“Digital Phones Are The Future” Anonymous 1976

by Dave Michels


Everything old becomes new again. Dumb terminals are dead (just don’t call a thin client dumb!). UNIX chat was so 80s (IMHO), and digital phones are doomed (to return).

We live in a connected world. The Internet has enabled our simple desktop computers and telephones to talk to all types of devices around the world. IP created an openness that today we take for granted, but it wasn’t always so. Microsoft’s Windows95 was their first operating system with TCP/IP built-in. Prior to IP ubiquity, data centers and corporate networks experienced major challenges trying to get their systems (DECnet, SNA, Novell, etc.) to talk to each other.

The telephone followed a similar but different evolution. The analog phone can still be used on most networks, but its simplicity only really works in residential environments. Business users wanted more features, so the PBX makers switched from analog to proprietary digital handsets. This standard lasted for several decades. For the past 10 years, the rage has been VoIP. Sometime around 2003 or so, VoIP phones outsold digital phones and continue to do so now.

The classic PBX vendors still offer digital phones, but the focus is clearly on IP. Mitel, for example, discontinued all digital phones but their primary workhorse the 4025. The VoIP phones are predominately deployed with proprietary protocols, but SIP deployments gain in popularity. Some IP “soft switch” manufacturers such as Digium and Broadsoft don’t even make phones relying on instead on popular SIP phones from companies such as Polycom and Snom.

I think the digital phones make a lot of sense and the [VoIP] manufacturers need to embrace it. Yes, I’ve been called a contrarian before.

This isn’t my first choice. I would much prefer the phone makers figure out how to make useful phones, utilitarian network appliances, something like a Blackberry or Android desk phone (see several posts in this blog about that idea here and here). But if you can’t beat them, join them. So if the manufacturers are determined to make stupid phones, devices with a singular purpose of voice communications, then digital solutions make more sense than IP phones.

The fact is, other than the six people that use the microbrowser functions, the IP phones used today don’t need IP to function. In fact, IP adds quite a bit of overhead to the solution. Click to dial, unified messaging, hold, transfer, soft keys, music on hold, intercom, BLF, speed dial, transfer, directory lookup, redial, do not disturb, and more – all work on digital phones just fine. Yes, they work fine on IP phones too – but at a cost.

1) The IP phones are 20-40% more expensive than comparable digital phones. (recently the manufacturers have started to raise the price on digital phones).

2) IP phones require Cat-5 cabling. Digital phones work on EXISTING cat-3 or cat-5 cabling.

3) IP phones require switched Ethernet ports.

4) IP phones need power – preferably POE infrastructure, but regardless, the additional power is not as efficient as the low-power, centralized power requirements of digital phones.

There are some common myths about IP phones we need to address head-on:

1) Soft keys only work on IP phones. Actually soft keys have been around for quite some time on digital phones. Some manufacturers extend the soft key functionality to the voice mail system (not particularly common in IP deployments).

2) Click-to-dial only works on IP phones. In most cases the click to dial solution works by creating a peer to peer relationship between the desktop and the IP enabled phone system, not the phone. Hybrid systems with digital phones are equally capable of supporting click-to-dial applications.

3) Unified Messaging Requires VoIP phones. Again, UM is accomplished by the phone system and/or voice mail system having an IP relationship with the email system, not the phone. In fact, in some IP solutions the voice mail is unable to light the voice mail indicator light on the phone (VoIP As You Are).

4) VoIP systems have better quality – HD Voice. HD voice typically accomplished by doubling the bandwidth from 3khz to 6khz range for the audio path is a nice contribution from the VoIP world. The 3khz limitation is largely from the PSTN, so the HD phones usually only work in HD mode on internal calls, but now some SIP carriers are beginning to support this as well. There is nothing about VoIP that enables this technology. It could just as easily be deployed on digital phones if demand warranted (I don’t think there is demand on the VoIP side, but it is nice if you have it).

5) IP Phones are easier and cheaper to move. This is true, digital phones need to be repunched and terminated. However, with increasing security concerns, and often the lack of availability of POE in all ports, moving IP phones usually isn’t seamless either.

This strategy is particularly obvious for the “classic” PBX makers that already include digital support in their systems. The Mitel ICP-3300 evolved from the maker’s TDM/digital SX-2000. The 3300’s software includes full support for all of Mitel’s digital phones. The code is there primarily for a migration strategy so the “mini-fridge” 2000 digital cabinet and cards can be controlled by the 3300. This migration strategy made sense, but what about new green field opportunities where digital phones make sense? Mitel’s only option is for the customer to purchase new 2000 hardware/cards thus reversing the potential cost savings. Mitel’s ATA (called an “ASU”) is actually a box that accepts their analog cards… why they don’t support using their digital cards and their digital phones greatly confuses me. The 3300 has no native support for digital phones.

It sounds trivial, but most VoIP installs replace an older digital system, and it isn’t uncommon for a warehouse location with limited wiring options to need to “upgrade” from digital to analog and lose their phonebook, feature buttons, intercom, and display features.

But most VoIP vendors support SIP. Today we can purchase analog VoIP gateways; often rack mountable with up to 48 analog ports. I like the sound of a digital phone gateway instead. A rack mounted box, SIP on one side, and 24-48 digital ports on the other. Just one problem, there isn’t a “generic” digital phone on the market. Hold that thought.

Let’s take a look at phone configuration. Proprietary digital and IP phones are typically configured from the PBX. SIP phones are usually configured outside of the PBX using vendor specific options. The world has not seen an open standard for configuration on digital phones, but such a standard would lead to mass produced inexpensive digital phones – a game changer.

What I want is a company to create a digital phone specification that lends itself to multi-brand phone systems that covers both signaling and configuration. Unfortunately, there aren’t very many companies that can pull this off. Digium may be the only one that could do it.

Digium doesn’t appear to be interested in making phones. They do embrace (live, breathe) open source. They have much to gain by lowering the total cost of a Digium solution, and if they design a decent amount of control/configuration in the specification, they could actually compete with proprietary systems in terms of managing/configuring the phones. They have the market share to pull this off, and I think they would actually gain support from competitors on filling this void.

As far as I can tell, companies like Cisco, Avaya, Mitel, Shoretel…. don’t like analog or digital. They have to support analog (fax, elevator, etc.) but many organizations just aren’t ready for total VoIP. I just watched a hospital buy an all Cisco all analog solution. The major proprietary vendors differentiate on features and architecture, but can’t do much with analog. A standard, inexpensive digital feature phone would be an improvement over analog without threatening their differentiation. The fact is the vast majority of IP phones aren’t used for much more than basic phone calls. Customers demand certain features – transfer, hold, conference, BLF, etc., and these can be hard in an analog world and the cost of an all VoIP solution can be high. A generic inexpensive digital standard can add a lot of value to the total VoIP proposition.