Last week, the IEBC called IT-savvy representatives of political parties to a meeting on Friday at the Sunshine Holiday Inn in Westlands, Nairobi, where the use and efficacy of the equipment to be employed for the election counting and reporting was to be demonstrated. A representative of the US National Democratic Institute (NDI) was also present.

Party participants had a whole range of questions on which they wanted reassurance from the IEBC, especially concerning potential failures carried forward from the past.

To demonstrate how the new system would obviate all this, some of the participants at the meeting were divided into five groups of three, each group a mock ‘polling station’. They were given mobile phones such as those to be used on March 4. The phones are loaded with the software and menus for completing the tasks at hand.

The remainder of the group sat watching the screen, waiting for the ‘results’ to come in.

That’s when the problems started. The five ‘polling stations’ were initially all unable even to log in.

After a few of the five did eventually manage it, the next problem arose. They were logged in but there was no connectivity with the ‘tallying centre’. The ‘polling station’ callers could not be authenticated.

Finally, after struggling for ONE HOUR, only ONE of these five ‘polling stations’ managed to transmit its results.

Now, much as we want to trust that the IEBC is going to do a good job, we have to ask – if four out of five ‘polling stations’ have problems in a demonstration meant to show the efficient use of this technology, what on earth is going to happen when 33,000 polling stations all try to log in and transmit results at the same time?

We hope the IEBC is trying to fix these problems but, on the basis of the evidence so far, and considering the sheer volume of the data to be transmitted, it certainly appears that there could be a massive system failure.