(Christopher Atherton)
Convener notes on the chart pad
Reach the idea of peak science, data sources come from multiple places. 3 billion people in Africa, a massive gap in capability, in addressing how app countries can participate together while Africa can’t. How do we start to address those problems, to address this gap?
To what degree do you think this is a technology problem vs a political problem?
1000’s of students but 10mbit connection… most universities don’t provide their academic username and password so lot of these get the students to setup the yahoo or Gmail addresses to do research. But for us, the institutions that have that capability can’t work in those countries like Africa that doesn’t have the same abilities.
EduGain, things are changed a bit now, eduGain is a metadata distribution service, it allows an identity federation to be created and the identities to be basically used between an identity provider and a service provider. If you think of the end goal, universities have collection of identities, and universities perform research activities, we are trying to do cross institutional cross research, eduGain is a layer to facilitate the research.
So, there is a certain amount which is for structure provision, if that infrastructure was provided, would the adoption be provided automatically. The African situation is a much political than technical issue.
Even for universities where they have the email system researches don’t want to use their identities.
Even if you accept that the SP side is still benefiting from federation. The diversity exists in Africa. So, I think we need to adjust the lens a bit to look at the detail. There is a number of initiatives of parties trying to access the reach.
SKY is putting up a 5G connection using pseudo satellite platforms, in a very stable part of the stratosphere, you can stay in the position very easily, at very low cost to get high bandwidth around to the people below. Obviously there have been attempts and they have worked.
Problem is getting connectivity for NRENs in Africa. Fiber submarine cables to NRENs in Africa that connect up to Europe costs a few million euros. It costs the same amount to go a kilometer inland once that cable lands in Africa. There are monopolies in place that prevent us from providing that service. Universities that are a km away from the shoreline. That’s why the 10mbit link is as expensive as the submarine.
Geographically, it’s challenging as there aren’t that many countries at the coastline, getting fiber through is very difficult. Getting the ISPs to understand that this is not about profit is almost impossible. A lot of the people that would end up being a commodity internet user are those that have been to universities. It’s really challenging to work with ISPs, specifically in Mali but in Uganda it’s the same. In some ways they stall their progress.
We have done it in Latin America. There is only one country who didn’t join RedClara and that was Bolivia and they wanted to do their own thing. Even in east southern Africa we managed. GÉANT helps to procure an IRU down the east coast and now they invested their money to build fiber infrastructure in their countries, they can afford to reinvest. They go into support for the universities and the fiber in their countries, but the next step seems to be trust and identity services.
Most of the region participates, east of India to Japan but not China, there wasn’t a heavy campus IEM infrastructure. Building baseline materials and now they are at the point now that we have that what are the tools that we can use like opensource packages. Since you mentioned South America, I am trying to think about what is going on there. Some email traffic from Asia or pacific, the only thing I can think of are the Chile research groups regarding astronomy. Not sure if there is participation from universities.
A little bit that I have seen is that we need the telescope in the federation. Wondering if there are lessons learned from APAN that apply for other regions.
Sounds like a capability challenge. We need to implement some processes. Materials to push the technology out and there is a probably a core of assistance stuff that can be packaged up. It might not be English as a de-facto choice.
There has been some of that point on, there were materials developed for APAN, if nothing else is a starting point, their choice was interestingly enough English. For the admins and who shows up at the conferences so far it has been ok.
Albert: I know very little about this, we got into the Africa bit. The rest of the world, we have been approaching this through NRENs, it doesn’t sound like Africa. It sounded like there is a strong mindset challenges that part of the world things about. They have different priorities and they perceive technology much differently.
Getting clear messaging out to the right communities is really important are not the folks working in IT but the government departments.
A: He gave two examples, in Namibia there are two universities, but as there are two universities it doesn’t seem like you need to have an NREN. IF you don’t have needs to collaborate in the nation itself.
So, you think it’s a conceptual need…
A: In the US we keep talking about the longtail problems but then you have the very small universities that don’t have the resources. I wonder if that idea can lead into something for universities and colleges that don’t want to run their own thing. I wonder if something like that would work in these regions. You need to provide a complete packaging.
To generalize is the existing tooling the appropriate tooling. DO we really need university credentials for that kind of stuff?
An example: Uganda, my convo with the CEO was saying how important was that he was able to get researches to be the voice of the NREN, it has been clear to me working in Mali, IT people finding out about eduGain, but they are not often the best people to sell it.
A: Language can be an issue looking at the map. Is there anything other than trying to provide support to NRENs?
Can something else be done?
One of the discussions we had before is that connectivity is pretty low and that’s for every single student. If we were trying to do something that is a centralized, nationalized IDP service that might fail because of the lack of infrastructure?
Doesn’t that get into a much more, why this is an issue? Back to the US, part of the problem is that the colleges that is not just about money, but they don’t see the purpose of the research, because they are community colleges.
In regions of the world where internet speed is inadequate, is there even a motivation to do this?
The idea is what at the moment research collaborations is for the northern hemisphere, they end up racing ahead. There is disconnect between the research that they do and the north. ESO are headquartered in Europe and their telescopes are in Chile, the data travels from Chile to Europe. There’re only a few researchers in Chile that take part in that.
I think that there are interesting projects in the south and that those collaborations can be good in some ways because it faced similar challenges. At least in that case they share the language.
A: Is the desire to even the playfield coming from the south countries?
In some cases, yes but not only. Botswana is one of the richest countries in Africa but no eduGain or NREN.
Some countries just choose not to invest into it. We as countries think that higher education is important, so we invest into that.
Some US universities want the single sign on, but they don’t want to support research.
There is a traditional model that NRENs fit into for service-based economies. Perhaps for resource/extraction-based economies the is a need for another type of infrastructure provider for education which is similar in function to an NREN.
