Strengthening ccTLDs in Africa

IGF 2010  
Vilnius, Lithuania
13 September 2010
Session 63
0900
Strengthening ccTLDs in Africa


Note: The following is the output of the real-time captioning taken during Fifth Meeting of the IGF, in Vilnius. Although it is largely accurate, in some cases it may be incomplete or inaccurate due to inaudible passages or transcription errors. It is posted as an aid to understanding the proceedings at the session, but should not be treated as an authoritative record.   


********

>> ALICE MUNYUA: very important issue to begin to think about seriously, and that's why we ask the question, what do we really mean when we talk about strengthening ccTLDs in Africa, we're asking all the various stakeholders, what does it mean from a regulative and consumer perspective?  For example, what does it mean from the ccTLD managers themselves?  
So I think that's the question I would like everybody to have and to begin to explore right now, also from a policymaker perspective, what does it really mean to have a strengthened ccTLD?  Does it mean creating new policy like we did in Kenya that begins to acknowledge ccTLD in law?  Or perhaps it's not enough?
So that's    it's a question we would like to, you know, take up as part of the discussion when we open up the floor.  We have, you know, in the light of IGF concerns, and for each of the East African countries, we have, you know, administered questionnaire to ISPs to regulate authority to policymakers and consumer organisations.
The questions were tailored broadly addressing concerns to the African IGF and to IFC.
And a little outline on the questions, we have looked at the structure and management of ccTLDs.  Operational issues which we going to hear from KENIC, how are the operation, and is there competition and at what level?  
We also looked at the issue of Cybercrime that affects the ccTLDs, what are the technical measures in place, to ccTLDs participate in other regional security ccTLDs?  Are there other policies that ccTLDs have put in place to reduce abusive registration and spam and others?
We were looking at the role of various stakeholders, so what is the role of regulative authority and parliamentary and other ccTLDs and the organisational structures, including the role of any organisation that has been created, does it    KENIC and organisations that have been set out to manage the ccTLD?  
We are looking at new area of ccTLDs consumer awareness that hadn't been done in any other ccTLDs in terms of the issues of price, service, consumer satisfaction, even though in some of our countries that register and sell dot ke and understand the    from a consumer perspective, how do we see ccTLDs and, perhaps, is that one of the problems in terms of the law of registration of ccTLDs of dot ke and the generic ones?  And do ccTLDs need to put more emphasis on marketing?  If they do, at what level?
Then the role of registrants and in terms of promoting a relationship between local and eCommerce, the relationship between local content and speak to that in terms of local languages and ccTLDs, you know, and protection of privacy.
Then for us, our region of issues of access, it's very, very important to issues of roadmap and Internet affect our ccTLDs, Kenya, for example, we felt that, you know, the ccTLD would only    KENIC going ahead and doing a marketing campaign without looking at background and other efforts made towards increasing the usage of Internet users is not really going to work if we don't have the level of Internet user that we are looking for.  So that's why it's important for an organisation of KENIC to work with other government institution working so hard to increase access to Internet.
So the specific objectives of this study were to identify transient management of ccTLDs, what are the good practice and bad practice, identify nonoperational issues, highlighting some of the, you know, causes.  Identifying good governance model, is it nonprofit better?  Is private public partnership or is it better to have a holder model?  
We also want to identify the current and ongoing policy and regulatory issues that affect not just the operation of the ccTLDs, but how it    I mean, all the other issues that affect it from the policy perspective.
Then exploring types of Cybercrime committed in the region, what are the strategies that we need to put in place to ensure that, you know, African ccTLDs are not being used as conduits for Cybercrime, identifying good practice management structures for efficient operation of registry and competition as well.  Highlighting consumer issues, consumer issues here ranging all the way from service all the way to relevant content, highlighting issues of access to broadband and the Internet.
Now, the initial responses we found have been very interesting from misunderstandings about technologies, especially ICANN technologies, not really understanding what they mean.  So even misunderstanding of some ICANN processes, like delegation processes and pleased we're doing this with ICANN and this is something to take up in terms of capacity building and ICANN delegation processes.  
Also, examples of some well established ccTLDs like KENIC and ad hoc policy also presence.  One of them was dispute resolution where we are finally thinking about formalizing it and adopting it into a national, you know, context.
And I think KENIC will give an example of how important it is with dot force, dot ke, and how we end up dealing with it.  
Also, very few responses    we got very few response from ccTLD managers, how they understand delegation, what    any of the problems they may have encountered in processes, yet in some countries, like Rwanda, processes ongoing and very few responses on that.
We also found we have very sophisticated policy regimes, and manual processing, some of them very interesting like KENIC, people are able to buy domain names, quite a number of domain names through more money.  There's a discussion about buying domain names against the backdrop of new generic level domain names.  The East African community, for example, is discussing the name of getting a dot ee domain name discussed in 2004 and is discussed more today because of the East African common market.  
However, we found out dot ea, East Africa, has already been resolved by IS and is belonging to some Spanish territories and we're not able to get to that in a moment, so we had to fall back to EAC or dot East African Federation or community that can be created.  There was a discussion in    it's like a chicken and egg.  We had to think about a regional TLD, so this is also against the background of the controversy against Africa.  
Should we really focus on strengthening our ccTLD or think of a dot eac and that may help strengthen our ccTLDs?  I think a question I throw back to the audience we may want to discuss.
So you see that map asking that question here dot eac and strengthening, you know, dot Kenya, Rwanda, dot ug from technical excellence, governance, policy structures, price and marketing, branding as well, which is important, and quality of service.
Thank you very much and look forward to further discussions on some of these issues.  Thanks.
(Applause)

>> Thank you very much, Alice, for that insight.  I think we discussing strengthening these, we need to think about what we mean by a strong the policy is it the infrastructure, the model, the prices as we discuss the challenges, we need to think about this as we go to open session after our presentation.  So I call about GLOCOM to do their presentation.

>> KEISUKE KAMIMURA:  Good morning, I'm Keisuke Kamimura, from GLOCOM.  I'm a colleague of Adam Peake, so many of you, I assume, know Adam Peake quite well.  
Today, this morning, I would like to talk about a little bit after our finding of our research and its policy implications.  I really appreciate this opportunity to be here.
Our study is not necessarily focused on Africa, but on ccTLDs globally, but we    I would like to give you some implications that    which may be applicable to the issue of strengthening Africa ccTLDs.
Let me give you some background of our study.  We started with a very simple question, how can we    can we understand and measure the governance of ccTLDs from an integrated perspective?  By the way, we have some handouts in the back of the room, so you have all    more or less the same version of the screen one, just    feel free to take one copy.
So we started with very simple question, we know ccTLDs started with local authority and is in many diverse ways, but despite the diversity, ccTLDs form a specific part of wider and local Internet, so it's vital to safe, secure and trustworthy Internet as a whole, not in part.
And earlier studies like McCarthy in time of the administration of the ccTLD and the security and trust of the Internet and showed necessities to ensure good governance in ccTLDs to show security and trust in the Internet like wider benefits of the ccTLDs like reasonable pricing of registration and so on.
So, we tried to find ways to crib to improving the security and trust of Internet and ccTLDs area in particular.
Could you go to next slide?  Our study is called country code governance or ccTLD in short.  We did not look at TLD we just focused on ccTLDs.  We tried to develop a ccTLD index for measuring the exclusive development and modeling it for better governance of ccTLD operation and administration.  And we also wished to implement, to some degree, our development to the best effect possible.
And our team is comprised of a couple of universities, including the international university of technology, international university of Japan and GLOCOM, Concordia university of Canada and other expert and advisors and our project lead from the university of technologies right there, so we may call upon him from further comments after my presentation or in the question and answers time.
And by the way, our funding is    comes from Japan, science and technology agency research institute of science and technology for society, it's a noncommercial academic grant, so we do not represent any commercial interest, so please understand our research findings are academic or independent.  Okay.  I might have spoken too much about housekeeping kind of issues.
Framework about our study, we employ the inductive approach based on the current practice of domain in administrations, so we looked at the actual state of play of the domain name use, first, and then we tried to tried to draw some implications from it.  And here's our framework we have a matrix of three    fixed sales.  It goes access and affordability, better service to local community, and secure and trust.
And horizontally    vertically it goes system level and content level, and we have    we have each    each one area to follow, but let me give you further details later.  Could you go to the next slide?  Before going to our policy implications, we would like to    I would like to give you some of our finding.  One is domain name pricing.
Is domain name affordable?  This is the first question we tried to answer, and the graph shows    the X axis shows the price range.  By the way, the price is floated in local of ten, so the number doesn't go linear, and Y axis shows the volume of the price, so the blue line shows that many ccTLD name spaces priced more or less around $50 a year, so it has a steep peak around that price range, but that's    that's a nominal figure.
So we tried to understand that number from a real    from real terms and we divided the domain name price by the monthly income of the ccTLD jurisdiction and the peak goes much lower and the prices range scattered around the X axis which means from a nominal perspective, the minimum registrations are priced, more or less, uniform price, but in reality from real terms, real    in real terms, domain name prices mean a quite differently to their income level.
Could you go to the next one?  We also provided analysis of DDP of the ccTLD jurisdiction and the number of registrations.  This chart looks a bit complicated, but the point is that domain name registrations grow as the GDP or local income level grows.  And the next one?
This slide is a bit technical.  Actually, when we calculated the number of registrations of the domain name price, we ISC, that stands for Internet system solution based in California, they provide a quarterly survey results of the domain name numbers.  Last year in Sharm El Sheikh, our project made a brief presentation on our survey results, then we had a very critical strong criticism because our findings was not based on actual data.  We relied on our finding on    we based our study on IC data, which is 100 picture of all ccTLDs.
So we compare IC based findings and findings based on actual observation.  So we found that 50 some ccTLDs published the number of registrations on the Internet publicly, so we took the correlation between the two sources and then we got the chart.  It shows the two sources show a relatively high correlation, which means from a nominal perspective it would be reasonable for us to draw some implications based on this IC data.
We do understand IC domain and survey does not provide an accurate profile of ccTLD, but we could use it for more generalized information, so we decided to continue for the further study.  Could you    yes.
This slide shows per capita analysis on GDP and the number of registrations, so in general domain name registration grows in line of economic development just as I mentioned, but deviation of pricing gets smaller as market develops.  In the chart LL stands for lower    lowest end GDP class, L stands for lower end GDP class root, and H stands for high GDP and double H stands for highest GDP group.
Oh, could you go back to the previous one?  Yes.  At the right bottom, you have standard deviation figures and LL has a standard deviation of .86.  L class has 1.2 and H has standard deviation of .7.  Highest group has .66, deviation of .66.  So you can see the price deviation gets smaller as the local market develops.  So another implication is in the lower end ccTLDs, registry tend to serve or sell domain names to foreign registerants who may or may not come from foreign countries and that's why it's larger in the lower end countries, so    lower ends countries.  That was our interpretation.
And this is another analysis of GDP based classification of ccTLDs.  So we    here in this chart, we compared the number of registrations and the registration fees.  HH again stands for the high end GDP class of countries.  CcTLDs.  So in the HH, the correlation between the number of registrations and registration fees is relatively high, whereas lower end ccTLDs have weaker correlation.  So the price is not    price does not reflect the actual market demand, so this is what we can draw from this analysis.
So we can go to    first of all is implication that is    so for the better user benefit, we should    in that long term strategies for future local market growth when designing and marketing ccTLDs locally.
Okay.  So we go to next finding, which is how ccTLDs can respond to legal and harmful content.  We do understand ccTLDs or domain name are criminal to content, so many registries choose not to interfere with the content of the domain name or the IP addresses, but as spam or phishing or other illegal content are prevailing on the Internet, we should do ever way possible to discourage or block these content.
And, actually, some CCs have policy to interfere with content and table shows how many CCs have explicit clause for canceling domain names in order to block or discourage content level or TLDs.  We looked at CCs from ATDs or middle countries, and we have 49 CCs in our statistics and we found that not a few registries have a clause that domain name may be canceled for other than procedural things like delayed payment or inaccurate things like details and so on.  Some CCs have clauses and some have them for content and so on.  We understand this is a somewhat tricky area because CCs should not interfere too much with content layer, but as    as the content is changing, we may need to have somewhat updated approach to ccTLD policy.
So we have the second policy implication which is this, act preemptively against illegal and harmful content.  We also found out that some CCs like dot HH and Hong Kong and Japan and ZX like Christmas Island, they have preemptive policy to discourage illegal content, so and some worked quite effectively to such behavior, so this is another suggestion or recommendation from our study.
And this is not for regulation.  This is for encouraging user    more user confident or user trust.
Okay.  Third implication, we had another implication from our study, which is to enjoy multistakeholder based governance.  I didn't give you precise funding here, but we found that open participatory governance and user benefits like demanding pricing are correlated, so an open participatory government stands for participation in ICANN processes or employing resolution or processes based on that kind of governance and consumer benefits like domain name price are correlated, so these two aspects should be in line with each other when discussing what kind of ccTLDs policy should be employed.
So    yes, so ccTLDs are a public use and really serious trust on behalf of the Internet community as a whole, and that's on one hand.  But on the other, ccTLDs has authority based on something like national sovereignty or self determination, but that should not exploit the rest of the community because ccTLDs is delegated to local entity and part of the wider and global Internet, so if domain names are to serve local users, registration fees or other, you know, user benefits need to reflect the income level of local users.
But if domain names are to serve nonlocal users like foreign corporations, the legitimacy of ccTLDs, we need a redefinition.  In extreme cases, ccTLDs need to be administered more efficiently than a government registrar than a local player, so we have to consider this seriously from an end user perspective.
So let me conclude, we have, based on our research finding, we have three implications.  One is to end long term strategies for future local domain name market, growth.  Number two, to act preemptively against illegal and harmful content.  Number three, ensure and make sure ccTLDs it reflects the local users' needs and serve local benefits.
In conclusion, we propose in order to ensure better governance in ccTLDs and better benefit, ccTLDs be organized and user principles and the charter    the next one, which is this, so this charter is drafted according to the matrix of six cells I chart earlier.  This is our recommendation from our point of view.
Thank you very much.
(Applause).

>> Joseph Kiragu: Thank you, very much, Keisuke.  Thank you for that insight from your research.  Very interesting to note from your findings that the price of ccTLDs is more domain name growth is the development.  The growth    also domain name growth and price is not market share and demand.  Then you pointed out growth and ccTLDs and security is very helpful insight.  Most ccTLDs in Africa would like to carry on and find out how to strengthen the ccTLDs.
Now, we have a presentation from    ccTLDs is on the National Information Structure.  They identify their country on the Internet.  They are actually countries and location and brand.  So it's the work of the presentation.

>>  ABIBU RASHID NTAHIGIYE:  Thank you, Chair, for this opportunity.  First of all, I would like to thank GLOCOM for sponsoring my participation to this meeting.  Basically, I will share with you the experiences on managing the registry, the history behind, and basically, I would like to talk on the Internet and DNS, the historical background, the operational issues, current status, the challenges and opportunities, and then I will conclude.
So the Internet, we all know it is for Internet and networking and global networking for    it is key networking for everyday activities.  Now, DNS is engine part of Internet, without DNS the Internet can't operate.  DNS is a critical resource.  Other critical Internet resources include IP addresses, telecommunication infrastructure, ISP with question mark because some people say ISP are not critical, so case of localisation content is a critical infrastructure.  Therefore, the registry, ccTLDs, they are part of DNS and are critical Internet resources.  
Now, on the registry management, the management of dot is done by Registry Information Centre, so they manage this for the tzNIC information name.  They are Greater Internet Service Providers, that is Public Private Partnership.  It was established in 2006, and through an exclusive process.  
It was an issue which is start on how it should be managed, so they came up with a report and recommendation for establishing an entity to do    tzNIC was launched in 2008 in May, and currently it is staff for forming the Secretariat, and this Secretariat to the board, which is called Policy Advisory Committee, and it has eight people with equal representation from the tzNIC members.
Before tzNIC Corporation, the .tzNIC Services were provided by individuals and representatives of the university.  And the registry management was Randy Bush, a technical contact, and the professor was the original contact.
Now, after operational in 2009, a delegation in April of 2010, the tzNIC and credited.  They're a ccTLD.  It is among five ccTLD KE for Kenya, for Rwanda and as I say, tzNIC and it was a fully dedicated on April of this year.  It is the East Africa that is so far the delegation.  It is about 600 in the domain but I give some explanations.  The number has gone down for some reason.
The model behind establishing tzNIC were three and these are local management of wire resources, that is infrastructure, development and management.  And also opportunity for local capacity, human capital development.  With the registry and countries, we have opportunities for capacity building like practical training opportunities for students at the universities, internship problems, et cetera, et cetera.  Contrary to if it would be managed outside of the country.
And also, opportunity to enjoy the benefits associated with .tzNIC operation based on local environments, financial issues, the money remains in the country.  Legal issues, business traditional and language as well.  So, the first reasons are part of the ICT policy objectives.  So we are    we are implementing the objective of the ICT policy.
Now, tzNIC policy are governed by the IIS netted members.  It is the electronic communication and it is called the TCRA and it is the action of 2003, the manager of the communication resources by consultation and it is facilitators because granting the institution until now.  Membership to tzNIC is open.  We have two founding members, but I'm facilitating this, so that in future it can contribute to local Internet community and policy Internet development issues.  Currently, the two global ISOC members and currently has 25 members.
On the operation, we rely on grants and currently we're granting the registry, and they grant this on temporary basis.  I was told initially on three years, but if you take into consideration the .tzNIC, it was 2009, but since we started operation in August of 2009, this has been extended and we believe that since the regulator, we invented the process and been reappointed for another five years time, they support us when we are sustainable.  And our sustainability is basically expected from the registry services that is registration and fees.
On the policy and rules, registration is adopted ICANN and customers see the local environment and currently we are in the process of appointing the distribution registration and registrars, we documented three other    registration of the complement of the registry, but the registrar does the registry.  We register the domain names, own do the registry,    80% of the domain names are done by the registry trust.  We currently have 12 register trust and that's why tzNIC is    we have cross registration, and that's why the registration of domain is complete for registration trust.  So maybe in future we going to open.  I'm not very sure.
We have a fixed registration fee and it's inclusive of Commission register Trust.  So the Registry Trust Charge an amount depending on domain, hosting and website hosting, but for the domain registration maintenance fee is fixed rate.  The fee is about 3.5 U.S. dollars for registration, and 3.5 to $13 renewal.  And so far we have done about two promotions whereby we waived the registration fee and we manage to register about 120 domains per month.
So as I say, the register is complete and other services, hosting and web designing and hosting, and at the moment, the domain can be registered within an hour, if it has been hosted.  But the maximum number is three days, taking into consideration weekends and public holidays.
Registry payments is automatic or manual.  It can be done through bank deposits internet banking can be done by check.  We allow also people paying by cash, bank transfer, and we started payment through and we are finalizing payment through another operator.  
On registry development, it's ready and one of our staff has attended a training for the deployment of DENIC and maybe before    on the current status, when we started operation, we had six level domains and now we have one for primary school, secondary schools and it is .sec, and the fee is $3.05.  This is done of 5,000.  That is USD 3.5 registration and 3.5 for maintenance for year.  It is    as I say, we are PBS ready and for deployment.  
The number of domains, I say I'm going    domains by 2010, we have 600 domains, but they have gone down to 3500.  The reason is in the past, domains were for free.  We gave a grace period for one year, which expired in July of this year, and most of the domains which are not used at the moment have been suspended or outsold, so the number has gone down.  But for now the number will be going high.
The upgrade per month, initially the registration went down by the university and individual, so there are not enough publicity and marketing campaign, so it was about 45 per month.  So with minimal publicity in market, we are reaching about 110 domains per month, registrations per month.
The tzNIC per month, generic domain name, some people generic domain names are superior than the dot tz local, that some people who don't really understand, they say dot tz are slow and dot coms are fast, so also the price competition, you'll find generic domain names are cheaper than dot tz.  
So if you take in consideration the price, then you going to straight go to dot coms, but if you take into consideration the benefits for sure, the price issue will not be a    an issue and you going to be registering at .tzNIC.
Most of the people do understand about the identity using the ccTLD domain names, but it has been the habit that people have been studying domain like abc.tzNIC.dot.com.  On publicity and marketing, this is challenging because we are in front of ccTLD marketing, so it    is a challenge to us and also the service is a bit technical to common man.  If someone comes to register at .tzNIC.main, he can change his mind.
Also the publicity, there is a lack of specialization on people from the area reporting the ccTLD issues.  I went to one of the meetings when we became operational and I gave the materials to many people, but what came out in the newspapers was very funny, so I don't have specialization reporting ccTLD issues.
Sustainability, this is a time consuming issue.  It going to take some years because the sustainability depends on the volume and prices as well.  So which one should I start?  This is about the chicken and egg issue.  If you need big volume, you need lower fees.  If you want quick sustainability, if you put higher fees, you going to end up with very few domains.
Another issue is service availability.  Dot com is available to common people, they go to hotmail without any fee.  And the challenge here is that the online business opportunities has not been realised.  You find some people distributing magazines for free, but they end up making money from the ad.  If they change and do online business, while at the same time providing free e mail account, this going to change the situation.
On the register service centralization, we find that for the case of tzNIC, we have    we have out of the    so it's a challenge.  We need to centralize the services around all the countries.  And also there is a challenge.  Even though we have fixed the prices for registration and renewal, these registrars mark up prices and they do it in a package.  They don't specifically tell what is the registration fee and what is the renewal fee, so at the end of the day you find end users paying a lot of money contrary to what we suggested.
So to conclude, the local management of the ccTLD should be emphasized and the publicity and awareness are the key taker for sustainability.  TLD    domain names can be developed through free e mail accounts they have through land advertisement and free e mail accounts.
While fees are crucial for content localisation, hosting companies should be encouraged so local traffic can remain local and this will contribute to diminishing prices.  Starting with the government, local content within the government, okay?
Coalition among the countries so ccTLD can succeed, so we need to allow like benchmarking exercises for other countries which have been rededicated.  About a month ago we, had Botswana so we can continued    should continue assisting our colleagues.  By doing this, we will be developing the future together.
That's my last slide.  Thank you.
(Applause).

>>  JOSEPH KIRAGU: Thank you, very much, for you.  It's interesting to know that tzNIC re delegation, and it was one of the few early re delegation that was not contested.  This can be a case study for the ccTLDs in Africa that are struggling.  They can learn this from tzNIC.  
It shows how ccTLDs are struggling.  The issue of price, do we lower the price down?  Do they pay?  I wish they would get the revenues coming.  This is an issue we need to address.  The perception issue, people think that tzNIC is less superior?  The same thing for dot ke, they think it is local?  This is something we need to address so much.  
Also on the challenges, you touched on some challenges that we are facing.  Interesting also to know about the role of ISPs and ccTLDs, some people have been trying to argue that ISPs are not critical Internet resources, ISPs are critical Internet.  We need them for ccTLDs to grow and domain space to grow up and local content issue.
I'm now going to welcome Anne Rachel for few remarks for ICANN, most ccTLDs in Africa are struggling and the issue of free delegation is key.  We would like to hear about the re delegation process from Anne Rachel.

>> ANNE-RACHEL INNE:  Thank you very much.  Good morning, everybody.  Good to be here.  Thanks for inviting me to be here, ICANN to be here.  So I don't have a presentation.  I'm going to talk to the subject because, you know, I think sometimes presentations actually distract from what we have to say.
So what is    what is a relationship between ICANN and ccTLDs, first of all?  ICANN, basically, is the repository of information from ccTLDs via what we call the function on ICANN.  If you go to www.ictanet.org, you get where the domain names are put in the system and when the Internet community created ICANN in 1998, some processes and procedures were put in place, one of them being re delegation or, for example, asking an operator of a ccTLDs can ask for day top ten use for what we call the sponsoring organisation, so the sponsoring organisation in our case is    the organisation that oversees the person, the moral authority under the domain that kind of gets    gives the technical contact, technical contact being the person on a daily basis who inputs the information on the servers of the domain, so this is what we have.  Those are the information that we have, the    for the people managing the domain.
Then underneath we have the list of main servers that serve that domain, wherever they are.  Okay.  So this you can actually find for every ccTLDs at the IANA pages, okay?  So in case    in the case we're talking about here, re delegation is only when you want to change the contact or reorganization of a ccTLD, okay?
So that's what a re delegation is.  I'm going to backtrack just a little bit.  ICANN doesn't coin the codes, the two letter codes.  We get them from a list called 3166 1.  From ISO.  And if you just type, you know, in any search engine ISO 3166 1, you will find that list.  
So this organisation is in charge of doing that, the two letter codes are not the only ones the organisation does, you know, there are three letter codes that are for other users like, you know, passport or whatever.  So from the beginning basically John Postel who used to be the one man care taker took the 3166 list and put it in the root for the DNS.
So in case you want to change information about the people who are managing those codes, you come to ICANN and there is a process.  ICANN doesn't choose the people who manage the domains.  This is a sovereign matter.  All we do is put the information that the countries, the communities give to us of the people who are managing those domains.  One of the thing we do, rest of the community, ISOC is here, the rest of the community and a few others is give trainings, you know, to ccTLD, country code operators on what we call the registry operation courses.  They have different levels.
They are    when they come to explain what a re delegation entails or give information, how did other people do it?  Why are they successful?  Being one of those, just like KENIC being one of these to use at the workshops to come tell people about their experience, some of the good parts over the delegation and others that are, you know, not that good, basically you heard Joe and Vika talk about contested re delegations, they're re delegations whereby the community doesn't agree on who should operate the domain.
It can be a question of the old operator not wanting to give the management to the new entity that the community has chosen, it could be now that, you know, for example governments are more and more interested in the subject, two departments absolutely want to manage a TLD.  ICANN cannot sit down and say, oh, so and so ministry is more legitimate to manage a TLD than so and so ministry.  So unless it's resolved at home, there's not much we can do.  
We can come and explain this is how other people rezoned it, this is how other people shared legitimacy and controlled operations of the TLD, but the community chooses what they wish to do, not ICANN.
And we don't stand on disputes referees.  We don't stand as referees.  It's not for us to decide.  It's a sovereign issue and it has to be decided by the community.  Okay?
So one of the things we do in terms of training is to    as I told you, so we do what we call the registry operation courses with people like NSRC, ISOC and the regional TLD associations like, for example, at the end of this month we going to have what is called the advanced registry operations course with the FTLD with 20 TLDs from the African region.  So basically this is divided into two parts, technical operations, sound technical operations.
What does it entail?  Security issues.  A little bit about, for example, things like DNSSEC?  How do you implement that?  What do you need to know?  Something in administrative in terms of what do you explain.  You have to understand that management of the ccTLDs changes.  There's about 248 of them around the world.
On a regular basis, we have a list of 10 that come every month to ask for either a change of service or change of name of a person because in the organisation that manage it is TLD, you have one person who has left for one reason or another, you know, so we have all these changes to do all the time.  And we always have to remind people on how those changes actually take place.
Some of the requirements in terms of having, you know, name service changes dawn, some of the issues we face on a daily basis, for example.  Somebody sends the name of a technical contact or just name server IP addresses to change, and they forget the technical contact and the administrative contact have to actually verify the information.  And this is for your own security.  This is for the community security just as well as ICANN security.
The administrative contact and the technical contact both have the verify that the changes that are asked for are legitimate.  So sometimes we go by for weeks and a lot of us have    are called up by INR to remind, you know, ccTLD operators that, please, you need to send that e mail or, you know, did you really ask for those changes then?  Because things like that happen.  It's the same process with re delegations.  Once there's several steps.  Once you have produced some documents, INR verify them.
They ask you if they need more information and paperwork, they will ask you to produce that paperwork.  If you don't do that in a certain time frame, we do what we call administrative close of the request, which means that administratively, the request is closed for the moment until you provide the information that is needed before the case can go forward.  
So re delegations have taken anywhere between a month to six years.  Six years why?  Because as I said, the community doesn't agree at home, and unless the community agrees on who's going to be the operator, what is the    show on the root zone, there can't be any changes.  There can only be one root operator.  
Some of the issues that we're facing in the regions    and especially in the African region, is really the fact that, you know, community at home does not come together to make sure that what is needed is done in terms of paperwork, in terms of people talking to each other, to make sure this is what I agree on.
We see disputes on just basic charges that have to be put in place, disputes about people, as I said, the government entities that should supervise, name it.  There's, really, you know, quite a few things that we've seen over the years.
So our    everybody that    the first step is, really, this is one of the things we do, is to say we have a paper called re delegation step by step at IANA.  This is one of the things we push to people who are willing to go for re delegation.  Read it.  Let's go through it together, let's see who is it that you're going to need and let's try and just do it together and make sure that you get it.
Or else it's going to take a lot of time, okay?  We're here for, you know, the hand holding.  We can tell you the way it's been done in different countries, the way other people resolved issues.  
Don't think that you're the only one having those issues.  We actually have plenty around, okay?  Everybody has, actually, a set of, you know, difficulties that are absolutely different because the context is different because the laws are different.  Because the simple regulations to put in place are different.  Because maybe your vision of what the ccTLD should come upon is different.  So all of that has to be taken into account.  
I'll stop here.  If you have any questions, that's how we've been working.  Happy to accompany ccTLDs being strength end in the region.  Thank you.
(Applause).

>> Joseph KIRAGU::  Thank you, Anne Rachel for that insight, and let us know how ICANN is supporting cities in Africa as you talk about growth.  It is interesting to know about the training that ICANN is conducting with ccTLDs in Africa.  AFTLD together with ICANN is organizing a workshop for advanced registry of operation, of course.  Most ccTLDs in Africa will be participating.
Also interesting to know about the administrative training aspect, although this has not been coming so much, I think most ccTLDs will expect this to become much as a technical training is coming.
Also, interesting to know that ICANN does not act as a dispute on the re delegation process.  Okay.
I'm now going to invite, we still have a few minutes, so we can have one short presentation by the professor and how to    the research of the language of university, very interesting topic.  Welcome.

>> Just give me a few minutes to make to our project.  I'm working together with Mr. Ntahigiye on the development of country governance project.  And the project is composed of two elements.  One is modern chart metrics he presented in his concluding remarks    it is six cell metrics, which I expected to be used for the valuation of the domain management.
And the other components is the CDG indicators, a set of indicators composed of maybe 22 study indicators.  And these two are expected to be used in combination to    by stakeholders to evaluate how the domain manage so well.
Look at the left bottom slide in front.  So this is the structure of the a set of indicators.  So it is composed of six cells.  The one is accessibility and the second one is openness and maybe centre one column is diversity, diverse issues.  And right on side is security and trustworthiness.
And all corresponds to the set of chart of components.  And in order to make more concrete idea of what we are doing, I put handout on back of this room, so if you don't have it, this particular one, I just include it few statistics with    we collect it from the Web.
So this is especially relating to the diverse issues.  So how the domains, ccTLDs are used for local residents to make them available    make them available on the net.  And so far, what we found from this so far is still very, very limited use of local, as seen in the Africa domain.
As seen on Asia domain, but in case of Asia, some higher    in case of Asia local languages.  But in case of Africa still, English, French or other    mainly used, the local language is less than 1% of the total event.
Only (off microphone) is extensively used in other countries.  Look at the right bottom slides.  So our data source is maybe three data sources.  One is our own data.  Collected by our own and the third one is from the international statistical sources.  So we combine this to make it usable, meaningful set of indicators.
So far our calculation of the indicators is not completed, but I hope in a month's time we can finish our work, so I    if you send me your interest and me or to    I will    once we finish our work, I will send you our report.  So thank you very much.
(Applause).

>> ALICE MUNYUA:  Thank you very much.  You can check the results to    in the back of the room.  Very interesting, you can go through that and see how many TLDs are using the web site in our own local languages.  We really need to do much in terms of this.  
We're now going to have an open mic session, questions to the panelists, a question and answer session.  As more are coming, let's be asking ourselves what we mean by strengthening ccTLDs.  We need to answer these questions.  So, welcome.

>> ALAIN CARISTAN:  Hi, this is Alain from Africa.  My question is regarding from the presentation from that the ccTLD now looking at the content and the domain name, and my question is going to be    what I say, the local law allows you to do that and have    you already have issues, legal issue with domain name based on the content?

>> ALICE MUNYUA:  Could you repeat the last part of your question?

>> ALAIN CARISTAN:  My question is, my issue from working domain name because of the content, have you already had such experience?  The question is    we did    you did a study, so the question is for ccTLD, domain name location based on the content?

>> ALICE MUNYUA:  I didn't look into individual case where domain names are blocked, but we just look at the principle they have and some cc do have principles for    yeah, for blocking or discouraging content level, and in a case of .jp, Japan, they do not have an explicit policy into content, but they do cooperate with local law enforcement agencies when spam or phishing cites are present, so they persuade them not to distribute illegal content.
Or in some cases, if I understand, the registrant's information is    they're able to close down the domain, that's how it works.  I'm sorry, I do not understand in actual individual cases, but that's what we have now.

>> ABIBU RASHID NTAHIGIYE:  Actually, if I can add, I know for sure from our region, from the Africa region, there are quite a few ccTLDs that have a list of what you would call morality or public order names or words that are black listed at their level and sometimes at their registrar's level and they give the registrar's    or if the registry's name directly, they are the ones who basically decide if a name is okay or not to go under a ccTLD.

>>  ALICE MUNYUA:  I want to add a little bit and throw back the question to the Kenyan policymakers.  For example, in Kenya pornography is illegal.  To look at what might be introduced that begin to deal with issues of pornographic content, how we going to deal with that, we don't have a policy yet that helps us at the CCK.  It's already dealing with content regulation very specific to broadcasting and to, I think, a new area that was used to generate content, so I think what does that mean for us at the ccTLD level and are we going to be seeing that framework being adopted or being used or    being used at the ccTLD level in types of domain names that carry content that we may think is, you know, malicious or does not adhere to our public, you know, morality and what we consider content that is    you know, that could be subversive    especially when we begin to look at Internet, a very new way of accessing the Internet that we see is going to    I mean is developing within the Internet context, especially with the introduction of eCommerce.

>> JOSEPH KIRAGU: Okay.  Just to that, the registry also had profanity fee, we block some strings    some streams of particular strings not allowed because of morality.  On the content we don't have a way of controlling so much, you could have a string which is okay, but the content is somehow not good.
So we try to do a lot of mediation, try to talk to the registrar and find out whether we can pull out the domain.  We had a case where the domain was registered KA0.K and the content that was put there was to the head of the Kenya authority, so we had to interview and do some mediation to try and move that content, but the policy is not clear in terms of that.
So I think most cities are struggling in terms of the policies we need to put in place to control the content they put on their web site.

>> ABIBU RASHID NTAHIGIYE:  Well, similar place for Tanzania, we don't have policy issues and we don't have specialist whereby certain domains can't be registered.  We use our experience where this domain does not make sense.  
At some point, for instance, there was one particular example whereby someone wanted to use a domain name using a president's name, so we had a content check.  If it was a real authority to register such a domain name.  There was a malicious content.  
Sometimes we get request from owner of the web site.  Some web site from    sometimes if they request, we do assist.  Sometimes we find without even receiving a request, if there is malicious content on a web site, we need to take some actions.

>> MARK SILBER:  I'm Mark Silber, South Africa Authority, I serve in a similar position on an international organisation.  I'm really fascinated by this approach to content regulation within the Africa ccTLD community and I've got to throw it out to my colleagues here.  Is that really the job of the ccTLD administrator, or is it just because they're on to other people in the country who understand the issues and so you're getting involved because if we're ever going to move towards significant automated systems which can handle the volumes of registration which our country should handle, then the ability to do things one by one by hand and to actually assist    and I use the example, I thought it was a very pertinent example, of somebody trying to register the president's name, you know, in an automated system you might have put that into some form of watch list of words which won't be registered and you and your staff will have a fun afternoon possibly with a few beers coming up with as many bad good key or objectionable words to throw into that list, but what you can do, you can't actually sit there and watch every registration as you come in, otherwise you need a stock of hundreds, thousands.
It seems to me you need an approach centred on keeping African ccTLDs as lifetime home cottage industries, craft industries where people are doing this by hand instead of having an automated system and process.  I'm also absolutely fascinated when you talk about a ccTLD manager getting involved in processes to remove or mediate objectionable content which may not even be hosted in the country, depending on where the main servers are.  Do you really think that's the job of the CC manager?  
Shouldn't the job of the CC manager be offering quick, efficient reliable name registration and leave content regulation to other agencies?

>> ALICE MUNYUA:  I'm not a registrar, obviously.  If my presentation gave you some impression that I am a proponent of content regulation based on domain name, that's my fault.  I'm not    I'm not saying that content should be regulated on domain name level.  But what I am trying to say is that when illegal or harmful content is online and    and blocking domain name resolution may be one of the solutions to it, how can we do that?  That's the point.
When law enforcement authority may want to stop it, currently we do not have a    you know, legal basis to revoke the domain name resolution.  So what I am trying to say is that put some system in place so that we can do that when we really need it.  So that's    that's my position.

>>ABIBU RASHID NTAHIGIYE:  :  Okay.  Also to answer that, I would say most ccTLDs are unique, depending on their country rules, so some see that as a deterrent.  Some of them are regulated depending on their authority, and some of them have to regulate particular names or trade, something like that.  They must have a way of    a way of blocking particular names.  For some    every city is very unique, depending on the kind of laws they have in their country.  

>>  ALICE MUNYUA:  Thank you very much for that question.  It's a very important one to think about as we begin to look at policy and regulatory.  Again, to asks ourselves, do we    ccTLDs really need to get involved in content regulation?  I think personally, my answer is no.  You know, because there are other institutions, you know, that do that.  Like Joe said, the regulatory authorities do that.  But again, to what extent should the collaboration be?
Who, then, has the authority, the legal authority to pull down a name or to request for a name or recycled content to be pulled down?  So it's a very good question you have to think.

>> Yeah.  The register is to concentrate for the registry services and this should be left to the regulator.  But the registry should be able to if he's requested to do so.  There's a new act called the Electronic and Postal Communication Act which are being drafted or reviewed, so this will be a subject that should also be discussed, it came at a good time.

>> PHILIP OKUNDI:  Yeah, I'm Philip Okundi, I'm Chairman of Commission of Kenya.  I think this issue of regulation is one of those which not in the past as we see in the CCK.  We have recently introduced content regulation, particularly in the broadcast sector, but I can see a move towards putting content regulation on the ccTLDs.  Because see, this area is already growing so fast and before you realise there's a server moving the place, and people say what are the structure in place moving this kind of place.
The people register the ccTLDs, they should try and put a wide spectrum on the question and try to restrain the    the operate of guy who's register not allow or not be able to register who's going to do some things or whatever, but it is an area which is still quite green, but I think regulators will be moved to make some regulations quite quickly because it's growing so fast.
I just saw recently there was a report on Cybercrime which is marking one of the countries of Africa now leading in Cybercrime.  Something which surprised me so much and surprised a lot of us.
The other point I would make is the presentation of Mr. Muruiki, sorry about the name, but you seem to say that registration of the ccTLDs are related to economic growth of our country.  The more economically able the people, the greater the technical license.
I seen from the presentation, that when they started the tzNIC, the level of participation went down.  Now, I just want to ask you, I saw that people were very good in the beginning but later on they depend to fall down in the need to register or take license.  But isn't it so that the youth, you know, the growth of the youth in our countries is getting on the Internet so fast the need to put the environment the need for them the operate with the information becomes higher so sometimes the economic state of affairs of the country, the pressure to register is very, very high indeed.  Is this considered in your study in
Lastly, ISP's, in Tanzania, I don't know who's hosting it.  I think there's a problem here.  We are trying to see if it can be hosted by Kenya or some independent body or by whatever.  Is it good that, say, a country's should host it or should it be hosted for another place or what do you consider in this particular case and what are the other advices?  Thank you very much.  Sorry.  Sorry for long time.  

>> I want to ask, tzNIC, your presentation, you're planning to deploy in 2011.  Why?  We fought and got site.  Why should we wait to 2011?  Why?

>> ALICE MUNYUA:  Good question.

>> ABIBU RASHID NTAHIGIYE:  The DNS deployment depends on the bandwidth.  We had problem with the bandwidth.  So I'm improving the bandwidth and deploy the DNS.  On the ISP, the ISP operation are managed by TSPA.  This is association of Internet Service Providers, but it's also facilitating the process, and granting some advance to direct the ISPs.

>> KEISUKE KAMIMURA:  Let me answer to one of the questions to me.  Unfortunately, in our economic analysis, we did not find any cause of relationships between economic development and the level of    and the number of registrations, so we do know these two elements correlated, but we do not know which is the reason for the other one.
So that's one of the reservation of our studies, but we    I think we could say that for a    for a good ccTLD governance, we can observe correlation between economic development, domain name registration and other governance elements like participation in ICANN processes or employment of uniform UDRP processes and so on.
So we can say    we can't say which one is the reason for the other ones, but we could say to be able to    to become a good ccTLDs registry, you have everything, so that's    that is the core message of our study.

>> Okay.  Let's take the question from Erick Ahon LACTLD Latin American region.  We talk about the statistics.  It is not clear any kind of question between the economic development and the number of ccTLDs domains, you have the cost like maybe one of the    administer to my names.  The other case is something like local identity.  In our case it's local monopoly.  My question is if we can't define that, how is your proposal to set    if we take your will be a good ccTLD, what mean good ccTLD in the case of all our different context and different problematics, not the same of (off microphone) and Japan?  The second is (off microphone) Africa America and    the same problem the ccTLDs, the government fry to increase a number of names, relation with our local communities, so how we can improve administer information between the Latin and how we the    and final question is how is possible as a member of the government a few minutes ago, how is possible to contact the national laws for some things with the national policies or the policies of the ccTLDs because the only real policy for ccTLDs is the 1591.  The ccTLDs have set the laws in some parts but what happened if the ccTLDs are outside that country?

>> I have two questions which followup into Elaine's questions while we are waiting, it's posed directly to tzNIC and mine is general.  I missed most of the presentations, so probably this was answered.  Mine was very generic to the African reason with relation to DNSSEC.  I'm looking at the point there's a lot of interest in e government going online, are they actually asking foreign enhanced security in the DNS front?
And the second is what are the DNSSEC now that we have the roots, what are they doing    what are the DNSSEC and what are the research that was done, did they find any interest on DNSSEC?

>> Okay.  Thank you.  I'll try also to answer a few on the DNSSEC issue.  The issue of interest from the community, we're not seeing that coming from most ccTLDs, that is not coming.  We don't have the registerants asking for that.  It's up to ccTLDs to clear that awareness.  It's up to banking sector and online business sector to conducting that transaction, we DNSSEC, we    for example, today if we say, (off microphone) what happens with the registerants, so the    it is not there.  So they're not talking to enjoy the full opportunities of using DNSSEC, so it's something they need to take up and clear the awareness in their communities.  Just like tzNIC, we are ready.
A few registrars before we go to full deployment so we show them the impact of DNSSEC first.

>> May I answer the questions from Erick?  What is it about ccTLD, good may not be the right word to refer to here, so    but ccTLDs have some responsibilities.  A couple of responsibilities in the domain name field, and major ones are technical ones, I suppose.  So they should provide reliable    technically reliable and stable domain name service.  But recently as the Affirmation of Commitments documents mention, consumer benefits or something like that, newer thoughts were being introduced in the domain name governance field, so ccTLDs or any other registries are expected achieve newer goals as well.
So for a good ccTLDs, they should respond to these traditional or conventional needs as well as newer requirements.  That's my understanding.  And why ccTLDs policy and national laws should be aligned?  Well, ccTLDs are given some kind of privileges in the name of national sovereignty and self determination?  Why not.
And what was the second one?  I think I missed something else.
Okay.  Maybe I can respond to you later.

>> ANNE-RACHEL INNE:  I actually want to comment on what Erick said.  I want to say a good ccTLD, a well governed ccTLD you serve the community.  This is not ICANN's ideas.  This is things I'm saying right now are really coming from somebody like    who has been in the domain for the past 15 years.
One of the things    some of the issues we have in terms of governance in Africa is just a fact, the sheer fact that to start with, the ccTLDs are hardly recognized at home.  I really would like you on a household basis to go to the government people or the community to say what is the domain of our country?  Half of the time they wouldn't be able to tell you.  Why?
Second, when they know about it, it's just a    the casual way, it's been managed by an organisation that does not take it seriously.  It's being managed by people who, basically, are doing it practically as a moonlighting job, more than a real day to day job.  And I see that every day.  Rules and regulations aren't in place.
You can't be thinking about competing against a dot com or dot original.  That is 24/7, 365 days a year online.  When you hardly have any electricity, when your primary server is down half of the time.  When basically, me as an entrepreneur, I can take a domain from my own country, I could be subject to my regulator at any time thinking, you know, somehow I've either got to have the right content or    and I've seen this    or I no longer belong to, you know, the good community because I used to be in the opposition, you know, and the government has changed so my site needs to come down, hello.
It's either we have sound rules, okay, that people can abide by that, you know, that actually give predictability in terms of the way the ccTLD is managed, or it will never really pick up.  The one reason why    we all said it    the one reason why people all go to dot generics, you know, is because they're easy, less expensive and they're predictable.  Okay.  If something happens, you actually know what the recourse way is.  You hardly have that at home.
Okay.  One of the issues that we have is also the fact that, you know, the local content that is    that could generate, actually, traffic and interest in people is not there.  Three quarters of Africa languages are reaching with the ASCII script.  There's hardly any African content.  African language is content, except as he said, from Arabic, you know.
A lot of our languages can be written either with the ASCII script or Arabic alphabet.  If you keep the northern part of Africa away from the rest of it, there's hardly any content in local languages.  There's not much to do to make it happen.  You know, most of our    the    you know, the critics that our languages are using are codified.  It takes the ccTLD a little bit of effort to do it and offer to people that again have it.
But then, if the ccTLD operators are not in charge.  If they're not really giving the means to function, they can't be thinking about that, you know.  Some of those that are in universities, how are universities managed in Africa?  Subsidies from government.  Do they put it in wiring the university, or do they give the money to the ccTLD operator, you know?
Some of them are managed by telcos.  What are they doing?  Telcos in general are thinking given that we're prioritized and this is not one of our assets, we're not putting money in it.  
So these are some issues that we're having and, really, the governance of a TLD absolutely requires that we think about rules and regulations in terms of what Mike was talking about is about the government putting rules in place in terms of, for example, you know, codifying what, as a local law, what can be a name protected because it's a geographic local name, because it's a heritage local name, because it's not    it's not for a ccTLD to do it.
When it comes to disputes later on those people aren't in charge of that come to a ccTLD operator to say this name wasn't supposed to be there.  Did you verify the person who has it actually was legitimate?  You know, so it's really    an operator, you know, can do a good job if they have the means to.  
One of the things we need is for the community, the government, the business, and everybody to be behind it.  But also, you know, for the TLDs to really come as multistakeholder operations that are there for everybody's sake, and not just one among the community.
It's the only way it can work because that's the only way you can trust.  If nobody trusts it, nobody will go for it.  So that's    that's what I wanted to say.  Thank you.

>>  ABIBU RASHID NTAHIGIYE:  Maybe I touch on the issue of a good ccTLD.  This could be on maybe customer perspective, can be on the good services like time to register domain, competitive fees.  As I presented, some people have notion that .ccTLDs are smaller than .ccTLDs.  Once they release a domain and realise they can work, technically it's the availability of the system based on the condition within our countries.
So if you can manage a ccTLDs around the clock and give registration in reasonable time, as I say in Tanzania, one hour time domain registration, they can say it is a good as a ccTLD or not.

>> On the deployment of DNSSEC, for sure it is a challenge.  Even awareness of .ccTLD domain name is still there.  Some have different perception.  So you can imagine for DNSSEC issue, but ccTLD need to lead the way.  I think we can    we can as well continue, the e force with publicity on .ccTLD as well as DNSSEC deployment.

>>  Okay.  Thank you very much.  We've come to end of our workshop No. 63, Strengthening ccTLDs in Africa.  Thank you very much for participating.  Let's give a round of applause to our presenters today.
(Applause).