Response to Abel Girmai

                                                                                (Anxiety of Cloud Computing Just does not Compute)

First of all, I would like to thank Abel for the detailed information about the project and his overall effort to make it a success. I now know the project is multi-dimensional and that cloud computing is just one part of it.

If my prior comment on the issue of cloud computing was the reason for your response, I don’t understand the motive for the negative tone of your introduction and conclusion. My comment was simply an invitation for an honest discussion to explore this fairly new technology in an Ethiopian context. In fact, we are both on the same page and I really applaud the project until you get to the cloudy part and it’s good to learn from your article that the proposed project does not entirely depend on cloud computing.

Referring to Ethiopia as a country on the other side of the globe, Seattle Times dated July 13,2009 reads as follows…“As U.S. companies begin exploring cloud computing this year, a school system on the other side of the globe has already leapt into the cloud.” (emphasis mine)

When I read this excerpt under a heading titled “Microsoft cloud gets down to earth”, coupled with a lot of cautious IT experts’ opinion, it’s only natural to raise many questions and doubt its feasibility, when the first ever landing spot for the cloud is in Ethiopia and that I think is not being pessimistic, it’s being cautious. Besides to the security, accessibility, and availability concerns that everybody is uneasy about, when I see Ethiopia with the current weak internet infrastructure jumping to use the cloud technology while US companies are reluctant to use it, it makes you feel like it’s being used as an experiment.

As you indicated storing web mails in servers provided by ISPs is an application of cloud computing that we use every day. But here we are talking about data of an entire student population of a country not about electronic correspondence between individuals (e-mails). I wouldn’t care about the physical location of the server that is hosting my e-mails as much as I would about my documents. This attitude might change in the future but most people at this time would like to have their documents on their hard drive instead of relying on a browser to access it.

As somebody else put it:  “Cloud computing is using the internet to access someone else’s software running on someone else’s hardware in someone else’s data center while paying only for what you use.”

Again, I would like to ask you regardless of the short term cost associated with building the infrastructure, what you think about the option of setting up a ministry- based data center in Ethiopia instead of using somebody else’s data center outside the country. That way you can navigate through the Ethiopian clouds to access the data. That was the question I posed in my comment raising doubts not about the entire project you have described that we all want to see succeed, but about the cloud computing part and the intention was to solicit opinions from professionals in the area.

You noted security issue as one of the disadvantages of cloud computing which can result in a loss of data, and as you know they probably won’t be held liable for that. That’s not to say ‘security is not an issue if the data center is inside the country’. But at least you own the problem and you deal with it the way you see fit.

The issue of accessibility becomes a concern, if for any reason the country defaults in the future on its scheduled payments for the services. If I miss two monthly payments for my internet service at home, it will be disconnected and I will lose access to my e-mails and the problem would be severe if I had my documents stored on their server. It’s as simple as that. You can imagine the severity of the problem when you have data of students of an entire country stored on their server.

After what we saw what happened to many companies that were once considered reliable and heavy weight in the current financial crisis, the issue of availability becomes a concern. Nobody knows how long the company that’s hosting the server is going to stay around. I am not saying this will lead to total loss of data, but what will happen afterwards becomes somewhat cloudy.

In conclusion, call me overly-cautious or conservative, but for the reasons I mentioned, I remain unconvinced of shipping data of that magnitude overseas instead of pushing to set up a data center inside the country. This was what motivated me to write a short comment to begin with, not any other part of the project you described and what you see is what you get. Again, Abel….it’s not what you think. This is not to de-motivate anybody as you implied in your conclusion. It’s rather a difference of opinion on an issue.

Thank you,

Abrerom 09/02/09