There has been a lot of noise on VOIP. Personal VOIP systems (Skype, Vonage, etc.) are taking off fast. Large organizations can clearly cost-justify VOIP because of infrastructure savings and multi-site support.
We needed a new phone system recently, and so it seemed a good chance to look at this technology, and play with some interesting products. The first shock was how few supplier choices there were for a small company solution, and that many were clearly not ready for prime time. An open source approach (Asterix) was plausible, but we judged it would take too much resource and not meet all our needs. Avaya looked good, but was linked to Exchange, which is unacceptable for us (they don't seem to want to sell the non-Exchange version to small companies). We ended up with Splicecom, which offers a system with both VOIP and conventional lines.
As a phone system it works fine. There is a BIG advantage in having a phone system running on Linux, with Web administration. Configuration changes are easy to do in house, backup is straightforward, software phone access is great, and getting voicemails delivered by email is very useful.
However, we don't use VOIP. VOIP phones as part of a company solution work out much more expensive than analog phones, and don't give much advantage on single site. We wanted to support a remote worker, but it turns out that using VOIP through a firewall is fraught with problems (in general, not just for this product). Similarly, it doesn't seem practical to integrate free Internet VOIP with the corporate VOIP system. This would have been a nice customer communication option.
Summary: VOIP for small companies has a lot of potential, but it isn't readily achievable yet.
... Steve Kille