unesco-itu global symposium on promoting the multilingual internet geneva 9 -11 may 2006

UNESCO-ITUUNESCO-ITU

Global Symposium Global Symposium on Promoting on Promoting

the Multilingual Internetthe Multilingual Internet

GenevaGeneva99-11-11 may 2006 may 2006

Daniel PimientaDaniel Pimientapimienta@funredes.orgpimienta@funredes.org

NETWORKS & DEVELOPMENT NETWORKS & DEVELOPMENT FOUNDATIONFOUNDATION

http://funredes.orghttp://funredes.org

http://funredes.org/LChttp://funredes.org/LC

Linguistic & Cultural Diversity in Linguistic & Cultural Diversity in CyberspaceCyberspace

1.1. A framework to understand the A framework to understand the complexity of the digital divide.complexity of the digital divide.

2.2. Its Its application to the understanding to the understanding of the diversity issue.of the diversity issue.

3.3. Analysis of different approaches for Analysis of different approaches for measuring linguistic diversity on the measuring linguistic diversity on the InternetInternet

APPLICATION APPLICATION OF THE PREVIOUS OF THE PREVIOUS

FRAMEWORK ICT4HD FRAMEWORK ICT4HD TO CULTURAL & LINGUISTIC TO CULTURAL & LINGUISTIC

DIVERSITYDIVERSITY

HOW DOES THE HURDLES HOW DOES THE HURDLES TRANSLATE IN TERMS OF LANGUAGE TRANSLATE IN TERMS OF LANGUAGE

AND CULTURE?AND CULTURE?

ACCESS/INFRASTRUCTUREACCESS/INFRASTRUCTURE

Technological neutrality is a myth (ref.) however the interfaces however the interfaces concentrated most of the probmes concentrated most of the probmes realted with cultural and linguistic realted with cultural and linguistic diversity.diversity.

Ref: Charles Ess, (2005) “Can the Local Reshape the Ref: Charles Ess, (2005) “Can the Local Reshape the Global? Ethical Imperatives for Human Global? Ethical Imperatives for Human Intercultural Communication Online, in Capurro, R Intercultural Communication Online, in Capurro, R & al (Eds), Localizing the Internet. Ethical Issues & al (Eds), Localizing the Internet. Ethical Issues in Intercultural Perspective, in Intercultural Perspective,

Ver Ver http://icie.zkm.de/congress2004http://icie.zkm.de/congress2004

ACCESS/AFFORDABILITYACCESS/AFFORDABILITY

Universal access must include the Universal access must include the consideration of consideration of a price coherent with the level of revenues of the with the level of revenues of the

target populationtarget population

Lenguages & cultures of poor Lenguages & cultures of poor segments of population (basically in segments of population (basically in the South but also immigrants in the the South but also immigrants in the

North)North)

ACCESO/SUSTAINIBILITYACCESO/SUSTAINIBILITY

Telecenters’ organization scheme should be coherent with local cultural should be coherent with local cultural

practices.practices.

Maintenance documentation in local Maintenance documentation in local language.language.

ACCESS/LOCALIZATIONACCESS/LOCALIZATIONIInterfaces shall allow access in mother tongue and be shall allow access in mother tongue and be

natural to user’s culture.natural to user’s culture.

IDNIDN Keyboard layoutKeyboard layout

Alphabet codificationAlphabet codification Online dictionnaries and other support software (Ref)Online dictionnaries and other support software (Ref)

Automatic Translation (in the future)Automatic Translation (in the future)

Ref: José Antonio Millán, “How much is a language Ref: José Antonio Millán, “How much is a language worth…” - Ver http://jamillan.com/worth.htmworth…” - Ver http://jamillan.com/worth.htm

ACCESS/LITERACYACCESS/LITERACY

The aim of universal acces is to reach The aim of universal acces is to reach knowledge and then start by a and then start by a a

functional literacy at 100%at 100%

USEUSE

Here starts the requirement for Here starts the requirement for digital literacy

which shall be conceived as a process which shall be conceived as a process which pays tribute to local languages which pays tribute to local languages

and cultures.and cultures.

TECHNOLOGICAL OWNERSHIPTECHNOLOGICAL OWNERSHIP

How to really own a tool which does How to really own a tool which does not understand your mother tongue?not understand your mother tongue?

MEANINGFUL USAGEMEANINGFUL USAGE

Local content production and virtual Local content production and virtual communities in local languages.communities in local languages.

SOCIAL OWNERSHIPSOCIAL OWNERSHIP

Network & Information’s Culture & Network & Information’s Culture & Ethic no are neutral and shall go thru Ethic no are neutral and shall go thru

a process of a process of sincretism with local with local cultures.cultures.

EMPOWERMENTEMPOWERMENTSOCIAL INNOVATIONSOCIAL INNOVATION

The closer we get to Human The closer we get to Human Development more the importance Development more the importance

switch into the cutural aspect.switch into the cutural aspect.

What is the meaning of those What is the meaning of those concepts in a given cultural context?concepts in a given cultural context?

HUMAN DEVELOPMENTHUMAN DEVELOPMENT

We will get close to this goal only if people’s We will get close to this goal only if people’s participation has been a reality along the has been a reality along the

described process, if persons are subject and no described process, if persons are subject and no object of the process.object of the process.

¿What meaning has “participation” and how does it ¿What meaning has “participation” and how does it get real in a given cultural context?get real in a given cultural context?

¿Does real participation exists if a language ¿Does real participation exists if a language different form mother tongue is imposed?different form mother tongue is imposed?

THE NEED FOR RELIABLE THE NEED FOR RELIABLE MEASUREMENTMEASUREMENT

HOW TO ESTABLISH POLICIES AND HOW TO ESTABLISH POLICIES AND FOLLOW-UP THEIR IMPACT WITHOUT FOLLOW-UP THEIR IMPACT WITHOUT RELIABLE INSTRUMENTS FOR RELIABLE INSTRUMENTS FOR MEASUREMENT?!!!MEASUREMENT?!!!

SITUATION OF LINGUISTIC SITUATION OF LINGUISTIC

DIVERSITY IN THE INTERNETDIVERSITY IN THE INTERNET

FROM RESULTS OF FUNREDES DIVERSITY OBSERVATORYHTTP://FUNREDES.ORG/LC

METHODOLOGY CREATED WITH CONTRIBUTION/SUPPORT OF UNIÓN LATINA AND SUPPORT OF FRANCOPHONIE

1998-2005

INDICATORS 2005INDICATORS 2005

INGLINGL SP.SP. FR.FR. IT.IT. POR.POR. RUM..RUM.. DeutchDeutch RES.RES. TOT.TOT.

SPEAKERS (millions)SPEAKERS (millions) 630630 375375 130130 6060 190190 3030 120120 53705370 60006000

SPEAKERS %SPEAKERS % 10,5%10,5% 6,3%6,3% 2,2%2,2% 1%1% 3,2%3,2% 0,5%0,5% 2%2% 89,5%89,5% 100%100%

INTERNAUTES (MILLIONS)INTERNAUTES (MILLIONS) 300300 8080 4949 4242 3838 44 7171 516516 11001100

INTERNAUTES % SPEAKERSINTERNAUTES % SPEAKERS 47,647,6 21,3%21,3% 37,7%37,7% 70,0%70,0% 20,0%20,0% 14,7%14,7% 59,2%59,2% 9,6%9,6% 18.3%18.3%

INTERNAUTES % POPULATIONINTERNAUTES % POPULATION 5.0%5.0% 1,3%1,3% 0,8%0,8% 0.7%0.7% 0,6%0,6% 0,1%0,1% 1,2%1,2% 8,6%8,6% 18.3%18.3%

% INTERNAUTES PER LANG.% INTERNAUTES PER LANG. 27%27% 7%7% 4%4% 4%4% 3%3% 0,4%0,4% 7%7% 25%25% 100%100%

WEB PAGES%WEB PAGES% 45%45% 4,6%4,6% 5%5% 3,0%3,0% 1,9%1,9% 0,2%0,2% 6,9%6,9% 33,4%33,4% 100%100%

PRODUCTIVITY PAGE PRODUCTIONPRODUCTIVITY PAGE PRODUCTION 1.571.57 0,660,66 1,141,14 0,810,81 0,550,55 0,700,70 1,061,06 1,321,32 11

WEB PAGE PER SPEAKERSWEB PAGE PER SPEAKERS 4.294.29 0,740,74 2,282,28 3,053,05 0,590,59 0,330,33 3,473,47 0.370.37   

(Sources: Global Reach & Funredes)

EVOLUTION % WEB PAGES EVOLUTION % WEB PAGES COMPARED TO ENGLISHCOMPARED TO ENGLISH

FUNREDESFUNREDES 1998-2005 1998-2005

Evolución de las lenguas latinas con respecto al Inglés

0.00%

2.00%

4.00%

6.00%

8.00%

10.00%

12.00%

14.00%

16.00%

18.00%

Español

Francés

Italiano

Portugues

Rumano

Aleman

Español 3.37% 8.41% 9.46% 10.95% 11.24% 11.36% 11.60% 10.83% 10.30% 10.19% 10.23%

Francés 3.75% 7.33% 7.89% 8.86% 9.13% 9.14% 9.60% 8.82% 10.18% 10.64% 11.00%

Italiano 2.00% 4.60% 4.93% 5.88% 6.15% 6.15% 6.51% 5.28% 6.09% 6.15% 6.77%

Portugues 1.09% 3.95% 4.44% 5.40% 5.57% 5.61% 5.62% 4.55% 4.36% 4.02% 4.15%

Rumano 0.20% 0.37% 0.33% 0.32% 0.35% 0.36% 0.33% 0.23% 0.41% 0.31% 0.37%

Aleman 5% 11.00% 11.43% 13.42% 13.74% 14.08% 14.41% 13.87% 15.37% 15.42%

Sep-98 Ago-00 Ene-01 Jun-01 Ago-01 Oct-01 Feb-02 Feb-03 Feb-04 May-04 Mar-05

EVOLUTION ENGLISHEVOLUTION ENGLISH% WEB PAGES & INTERNAUTES % WEB PAGES & INTERNAUTES

1988-20051988-2005Porcentaje pagina web en inglés (Fuente: Funredes)

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

Porcentaje ingles

Porcentaje ingles 75.0% 60.0% 55.0% 52.0% 51.0% 50.7% 50.0% 49.0% 47.0% 46.3% 45.0%

Sep-98 Ago-00 Ene-01 Jun-01 Ago-01 Oct-01 Feb-02 Feb-03 Feb-04 May-04 Mar-05

porcentaje de internautas en inglés (Fuente: GlobalReach)

0.0%

20.0%

40.0%

60.0%

80.0%

100.0%

Sep-98

Ene-99

May

-99

Sep-99

Ene-00

May

-00

Sep-00

Ene-01

May

-01

Sep-01

Ene-02

May

-02

Sep-02

Ene-03

May

-03

Sep-03

Ene-04

May

-04

Sep-04

Ene-05

% Internautas en inglés

% Internautas en inglés 60.5% 58.0% 48.5% 46.0% 45.1% 44.4% 42.5% 36.9% 38.6% 34.8% 28.6%

Sep-98

Ago-00

Ene-01

Jun-01

Ago-01

Oct-01Feb-02

Feb-03

Feb-04

May-04

Mar-05

OBSERVED TRENDS SINCE 1998OBSERVED TRENDS SINCE 1998

CONSTANT REDUCTION CONSTANT REDUCTION OF ENGLISH OF ENGLISH DOMINANCEDOMINANCE

SLOW INITIAL GROWTH OF SLOW INITIAL GROWTH OF FRENCHFRENCH COMPARED COMPARED TO TO SPANISH & PORTUGUESESPANISH & PORTUGUESE THEN SWITCH THEN SWITCH

OBSERVATION LEFT TOOBSERVATION LEFT TO MARKETMARKET … … WITH LACK OF SERIOUSNESSWITH LACK OF SERIOUSNESS

MARGINAL PRODUCTION OFMARGINAL PRODUCTION OF DEVELOPING DEVELOPING COUNTRIES COUNTRIES & NO PROGRESS& NO PROGRESS

SURVEY OF METHODOLOGIESSURVEY OF METHODOLOGIESFOR MEASUREMENT:FOR MEASUREMENT:

INTERNAUTES PER LANGUAGEINTERNAUTES PER LANGUAGE

Global Reach http://global-reach.biz/globalstats/index.php3

Unique Transparent Compilation of various not necessarily

coherent sources therefore approximative However stable and frequently updated

SURVEY OF METHODOLOGIESSURVEY OF METHODOLOGIESFOR MEASUREMENTFOR MEASUREMENT

WEB PAGES PER LANGUAGEWEB PAGES PER LANGUAGE

SIMPLE USE OF SEARCH ENGINESSIMPLE USE OF SEARCH ENGINES

VARIOUS MARKETERS STUDIESVARIOUS MARKETERS STUDIES

ALIS / OCLC STUDIES (1998 – 2002)ALIS / OCLC STUDIES (1998 – 2002)

FUNREDES (1997 – 2005)FUNREDES (1997 – 2005)

LANGUAGE OBSERVATORY (STARTING)LANGUAGE OBSERVATORY (STARTING)

SIMPLE USE OF SEARCH SIMPLE USE OF SEARCH ENGINESENGINES

Complement of empty space methodComplement of empty space method Simple but not stableSimple but not stable Used by many after Funredes (1988)Used by many after Funredes (1988) Offer an approximation of the search Offer an approximation of the search

engine own estimate based on engine own estimate based on language recognition algorithmlanguage recognition algorithm

Gross estimation but unreliable and Gross estimation but unreliable and seems to favour Englishseems to favour English

MARKETERS STUDIESMARKETERS STUDIES

DOES NOT REVEAL METHODOLOGY DOES NOT REVEAL METHODOLOGY THEN UNTRUSTABLETHEN UNTRUSTABLE

INKTOMI (2001) : EXCELLENT INKTOMI (2001) : EXCELLENT MARKETING BUT GROSS ERROR!!!MARKETING BUT GROSS ERROR!!!

DESINFORMATION CHANNELS WITH DESINFORMATION CHANNELS WITH POWERFUL VOICE POWERFUL VOICE

ALIS/OCLCALIS/OCLChttp://www.dlib.org/dlib/april03/lavoie/04lavoie.html

TRANSPARENT METHODOLOGY (although not the TRANSPARENT METHODOLOGY (although not the result process)result process)

BASED ON SAMPLING 8000 RANDOM IP NUMBER BASED ON SAMPLING 8000 RANDOM IP NUMBER AND APPLYING LANGUAGE RECOGNITION AND APPLYING LANGUAGE RECOGNITION ALGORITHMALGORITHM

STATISTICAL FLAWSTATISTICAL FLAW

NONE FREQUENT FOLLOW-UPNONE FREQUENT FOLLOW-UP

IMHO FAVOUR ENGLISH FIGURESIMHO FAVOUR ENGLISH FIGURES

FUNREDES/UNION LATINAFUNREDES/UNION LATINAHTTP://FUNREDES.ORG/LCHTTP://FUNREDES.ORG/LC

TOTALLY TRANSPARENT METHODOLOGYTOTALLY TRANSPARENT METHODOLOGY

BASED ON SEARCHING A SAMPLE OF 57 BASED ON SEARCHING A SAMPLE OF 57 CONCEPTS CAREFULLY SELECTED FOR CONCEPTS CAREFULLY SELECTED FOR EQUIVALENT MEANING AND REACH IN THE 8 EQUIVALENT MEANING AND REACH IN THE 8 STUDIED LANGUAGES (3000 WORDS) STUDIED LANGUAGES (3000 WORDS)

SOUND STATISTICAL PROCESSSOUND STATISTICAL PROCESS

METHOD ALLOW OTHER SPACES AND CREATION METHOD ALLOW OTHER SPACES AND CREATION OF FURTHER RESULTS BY LANGUAGE OR OF FURTHER RESULTS BY LANGUAGE OR COUNTRIESCOUNTRIES

FUNREDES/UNION LATINAFUNREDES/UNION LATINAHTTP://FUNREDES.ORG/LCHTTP://FUNREDES.ORG/LC

MANY LIMITATIONS THOUGH:MANY LIMITATIONS THOUGH:

LIMITED NUMBER OF LANGUAGESLIMITED NUMBER OF LANGUAGES MARGINAL COST PER ADDED MARGINAL COST PER ADDED

LANGUAGE LANGUAGE MEASURE THE INDEXED INTERNETMEASURE THE INDEXED INTERNET MEASURE % COMPARE TO ENGLISHMEASURE % COMPARE TO ENGLISH DEPENDANT ON SEARCH ENGINESDEPENDANT ON SEARCH ENGINES

LANGUAGE OBSERVATORYLANGUAGE OBSERVATORY

UNIVERSITY NETWORK WITH STRONG UNIVERSITY NETWORK WITH STRONG TEAMTEAM

IN THE STARTING PHASEIN THE STARTING PHASE WILL BE BASED ON LANGUAGE WILL BE BASED ON LANGUAGE

RECOGNITION AND CRAWLINGRECOGNITION AND CRAWLING NO LIMITATION ON LANGUAGESNO LIMITATION ON LANGUAGES TRANSPARENTTRANSPARENT

A PROMISING AVENUE FUNREDES HAS A PROMISING AVENUE FUNREDES HAS CALLED FOR YEARSCALLED FOR YEARS

SURVEY OF METHODOLOGIESSURVEY OF METHODOLOGIESFOR MEASUREMENT:FOR MEASUREMENT:

OTHER SPACESOTHER SPACES

FUNREDES/UL METHOD ALLOW TO MEASURE ANYTHING WHICH IS INDEXED AND SEARCHABLE (E.G. NEWSGROUP)

FUNREDES/UL HAS DONE A FIRST APPROXIMATION OF CULTURAL MEASUREMENT

FUNREDES/UL METHOD ALLOW CONSTRUCTION OF SOME INTERESTING INDICATORS

WHAT’S NEXT?WHAT’S NEXT?

THE NEEDS GO MUCH BEYOND WEB THE NEEDS GO MUCH BEYOND WEB MEASUREMENT.MEASUREMENT.

HOW TO MEASURE LANGUAGES USE HOW TO MEASURE LANGUAGES USE IN EMAIL AND MORE DYNAMIC IN EMAIL AND MORE DYNAMIC LANGUAGE BEHAVIOUR?LANGUAGE BEHAVIOUR?

ALEXA INSPIRED METHODS COULD BE ALEXA INSPIRED METHODS COULD BE THE SOLUTION…THE SOLUTION…