Why Communications Authority’s decision to measure internet service quality is welcome

In what is arguably a first in Africa and possibly the whole world, Kenya’s communications regulator gazetted regulations and guidelines on how it will be measuring the quality of internet service from providers.

Whereas measuring the quality of voice communication is a well established practice worldwide, many jurisdictions shy away from comprehensively measuring the quality of internet or data services.

This is mainly for two reasons.

The first is that competition is the best mechanism for raising quality. Subscribers voting by their feet is the best way to keep service providers on their toes in terms of ensuring they are offering high-quality services.

DIFFERENT QUALITY-OF-SERVICE EXPECTATIONS

The fear of consumers or subscribers moving to competitors’ networks ensures that operators are working round the clock to ensure that the internet is always available, has good speeds, is not congested and subscribers have excellent customer service.

These are some of the factors that the Communications Authority intends to measure, and will penalise providers who fall below the various specified performance indicators or thresholds.

The second reason most regulators shy away from measuring quality of service in the data market is that measuring internet quality is more complicated than measuring voice services.

This is because it involves measuring multiple services that have different levels of service expectation – and yet run on the same internet platform.

As an example, customers would have different levels of expectations depending on whether they are sending an email, an SMS, voice or videos over the internet. If a service provider delayed an email or an SMS from one subscriber to another by five minutes, the recipient may not even notice the delay.

EXPLOITING USER BEHAVIOUR

However, a five-minute delay for voice or video communication would immediately be noticeable and unacceptable. Both services run over the same internet platform – and yet have very different quality-of-service expectations.

Which performance indicator would one use to capture this quality disparity?

The general practice for operators is to dimension their network capacities to somewhat lie between these two extreme demands. After all, not everyone uses the expensive, high-end services that are bandwidth-hungry.

And furthermore, of those who use these high-end services, such as video-on-demand, it is unlikely that they do so all at the same time. It is this user behaviour that has seen operators get away with sharing a single link among three, six or even ten subscribers.

The technical jargon for this is contention ratio.

If a subscriber pays for a 1mbs link per month, it is not always true that he or she is the only one hooked up on this link. Most of the time the subscriber will be in contention or sharing the same 1mbs link with other subscribers.

SUBSCRIBERS AND NETWORK CAPACITY

The end result of an over-contention link is slower speeds, increased jitter and in some extreme cases non-availability of internet services.

The regulator’s recent move to measure quality of service is therefore welcome since it would help pick out service providers that notoriously acquire new subscribers without making commensurate efforts to expand their network capacities.

Another notorious move by providers is to market what seems to be affordable prices, while in essence they are piling multiple subscribers on very restricted network capacities.

Even though measuring internet quality from a regulatory perspective is a new practice, it has become necessary, particularly in developing economies, where competition may not have sufficiently matured enough to a level where it can sort out quality issues.

Lets us look forward to the first report and see if and how it would improve internet quality.

Mr Walubengo is a lecturer at Multimedia University of Kenya, Faculty of Computing and IT. Email: [email protected], Twitter: @Jwalu