Do you remember when we all abandoned vinyl records because the music industry told us that CDs were much better? I remember watching Tomorrow’s World where the presenter conclusively demonstrated this superiority by spreading honey on a CD and then playing it.

But he didn’t say that CDs don’t sound as good as records, even when coated with a delicious layer of honey. Wasn’t that the point of listening to music? But the industry could charge us all over again for the music we listened to already and, really, that was the point.

In my last blog I talked about the importance of objectively assessing and preparing for the risks and real costs of introducing new technology such as virtualisation into the business. It’s even more important to do this when an industry pushes a new, often immature technology onto otherwise unsuspecting customers. There’s no better example of this push than VoIP and IP Telephony. To illustrate this, let’s explore some of the issues that VoIP adopters have experienced around cost, capability and quality.

Cost Savings

Everybody knows that moving to VoIP and IP telephony saves money over traditional voice solutions, right? We’ve all read those vendor whitepapers that tell us so. The real answer, of course, is that it depends. It depends on what type of business you are, what you use telephony for and how much, and what type of networks you currently have. If the combination is right, then you should save money. If not, then you won’t.

I was talking to a customer recently who said that she’d like to move to IP Telephony but “we’d have to replace all of our head-office switches, plus the handsets are too expensive – I can buy my existing handsets for £9 a go, with VoIP its upwards of £40”.

So where are the savings coming from in that scenario, for migrating head offices to IP Telephony? Well, vendors will tell you about the considerably simplified moves and change management. In theory that’s true because handsets take their telephone numbers with them when you move them, but in practice I’ve seen a lot of IT organisations wrestling with immature call management IP Telephony software too.

In fact, the relative immaturity of VoIP solutions almost always means that IT organisations spend the first 6 months after deployment chasing down voice problems that just don’t happen with traditional solutions. But I never read about that in those vendor whitepapers.

When we move from intra-office to inter-office voice VPNs we come to the classical assertion that VoIP saves money because, whereas you needed two networks to support your data and inter-office VPN calls, now you can share the same network.

But that’s only true if it’s the right network. Anything below an MPLS WAN isn’t going to support business-quality voice (business won’t tolerate the voice quality you get when you phone your Auntie in Canada on Skype), so you’ll have to replace your existing WAN in these circumstances. To carry voice traffic successfully, you also have to pay a premium for that bandwidth (it has to be Class 1 QoS MPLS), so the savings calculations, including capital outlay, become more marginal.

So while VoIP definitely can save money, you really have to get down to detailed specifics to find out if it will save money in your particular case. What will the capital outlay be for intra-office and inter-office equipment and networks? What exactly is your current call profile so that the WAN can be correctly specified for Class 1 bandwidth? Have you allowed for greater trouble-shooting costs in the early months after deployment? And so on. People who don’t make these calculations always have some explaining to do later to their CFO.

Capability

Clearly, IP Telephony can bring some great CIT benefits in specific settings like call centres. But for the average office? I remember, whilst working for Cisco, that we had a competition to think up services that could ship on those expensive VoIP handsets. In the end this amounted to checking stock prices, getting weather forecasts, and playing funny noises as ring tones. Of course, it was just dumb – you’d use the PC on the desk next to the phone to do all of that stuff, but it did illustrate that it’s pretty hard to think up meaningful IP Telephony services for the average user to justify those handset costs.

What we did get were handsets that had to be rebooted every so often, like PCs, because they crashed. You never saw that happen when Holmes phoned Inspector Lestrade in Hound of the Baskervilles.

Quality

And finally, let’s get back to the question of sound quality. Did you know that, if you measure the quality of popular VoIP codecs against the industry-standard MOS voice quality scale, they score at best “some users dissatisfied”. ‘At best’ means that the network and equipment carrying the voice signal introduce no further signal degradation, which, in practice, they do. So, in promoting VoIP, the industry has quietly reduced the quality of calls because the technology can’t support the previous standard. Maybe a layer of honey could help.

What it means to IT organisations is that business users will perceive a reduction in the quality of voice once switched over to VoIP.

So does this mean that VoIP is an inferior technology to existing circuit-switched telephony? Of course not – VoIP can bring real savings and real flexibility to businesses. It simply means that VoIP adoption has to be approached carefully, costed accurately and that, like all choices in life, it involves compromises.

Honey on the other hand, is nutritious, tasty and it’s good for the environment too.

Here at Sumerian, we’re loving the World Cup. This is despite (or for some people, because of) England’s poor performance. We’re running a Dream Team competition and organised a Sweepstake which pays out if your team reaches the semi-final. But these things are notoriously hard to predict – spare a thought for Alastair, who drew Brazil and must have thought he was a dead cert to win something, or Roly, who I personally assured when he drew them that Italy stood a great chance of going far in the competition. And who would have predicted a Uruguay-Holland semi-final even on Friday morning?

But what has this got to do with IT? Anything?

Maybe nothing, but maybe your IT estate has been affected in ways you wouldn’t have thought of. How was your Remote Access Service on Wednesday 23 June? That was the day England played Slovenia at 3 pm UK time, and a large number of employees (uncoincidentally) chose to work from home that day. Would you know how much extra load this would cause on your RAS system, and whether it can cope?

Of course, not all companies are capable of letting people work from home. Some companies chose to let their employees finish early, or showed the matches in the office, but most expected their staff to keep working. This is the first World Cup where both the BBC and ITV have been streaming matches live on their website, so some people will have had one of them running in the background on their computer. Video streaming can be quite intensive on bandwidth, and many connections have a Class Of Service configuration on them, which prioritises video traffic over voice traffic over data traffic. That’s all well and good when the video is a WebEx conference with a client, but when half the office are watching football, you could find that the transmission of important files are being delayed due to traffic from the BBC website.

Some savvy IT managers have rigged up systems whereby a single video stream is downloaded to the office, then everyone else links up to that, cutting the traffic on your main internet link to a minimum. But then, at least the timings of the fixtures at the World Cup are predictable.

Last year, when Michael Jackson died, the internet usage at one of our clients went sky-high. When news breaks, everyone wants to be reading the story and watching the latest interviews. You can’t predict when it will happen, but you can plan ahead and be ready when it does.

IT systems will always have an element of unpredictability to them for as long as they are used and maintained by real people living in the real world. Maybe we can’t tell you how many of your employees would have been working from home the morning after, if England had played in the final, but at Sumerian we can tell you how many your existing RAS system could cope with, and where the problems would occur if you had too many.

Not that it’s going to be a problem this year. Cricket, anyone?

Posted by Nick Maher, Sumerian Analyst