SIP vs. Minet for Managed VoIP Telephony Applications
The major differences between the SIP protocol and the Minet protocol can be traced back to the radically different philosophies adopted when the two protocols were first developed.
SIP (Session Initiation Protocol)
SIP (Session Initiation Protocol) was developed at the IETF (Internet Engineering Task Force) for computing platforms to make media (voice) connections directly with other computing platforms. With SIP, the endpoints act as peers. SIP is a “peer-to-peer” protocol because developers believed endpoints could be made increasingly smart and so there was never intended to be a third party mediating or supporting their connections. Consequently, SIP is in no way optimized for hosted service. It can be argued that SIP was actually intended to make service providers unnecessary. Of course service providers do a lot more for customers than just connect calls. Businesses need service providers to keep records, manage resources, simplify services, manage change, add value, and just plain keep things working. Because of the philosophy behind its design, SIP does not make those things easy for service providers. As evidence of the inefficiency of the SIP protocol in hosted Voice over IP (VoIP) environments, note that all major VoIP PBXs work better when their customers choose that vendor’s propriety VoIP protocol over SIP.
In contrast to SIP, Minet™ was developed by Mitel™ so that an intelligent server is always in charge of managing telephone calls. The Minet philosophy is that there is a controlling server, and all of the intelligence naturally resides there. That’s where functionality is most conveniently maintained and updated. The user simply presses a key on their phone and the server is told which key was pressed, and then the server uses its latest software and configuration to decide the appropriate actions. Minet is thus a “stimulus-response” protocol. Stimulus protocols minimize phone software and don’t attempt to set the phone up as a peer to the server.
Who’s in Control?
When you buy a hosted VoIP solution for your business, you expect the service provider to manage all of the technical details. You only want to add and modify users. You only need control over the end-user features: forwarding, speed-dials, parking, intercom, group features, etc. But even for those end-user functions you need the service provider to be in a position to see everything and assist immediately when there are difficulties. For all of those reasons business VoIP providers, SIP-based or not, try to control their services from hosted servers. These VoIP servers “in the cloud” give the service providers both control and visibility. But SIP-based hosted service providers are always at a disadvantage, because phones using the SIP peer-to-peer protocol inherently hide things from the service provider.
- They collect dialed digits and decide on their own when to invoke the server.
- More generally, buttons pressed on a SIP phone are not seen directly by the associated server or provider. End users can make procedural mistakes in handling a call that the service provider cannot see.
- They hold critical configuration information, commonly provided when the phone is shipped or updated, and the configuration on those phones can change or be out-of-date without the service provider knowing.
Providers of SIP-based services have gone to great lengths to mitigate all of these issues, but there is still no way of their knowing for sure exactly what is happening on the phone or what might have been done that altered the phone. This creates support problems for both the end customer and the service provider.