The .it Registry: a short overview Delegated to CNR on December - - PowerPoint PPT Presentation

The .it Registry: a short overview

• Delegated to CNR on December 23rd, 1987
• More than 1,860,000 domain names
• New synchronous registration system from September

28th, 2009

• Coexistence of the two systems until June 2011
• About 1,900 Maintainers and 160 Registrars
• Open to EU juridical and natural persons

Operations Legal External Relations & Communication

International Relations

Director IIT/Registry Systems & Development Management Committee Rules Committee

The .it Registry: organizational structure

About 70 people including staff-operators, administrative persons and technicians

The .it Registry: the Rules Committee

• Technical advisory body constituted on April 1st, 2004
• Defines the Regulation for the assignment of the .it

domain names

• Formed by:

– 6 members proposed by the Local Internet Community (LIC)

• 4 representatives of the Internet Providers
• 1 ISOC representative
• 1 users representative

– 1 member of the GARR Consortium – 2 members of the .it Registry – 1 representative of the Ministry of Communications

• Members are renewed every two years

Synchronous vs Asynchronous: new registrations

New synchronous domains from 2009/09/28 to 2010/02/28: 141,084 New asynchronous domains from 2009/09/28 to 2010/02/28: 50,098 Total of new registrations from 2009/09/28 to 2010/02/28: 191,182

The Internet Diffusion Project

• Launched in 2001
• Aims to study the Internet diffusion in Italy

using .it domain names

– Use of domain names as endogenous metric

• The hostcount metric underestimates the Internet

diffusion

• Esogenous metrics (interviews, questionnaires,

number of ADSL contracts, PCs sold, etc.) are less reliable

Domain name metric

• Advantages

– Identification the Registrant

• Registrant characteristics
• Geographical characterization of the Internet diffusion
• Disadvantages

– underestimation of the Internet diffusion

Main Research topics

• Analysis of the Italian Internet diffusion

and Digital Divide:

– among different registrants categories – at national level, macro-area level (North, Centre and South of Italy), regional, provincial and municipality levels

Registrants classification

• One Registrant per domain has been used to avoid overestimation

cases

• For each Registrant a double characterization has been performed

– Kind of Italian Registrant

• Individuals
• Firms
• Public bodies
• Freelancers
• No-profit

– Geographical area

• National level
• Macro-Area level
• Regional, provincial and municipality level

Registrants Categories in 2009

Statistical indicators

Firms: the first 10 Regions

Regions PR SR % domains Trentino AA 32.57 1.46 2.76% Lombardia 28.53 1.28 23.55% Emilia Romagna 27.44 1.23 10.86% Lazio 26.88 1.21 10.59% Toscana 23.95 1.07 8.23% Veneto 21.99 0.99 9.09% Friuli VG 21.92 0.98 2.09% Umbria 21.27 0.95 1.50% Marche 21.12 0.95 2.87% Sardegna 20.39 0.91 2.15%

Individuals: the first 10 Regions

Regions PR SR % domains

Lazio 124.89 1.57 14.73% Trentino AA 105.27 1.33 2.18% Toscana 95.78 1.21 7.64% Lombardia 86.29 1.09 17.70% Liguria 84.86 1.07 3.01% Valle d'Aosta 82.56 1.04 0.22% Emilia Romagna 80.66 1.02 7.45% Umbria 80.61 1.02 1.54% Marche 78.69 0.99 2.62% Friuli VG 75.42 0.95 2.00%

No-profit

Percentage of Domains Registered by No-profit Corporations

• Table. Percentage of non

profit Corporations in Italy Source: ISTAT

No-profit sector: the greater diffusion of associations on the Net may be due to the fact that in Italy associations are more diffused than

• ther categories (Istat, 2001)

No-profit: Penetration Rate

No-profit: the first 10 Regions

Regions PR SR % domains Lazio

47.29 2.01 15.30%

Lombardia

27.55 1.17 16.71%

Toscana

23.94 1.02 7.95%

Campania

23.91 1.02 5.64%

Emila R.

23.23 0.99 8.27%

Veneto

22.13 0.94 8.41%

Abruzzo

21.47 0.91 2.13%

Liguria

20.87 0.89 2.77%

Piemonte

20.75 0.88 7.76%

Marche

20.55 0.88 2.93%

General: the first 10 provinces

Province PR SR % domains Bologna 602.73 2.09 3.48% Milano 535.15 1.85 12.28% Bolzano 533.67 1.85 1.47% Pistoia 465.59 1.61 0.80% Rimini 456.16 1.58 0.80% Roma 429.23 1.49 10.14% Firenze 428.49 1.48 2.49% Pisa 392.71 1.36 0.95% Trento 386.82 1.34 1.14% Siena 362.05 1.25 0.58%

Province PR SR % domains

Roma 143.40 1.80 12.33% La Spezia 140.08 1.77 0.68% Rimini 134.33 1.69 0.86% Bolzano 121.81 1.54 1.22% Milano 115.61 1.46 9.66% Firenze 113.79 1.43 2.41% Pisa 108.95 1.37 0.96% Lucca 107.78 1.36 0.91% Siena 105.22 1.33 0.61% Bologna 104.25 1.31 2.19%

Individuals: the first 10 provinces

Individuals: generational and gender digital divide

Firms: 2004-2009

2004 2009

Firms variation 2004-2009: the first 10 Regions

Regions PR 2004 PR 2009 Δ %

Sardegna 6.49 20.39 214.16% Emilia R. 10.22 27.44 168.46% Basilicata 4.92 11.81 140.07% Lazio 11.29 26.88 138.11% Abruzzo 7.42 17.61 137.31% Campania 7.67 17.71 130.90% Trentino AA 14.35 32.57 126.95% Umbria 9.41 21.27 126.08% Marche 9.39 21.12 124.96% Calabria 5.38 11.94 121.91%

Individuals: 2004-2009

2004 2009

Individuals variation 2004-2009

Regions PR 2004 PR 2009 Δ %

Puglia 18.41 62.54 239.69% Molise 16.51 55.90 238.58% Calabria 17.37 51.61 197.15% Liguria 29.19 84.86 190.72% Campania 24.16 69.41 187.29% Sicilia 20.5 58.28 184.28% Basilicata 16.19 45.80 182.87% Marche 28.89 78.69 172.36% Abruzzo 27.52 74.65 171.25% Veneto 28.43 75.32 164.92%

No-profit: 2001-2009

2001 2009

No-profit variation 2001-2009

Regions PR 2001 PR 2009 Δ %

Puglia 3.61 18.28 405.63% Abruzzo 4.26 21.47 403.61% Sicilia 4.13 18.27 342.64% Valle d'Aosta 3.73 16.34 338.44% Marche 5.05 20.55 307.09% Campania 5.89 23.91 305.81% Calabria 4.44 17.98 304.45% Veneto 5.57 22.13 297.37% Umbria 5.20 20.14 287.67% Liguria 5.68 20.87 267.75%

Gender Digital Divide: 2004-2008

2004 2008

Generational and Gender Digital Divide: 2004-2008

2004 2008

Registrar diffusion

Regions Mean % domain Toscana

3001.17
 38.39%


Sardegna

2281.92
 3.95%


Abruzzo

1245.26
 3.57%


Lombardia

573.47
 24.38%


Lazio

572.85
 8.93%


Trentino AA

416.84
 1.78%


Emilia R.

351.24
 4.98%


Piemonte

332.59
 4.48%


Puglia

269.34
 0.95%


Molise

254.68
 0.37%


• In order to identify the degree of concentration of

the number of domain names registered by Italian Registrars in the different regions we used two indicators:

• The HHI (Herfindahl-Hirschman Index)
• The Gini Index

The HHI Index

• Widely used in literature, measures the degree of competition in the

market.

• HHI is calculated by adding the square of the market shares of each

firm.

• It can be obtained through the following formula:

HHIk = S1

2 + S2 2 + S3 2 +…………+ Sk 2

• Where Sk is the market share of a firm measured in percentage

terms

• For example, in the case of a market formed by four firms with

shares respectively of 30%, 30%, 20%, 20%, HHI is equal to 2,600 (302 + 302 + 202 + 202)

• The index is structured in a way that it increases both when the

number of firms in the industry decreases and when the gap between firm size widens

• An index lower than 1,000 indicates a competitive market
• The markets in which HHI ranges from 1,000 to 1,800 are usually

considered moderately concentrated

• If the index is greater than 1,800, the degree of monopoly power

becomes more significant

• The HHI index at national level is equal to 542.75

– this shows that, at national level, Registrars are similar in size (in terms of registered domain names) – it is not possible to talk about monopoly, and moreover the number of firms at national level proves to be high (about 2,000 Registrars)

The HHI Index (cont)

The Gini concentration index

• Unlike HHI, is a standard index (in statistics a standard

index ranges from 0 to 1 and from –1 to 1)

• The Gini index varies from 0 to 1
• 1 if there is the maximum concentration
• Taking into consideration the income distribution in a country, the

Gini index is equal to 1 if there is only one individual getting the entire country income

• 0 if we have a situation of even distribution
• When all individuals have the same income

Lorenz Curve

• The Gini index at national level

is high. It is equal to 0.87

• 0.87 is justified by the fact that
• nly 10 Registrars out of

around 2,000 register the 46.30% of the .it domain names

Lorenz Curve

Determinants of Digital Divide

• To identify the key factors contributing to the

existence of the digital divide at a provincial level we have used a linear multiple regression model

• Dependent variable is represented by the log of domain names

registered in the 103 Italian provinces by individuals

• The regressors (independent variables) used in the analysis are

demographic, economic and infrastructural variables

• The regressors have been extracted from various sources (ISTAT -

Italian National Statistical Institute -, Infocamere and Tagliacarne Institute) which provide data at a provincial level

Determinants of Digital Divide (cont)

adjusted R2 = 0.647

Determinants of Digital Divide (cont)

• Significant variables are:

– Per capita GDP – The index of presence of telecommunication networks – The number of Registrars – The old age index

• in the previous table is negative and this means that provinces where the

index is lower are more inclined to register domain names

• Not significant variables are:

– The log of importation – The entrepreneurial density – Populations density – Number of foreigners per inhabitants – The jobless rate

Conclusions and Future Research

Conclusions and Future Research (cont)

Conclusions and Future Research (cont)

• Collaboration with other European

Registries interested in the same field of research, to analyse and compare the Internet diffusion in different countries and the eventual key factors leading to such results

• Is CENTR interested in promoting this

project in the CENTR community?

Thanks for your attention!