Home & Office

Fit for purpose, not just for headlines

With the OPEL bid cancelled and procedural questions dogging the FTTN bid, Australia is currently in something of a technological limbo.
Written by David Braue, Contributor

With the OPEL bid cancelled and procedural questions dogging the FTTN bid, Australia is currently in something of a technological limbo.

ADSL2+ rollouts are underway, but Telstra is already declaring it obsolete with talk of 100Mbps services.

Unwired bravely committed $500m to a national WiMax network, but prominent analyst Paul Budde is already labelling the plan unworkable. And CDMA, which was working perfectly well for nearly two million Australians, has been replaced with an alternative that is, depending on who you ask, either the same or far worse.

There is so much fear of investing in the wrong technology, it appears, that many feel the best solution is to implement nothing at all. Consider the recent outcry (read this editorial for one example) over the possibility that Melbourne's archaic rail communications network, which apparently cannot contact drivers or trains over more than half the network — might be upgraded with GSM-R technology.

Now that 3G is so popular, opponents argue, GSM is simply anathema to progress. But they're forgetting that GSM-R is an offshoot of the world's most widely used mobile technology, engineered to improve service reliability and provide group-conferencing features suited to the unique task of keeping thousands of tonnes of steel and human flesh moving at high speeds without hitting anything.

Why is GSM-R different? GSM-R networks maintain continuous service even on the superfast trains criss-crossing Europe at speeds of over 300km/h. At these speeds, traditional GSM networks struggle to keep up as handsets and terminals rocket between cells; consider how often your GSM or 3G mobile drops out when you're driving at just 60km/h.

GSM-R, on the other hand, is rated to speeds of 500km/h without communication loss; an on-board GSM-R modem maintains continuous contact with the network operations centre and stops the train if the connection drops. With fail-safe design like this, it is little wonder that GSM-R has been selected for use by rail networks in 38 countries.

The history of disasters surrounding Australia's transport systems — the lack of high-speed trains, governments dumping billions on unnecessarily complex ticketing systems, the installation of different rail gauges in NSW and Victoria — is well documented. In each case, one of the common threads is an almost paralysing fear of compromising technical perfection for everyday utility.

When did our fear of obsolescence become so gripping as to override rational thought? In today's environment of constant advancement, any technology is outdated by the time it hits the streets. That doesn't mean it's useless — but the claims of those who argue otherwise continue to pollute high-level discussions about every part of Australia's telecommunications infrastructure.

One of the common complaints about the now-defunct OPEL WiMax bid, for example, was that it would be superseded whenever the FTTN network was built out. I still don't get that one, since WiMax would have been available far earlier and provided much-needed broadband to many areas that I reckon still won't have a sniff of fibre five years from now.

This week I sat down with some technical types from Ericsson, around whose equipment Telstra's Next G network is built, and asked them to explain to me in short, simple words, why HSDPA was so superior to WiMax as a landline replacement (Ericsson last year cancelled all WiMax R&D, likely because of the technical success of Next G).

WiMax and 3G aren't that dissimilar technically, I was told, but issues of scope and infrastructure investment will prevent WiMax from competing with networks that are already in place. Coming upgrades to HSDPA will apparently allow Next G to provide the wireless equivalent of a copper local loop connection to the home — WiMax already does this.

"So, this is what Telstra is saying, right?" I pressed. "This is how Next G will replace the local loop as the primary conduit into the home?"

The answer was a little surprising: "Will you have an HSDPA connected set-top box delivering video to your lounge room? No," was the answer. "There just isn't enough bandwidth."

Excuse me, but isn't every analyst in the world rabbiting on about the importance of triple-play services for tomorrow's telcos? Given that nobody in the world needs 42Mbps for their mobile phone, why in the world is Telstra so hell-bent on pushing speeds up so fast, if not to position the network as a landline replacement so it can push richer and more profitable content to its customers?

Surely there is a point where bragging rights become less important than actual services. Next G seems to be a good network when (and where) it works, but Telstra has been so dogmatic about it that it just hasn't stopped to consider what many customers actually need — such as a decent, reliable Internet connection to their home.

People do still visit their homes sometimes, right?

No matter, though, because the coming FTTN contract will deal with that. And, as I pointed out to the Ericsson guys, that contract will provide an instant speed boost for the many ADSL customers who aren't currently getting the speed they should; after all, millions of Australians already have ADSL modems that can provide decent speeds if they only have to send signals over the 1km or so between the home and the nearest fibre termination point. Right?

I would have gotten a less shocked response if I'd been advocating a move back to telegraph services. FTTN will use VDSL2 to link the home with the fibre, I was reminded — VDSL is a faster version of ADSL rated to around 100Mbps if you're in the same room — and rather than supporting a range of ADSL equipment it's much easier to just give everyone new modems.

"But ADSL would also work, right?" I pressed, knowing full well the answer I would get. "For most people, FTTN would be ideal if they just woke up one day and their current equipment ran faster."

The answer basically went like this: it could work like that, and in fact Ericsson gear would allow that. But ADSL is passé, and VDSL2 is the future, so why bother with ADSL?

Actually, LTE is the future, and recent working demonstrations of the technology prove that Next G and every HSDPA network is, technically, already obsolete. Maybe all of our 3G minded carriers should forget about their 3G investments and start preparing for the real future.

Or maybe, just maybe, we should all just get over this hang-up about always having the latest-and-greatest, and stop for a moment to think about what we actually need.

Editorial standards