• but why not base the technology on speech?
– no need for written form
– works for all languages
• plus lots of graphics
21st September 2004
localisation and the digital divide
10
Cultural and Social Divide
• Standard localisation for dates, numbers, etc
– but what about lakhs and crores?
– other date systems?
• Software embeds (best) western practice
– ERP systems taken as unproblematic
– problems even within Europe
• Cultural models
– Hofstede’s four dimensions
21st September 2004
localisation and the digital divide
11
Sustainable Development
• country must be able to do it all for themselves
– this means education in underpinning knowledge
• basic software technologies
– Unicoded writing system
– open source developments
• language resources
– language corpora
– dictionary to standardise spelling
• ethno-methods
– locally produced software
21st September 2004
localisation and the digital divide
12
NeLRaLEC – proposal to the EU
•
•
•
•
for Nepali, though not endangered
written in Devanagari (almost) – in Unicode
already couple of OT fonts, and Nepali Linux
produce
– Nepali National Corpus
– Nepali dictionary
– basic office software in Nepali
• including spell checkers
– speech generation
• trials in schools
• computational linguistics course in University
21st September 2004
localisation and the digital divide
13
Conclusions
• will lack of technology help kill languages?
• what would computers be like had they been
invented in another part of the world?
• enable people to localise imported software and
websites, or create their own
• local Open Source movements very promising
but not sufficient
21st September 2004
localisation and the digital divide
14