VoIP -- is it ready for the small-time?

You don't get something for nothing. If it's cheap, there's always a trade-off. You get what you pay for.

All these familiar clichés are effectively saying the same thing, and it's hard to think of a market where this doesn't hold true. Exceptions, where they exist, usually turn out to be loss-leaders. And it's true in the VoIP market too -- and in a week where voice-grade equipment vendors such as Nortel are cutting back heavily as sales plummet, can VoIP rise to the challenge of filling the gap between expectations and reality?

You might think that it's a bit late in the day for this discussion and to an extent I'd agree. After all, the industry hype about 'the year of VoIP' happened a three or four years back. Now the dust has settled and VoIP has made considerable inroads into businesses large and small.

But the voice market -- or rather, the way that people use voice and the expectations they have of it -- is quite different to those in other parts of the IT forest. For example, connectivity does fail from time to time, due to external influences -- such as extreme bad weather, an over-enthusiastic trench-digging JCB or finger trouble at the local exchange.

We're used to that now. We don't like it, we get annoyed if not downright angry that we cant get our emails, but we accept that systems do fail. The same goes for almost all IT systems. We've been trained to accept that this new -- now not-so-new -- technology is not fail-proof.

VoIP's unique features
When it comes to VoIP, though, things are different. The trouble is firstly that voice is an intensely personal way of communicating, so we accord it great importance. The second is that the voice carriers have had a century to hone their circuit-switched systems to such a level of refinement that we just don't expect voice to fail.

And in general, it doesn't. Add to that the fact that many point-of-sale systems system still use modems, and fax machines -- those ugly duckling remnants of 1970s technology -- remain in widespread use, and it's clear that voice circuits still retain a number of unique advantages.

This is the challenge faced by vendors of VoIP systems. You've got to be bullet-proof, you've got to look like a voice circuit to a fax machine or modem, and you've got to sound like a voice circuit to a human being.

VoIP's big plus from an end-user perspective is that it's cheap, especially for long-distance and international calls. What's more, say proponents, VoIP's ability to add features, such as shared contact books and the like, are compelling.

So far so obvious. But the experience of many users -- most users? -- is that it's nowhere near as reliable as it's made up to be, and often, the quality isn't that great either.

Small companies need not apply?
Here's an example. One small company I know uses VoIP exclusively and, most of the time, it works fine. But every now and again, one senior member of staff -- the most technically knowledgeable person in the office -- has to spend significant amounts of time either liaising with the VoIP provider and carrier, or both, and fiddling with the router configuration. He's not paid to do that -- it's a waste of his time.

This example not unique, especially in companies with zero expertise on tap.

And what about those added features? As one commentator points out, "on a PBX system with 40-50 users, less than half will know how to use the recall button and you won't need the fingers of one hand to count the number who use hold or mute."

Of course, if you throw enough money a VoIP system you can have that famed reliability and probably high quality voice too. But for everyone but the large enterprise, there remain a number of caveats.

You'll still need a backup line to the PSTN for emergency calls and just-in-case. You'll have to be prepared to spend more time than you otherwise might fixing and fiddling with configurations. If you rely on Skype, with its highly proprietary technology, you'll still need a backup for when it goes down -- as the system did in August last year. And if you go for a technology that's a bit more robust and which offers business-quality, VoIP forums are awash with people having problems getting it to work reliably.

All this is not to denigrate VoIP generally. It's big step forward from what we had in many ways but we seem in many ways to be stuck in an interim phase, where neither the networks nor the software are quite ready for voice.

However, it is clear that, now VoIP has reached a broad mass market, expectations are high. This means the technologists -- largely in the shape of the vendors and service providers -- need to pay attention to user demands for ease of use, for reliability and for quality if the reality gap is to be filled, especially at the low to middle ends of the business VoIP market.

And a bit of training on how to use all the whizz-bang features probably wouldn't go amiss either.

Net neutrality to boost broadband prices?

The price of broadband is set to rise massively if European regulators enforce strict net neutrality, according to a recent story in the FT. It could go up by 30 percent, one analyst reckoned. This raises a number of issues.

But before we get into those, let's lay out the facts. The concept of net neutrality means that a carrier will treat all data alike. Whether it's from the carrier's biggest client, whether it's kids downloading illegal software, or if it's just your emails, all data is given the same priority. The principle behind it is that the deepest pockets should not determine which traffic is prioritised; it doesn't mean for example that your voice over IP phone traffic cannot be prioritised.

One of the fathers of the Internet, Vint Cerf, reckons that net neutrality is vital: “It's vital to innovation. Companies like Google and Yahoo, and eBay and Amazon, and Skype, and so on, got their start without having to get permission from any ISP or any broadband provider to offer services,” he said on YouTube.

What the carriers are currently bellyaching about is a proposal from the EU that net neutrality be legislated. In other words, it'll become illegal for a carrier to discriminate in favour of traffic originating from its biggest client.

The proposal is part of a new telecoms package that aimed at overhauling the EU's telecoms laws. EU ministers are due to discuss the proposal in November.

But consultants at Copenhagen Economics say in a survey to be published next week that the imposition of net neutrality would result in higher costs being passed onto consumers, and so depress demand for broadband.

They even provide sample costs: Swedish broadband rates could rise from €33 to €44 in Sweden and from €29 to €39 in Germany. That's quite a rise.

The Centre for European Policy Studies reckons in a report that the mandating of net neutrality would be undesirable. Compromises proposed include by the CEPS softer legislation which would ensure that consumers are informed about a lack of net neutrality.

However, whatever you think of net neutrality -- and I for one am nowhere near alone in being a fervent supporter of the concept -- I can't help wondering why prices would rise so much if the concept were abandoned.

I've always assumed, it would appear incorrectly, that we currently operate under a regime of net neutrality, that my data take their turn with everyone else's, and are dealt with by the Internet's routers in a first come, first served fashion. If the legislation were passed and put into force, and prices rose by one-third as a result, imagine how much money ISPs are making in return for preferring MegaCorp's data over those of smaller companies or, for that matter, yours and mine.

Would legislating net neutrality really make that much difference to prices? And if it would, why haven't the carriers told the rest of their customers -- the disadvantaged ones -- about it?

Can ZigBee succeed?

The killer app for ZigBee is energy management. That's according to Bob Metcalfe, to whom I've just been talking in Boston, at a gathering of the Metro Ethernet Forum.

Metcalfe is the inventor of Ethernet, the physical underpinning of pretty much every network on the planet, from your home Wi-Fi to the high-speed interconnects used in data centres and the international links that tie the world's telecommunications together.

Metcalfe's been a partner in VC firm Polaris Ventures since 2001, and one of his current projects is Amber Wireless, a company that develops ZigBee components. "As CSR is to Bluetooth, so Amber will be to Zigbee", was his comment.

CSR? Your mobile phone probably has a CSR-branded Bluetooth chip inside it -- the company claims to dominate the market for Bluetooth silicon.

So what's ZigBee? It's a 2.4GHz wireless technology that's designed to connect embedded devices. We're talking light switches, environmental sensors, solenoids and so on -- devices that currently tend to work standalone and whose operating principles are based on whatever is programmed into them at design time.

Metcalfe's a network guy so his take is that such devices need to be connected, not just for control purposes, which was the original rationale for the development of Zigbee, but also to manage their energy consumption.

It's probably fair to say that, since customers for ZigBee are designers and manufacturers of embedded devices, unless you're one of those, chances are you won't hear much about it. But Metcalfe's argument for ZigBee's likelihood of success is that there are some 10 billion embedded devices out there -- even more than the current tally of mobile phones.

Is he right? With the world now alert to the need for energy savings, it's possible. Metcalfe freely admits that the Ethernet label is slapped for marketing reasons on pretty much any networking technology these days, so the Ethernet name has in effect acquired a life that doesn't have a direct connection with Metcalfe's invention. So Ethernet's success outside of the local area network isn't all Metcalfe's doing.

As a predictor of the future too, his track record is mixed: he once reckoned that the Internet would be dead by 1996.

But then, no futurologist or analyst has ever been 100 percent correct and, on the energy front, it's more likely that Metcalfe is pretty much on the button. Whether Zigbee becomes a core technology in the struggle to sustain human activity and energy consumption on an overcrowded planet remains to be seen.

But his views on the future importance of video (plug, plug) must surely be spot-on though...