Supercharge your infrastructure

There are, it seems, 44 countries in the world with better broadband download speeds than Australia, according to the latest Netindex release by Ookla.

This has brought a predictable round of soul-searching, particularly as Mongolia appears higher on the list than Australia (it scored an average speed of 13.79 Mbps, while Australia could only limp along at 13.09 Mbps).

Without any context at all, however, it's hard to see the data as particularly newsworthy.

I'm not going to get into the ins and outs of methodology here, but before Australians gnash their teeth over being in position 45 on the speed table, there are some things worth considering.

It shouldn't surprise anybody that Macau, Malta, Jersey, Andorra, the Aland Islands or Liechtenstein perform better than we do.

Let's take the Aland Islands as an example. This tiny territory inherits its international connections from Sweden and Finland, and needs to provide services to a population of around 28,000 people in around 11,000 households. It's in position 29 and gets an average speed of 18.31 Mbps, which should surprise nobody.

The “Australia is large” isn't the only consideration I have in mind, however.

A crucial datum that's missing from the interpretation of the Netindex data is the household/business service split. Once again, I don't plan to try and data-gather for the 182 countries on the list, but I did gather it in 2006 for the short OECD list (still a daunting task that I don't plan to repeat unless it's for money).

A very broad-brush statement that I think still holds true: in a well-developed broadband economy, household customers far outnumber business customers. Look at Australia: including mobile broadband, we have nine million household subscribers and 2.8 million business subscribers.

In general, if there are more business broadband connections than household connections, it's either because services are too expensive for individuals, or because the infrastructure clusters around business – either case suggesting a less-developed nation.

Let's look again at Mongolia. Of its population is 2.8 million individuals, it has just 72,000 broadband connections (according to Wikipedia). This suggests two things to me – that its Internet infrastructure is concentrated in its cities; and that it's probably dominated by business users.

I'm pleased to see that Ookla eliminates the effect of international cable density by measuring speeds “where the mean distance between the client and the server is less than 300 miles”.

And with all that said, I'm still surprised that Australia scored an average download speed of more than 13 Mbps – and that this is fairly consistent over time.

About seven million of our 12 million broadband subscribers – fixed and mobile – are on services with an advertised download speed better than 8Mbps. Many of those are mobile services which (4G subscribers excepted) can't really expect to approach 13Mbps as their routine service level.

There's a million cable subscribers and 90,000 fibre subscribers, all of which can expect to beat the 13Mbps average speed, but that still leaves the DSL subscriber base to drag down the averages. Apart from a spread of subscribers across different packages, it's difficult to believe that the average DSL performance is 13Mbps.

Could it be that a lot of people who run speed tests are doing so from their office desktop, from behind a high-speed business service? And if it were true for Australia, might it also be true in other countries?

The Ookla results are interesting, and useful for advocacy, but I don't think they're a basis for Australians beating their breasts and crying woe. ®

Email delivery: Hate phishing emails? You'll love DMARC