x
<<   <   Page 4 / 4
dljvjbsl 12/5/2012 | 3:15:10 AM
re: Does VOIP Business Add Up?
SIP really doesn't do telephony all that well. Feature transparency is still a work in progress. Common features like 1A2 key system emulation are incredibly complex to implement. The whole presence concept is half-baked. SIP is just an evolving set of protocols


Now that is talking!

SIP does not do telpehony very well because traditional telephony features were designed for the late 70's and early 80's technological environment when they were developed. In particular, they were not designed for an environment of ubiquitous personal and device mobility. SIP was designed for just such a communications environment and more than that was designed to create applications within that environment. The traditional telpehony feature set has been made obsolete along with the ystems that provide them.

I wonder what use a 1A2 key system will be in a world in which SOHO users will be connected to DSL lines. I do not have the same puzzlement when I think about SIP services and SOHO users.

Feature transparency is an artifact of the technological past. It is a means to try to synchronize the workings of isolated call control feature logics. It has obvious limitations in the types of features that can be supported. One gets the standard lowest common denominator features and is unable to create any feature that would be of specific use either personally or in respect to an enterprise need.


The sooner that feature transparency is dropped as a goal the sooner this inustry will be able to create applications that will be actually ueful and actually attract customer interest. No more check box feautrez that nobody uses in other words.

The SIP purist on seeing the 3GPP architecutre would wonder why it took them so long to create so very little. The 3GPP architecture is just the SIP architecture dressed up in endless acronyms. SIP has no problem with application servers. Indeed it was designed to cater to the needs of application servers. It can route and reroute audio and data almost at will to coordinate the operation of multiple application servers

As an aside, ISDN was not about integrating the backbone, it was intended to create a single network for all services that would be accessed via a single socket type. The ISDN S bus was a long way both physically and conceptually from SONET rings.

ISDN failed for two main reasons. Firstly it failed because the telephone compnaies realized that the best palce for ISDN (like all digital network systems) was at the periphery. This was a threat to their monopoly and so they made sure that ISDN could not provide services at the periphery. In doing so, they made sure that ISDN could offer no useful services and so ensured that ISDN could find no customer base. Secondly, ISDN was developed at the time of the Internet expansion and just as it became available the web was developed. ISDN could not hope to compete agaisnt these two technologies.
alchemy 12/5/2012 | 3:15:05 AM
re: Does VOIP Business Add Up? dljvjbsl writes:
b) special feature handling is required for this type of call. For example, a Do Not Disturb feature can not be honored in the face of an emergency call like this. This can cause dfficulty in a traditional system becuase it tends to require that all affected users must be served by the same switch. In traditional switches, the internal feature logic acts differently than external services. For a SIP implementation, the internal/external difference is non-existant and so feature operation is simplified fo this service

Say what? Most SIP phones implement DND as a station feature. If you fire it an INVITE, it's a black hole. It won't know squat about emergency calls. This is the whole problem with fully distributed call processing. If the feature is done by the phone, you have to rely on potentially thousands of different implementations to all get it right. That's why the cellular guys in IMS/3GPP put all the features inside the network subtended from the S-CSCF.
dljvjbsl 12/5/2012 | 3:15:04 AM
re: Does VOIP Business Add Up?
Say what? Most SIP phones implement DND as a station feature. If you fire it an INVITE, it's a black hole. It won't know squat about emergency calls. This is the whole problem with fully distributed call processing. If the feature is done by the phone, you have to rely on potentially thousands of different implementations to all get it right. That's why the cellular guys in IMS/3GPP put all the features inside the network subtended from the S-CSCF.


The standard SIP architecture user preferences being enforced at the proxy. This is where CPL is active. There are many reasons for this but a primary one is that the SIP phone may very well be powered off. So the S-CSCF or XW-q/681Z704) or whatever acronym that 3GPP has dreamed up is identifying a standard SIP functional block. SIP-servlets and SIP-CGI are both active in this block so it is well known.

Doing it at this functional block will also ensure that the priorites between personal, enterpise, emergency (911 etc.) can be idenfied and maintained.

alchemy 12/5/2012 | 3:15:03 AM
re: Does VOIP Business Add Up? There is no "standard" SIP architecture. Virtually every feature has both an endpoint-centric method and a core-centric method.
<<   <   Page 4 / 4
HOME
Sign In
SEARCH
CLOSE
MORE
CLOSE