IP Convergence: Beyond VoIP, Beyond Cost Savings

I recently discovered an interesting report on the history of the VoIP industry, from 1995-2007. The report is titled, “The 10 That Established VoIP”

Below are stages of VoIP development from the report:
1)    Articulating the concept of telephony over the Internet
2)    Interworking of VoIP with PSTN
3)    Making cross network (IP/PSTN) telephony a commercial service reality
4)    Establishing VoIP and displacing TDM in international long distance
5)    National long distance (North America) scalability in VoIP systems
6)    Making national long distance a commercial service reality
7)    Facilitating adaptability and migration strategy for incumbent service providers
8)    Connecting VoIP islands IP-IP
9)    Establishing large scale retail VoIP services
10)    Extending VoIP across all networks (fixed and wireless)

Later in the report, and the main point of the report are a shortlist of candidates that made a important contributions to the development of VoIP (in the opinion of the author I should say).

1)    VocalTec – there’s little arguing this selection, as they were the documented first to allow users to have real-time, full duplex voice communications over the Internet.

2)    Level 3 – this selection on the list is a little bit more debatable, but it would be a shaky argument.  Level 3 can claim to have coined the term “softswitch”, with the initial focus being on segregating the signaling of voice sessions from the elements that handled the bearer (voice).    The big driver at this stage of the game
was modem traffic.  Modem access to the Internet saw a dramatic shift in the usage models of phone swithes, where folks would dial “into” the Internet – these long phone calls were much longer than voice communications, and phone company’s switches were getting tied up in rates much greater than was ever observed prior.  PRI offload is another term used to describe the method to address this phenomenon of the time.  The goal was to offload the data traffic from facilities that were meant to handle voice.

3)    ITXC – this company makes the list due to their aggressive attack on the International Long Distance market. 

4)    Sonus Networks – Sonus capitalized on the technology of VoIP by making it scale to volumes of calls (minutes, etc) that had been a pain point.  Due to Sonus’ mastering of the scale problem, adoption of the high-density media gateway started in earnest.  

5)    3GPP – the 3GPP is a consortium of companies that created the blueprints for IP Multimedia Subsystem (IMS).  The most important thing they did was establish SIP as the core protocol for the network Architecture.  This endorsement ensured that there was no longer a valid H.323 versus SIP argument.

6)    Cisco – the major contribution Cisco made to VoIP adoption was in the enterprise.  Enterprises had (and still do) a compelling business plan to spend capital on VoIP due to the associated cost savings in toll arbitrage.

7)    Skype – this peer-to-peer application has been around for quite a while (since 2002), and today boasts over 300 million accounts!  I have been a frequent user of Skype, and use it often while traveling for voice communications.  The application is well used and loved among the communities that place frequent International calls.

8)    Softbank BB – this is a company that has a very similar business model to Vonage.  They were the first to market a voice over broadband (VoBB) application, and today have over 15 million subscribers in Japan.

9)    Nextone – this is an odd selection to me.  I think that it should be Acme Packets instead.  Be that as it may, both are Session Border Controller vendors.  Their gear help bridge VoIP islands in the carrier space, and facilitate “SIP Trunking” in the enterprise worlds.

10)    SpectraLink (now part of Polycom)– this selection is motivated by Voice over WiFi.  Many pendants in the industry believe that without a wireless component, VoIP cannot dominate as a technology.

Above is what the report states, and I have added my comments.  If you have any of your own, please share.

Adam “voiploser” Uzelac

DISCLAIMER: The comments here are mine only. They don’t necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

Back in November of 2006 I posted on this blog some statistics about the our (Global Crossing’s) VoIP interconnections.  We conduct interoperability testing for all the VoIP Interconnections that we establish today.  This is similar to a "turn-up" procedure to ensure that our "voip stuff" talks nicely with our customer's "voip stuff".  I like to think of this process as a value-add.  We have a large and growing “SIP Trunking” suite of products.  As a guy that is interested in technology in general, and also interested in trends I, once again, mined data points from interconnection testing. 

Below is the information:

11/2006 (Sample size = ~500)

Percentage Open Source         24.80%
Percentage SBCs                    32%

6/2008 (Sample size = ~900)

Percentage Open Source         26.45%
Percentage SBCs                    41.28%

Both categories have grown, with the adoption of SBC technology growing at a stronger clip.  Some interesting points can be made from this very informal analysis.  Firstly, more folks are using Session Border Controller technology, and this increase was observed both within the Enterprise as well as in the Service Provider space.  Good, bad or indifferent, it’s happening.  The second point that I would like to make is that Open Source VoIP (Asterisk, SIP Express Router, etc) is also growing, but more not as fast as I would have liked. 

Adam "voiploser" Uzelac

DISCLAIMER: The comments here are mine only. They don't necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

There was an article in Processor Magazine that starts to address some points that should be taken under consideration for IT managers looking at packetized communications for their Enterprise.  Below are the highlights and my thoughts:

Look For SIP Support
I couldn’t agree more!  SIP is a protocol used to establish, teardown, modify, etc communication sessions.  It’s very diverse and relatively simple when compared to past mechanisms.  Most importantly, it has become the defacto standard within the world of telephony.  There’s native SIP support in nearly all the major vendors that supply VoIP gear. (Cisco, Avaya, Siemens, Microsoft)

Consider The Benefits Of Hosted PBX
This topic has be discussed numerous times in the past, and even before that within a TDM context (PBX vs. Centrex).  The thing that’s different within an IP context is the feature and functionality available.  When comparing a PBX to a Centrex offering, one key difference was additional feature and functionality in a PBX.  Centrex offerings didn’t have the same "whiz-bang" features.  In today’s Hosted Telephony offerings, there’s near feature parity, so the key determining factor becomes cost of ownership.

Think Unified Communications
VoIP (or Telephony) MUST be seen as a stepping stone to the ultimate goal of Unified Communications.  IT managers should consider the roadmap to UC when choosing a Telephony solution.  Real-time communications need to become multi-modal, meaning there should be options to transition communications from IM to voice to video to online collaboration on a document, and then back again - all within the same context and within a common look/feel.

Traversing NAT
Though the issue of Network Address Translation (NAT) is well known to negatively impact SIP sessions, the real point for consideration here for the IT Manager should be around considering the deployment of a Session Border Controller (SBC) within their Enterprise as part of an overall design.  

Know & Apply Codecs
There are more ways to packetize voice and video communications than one can shake a stick at.  The author points out the predominant technologies of G.711 and G.729.  Issues of bandwidth consumption and quality of user’s experience must be balanced.  Generally speaking, the more bandwidth consumed, the better the experience.  But the more bandwidth used, the greater the cost to upgrade the LAN/WAN infrastructure to accommodate.  If you skimp on cost, the result would be poor quality, and then adoption and experiences would suffer.  It’s a delicate balancing game.

Some further comments captured in the article..

• Make sure to have 100k in bandwidth free and available for every conversation when determining whether the enterprise really has enough bandwidth for VoIP, according to Andy Abramson, blogger at VoIP Watch and founder and CEO of Comunicano (www.comunicano.com). With this understanding, it’s easy to see that DSL or a cable modem line won’t cut it.

• Get VoIP phones that are both wired for Ethernet and wireless for Wi-Fi connectivity, notes Abramson. “That way, people can wander, and all internal calls within the building are free of charge because they stay on the network.”

• “Make sure the vendor is going to be around to support the purchase,” Abramson says. An older vendor with roots, commitments, and financial means is an obvious choice. A new vendor with strong management, skills, and reputation who proves out through considered research can also be a sharp choice.


Adam "voiploser" Uzelac

DISCLAIMER: The comments here are mine only. They don't necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

I think everyone has at least one horror story about navigating through an Interactive Voice Recognition (IVR) menu over the phone.  You know what I am referring to...those “Press one for this, press two for that”.   One example of the miserable part about these systems is when you have to listen through the entire thing only to find out that the option you need isn’t an option at all –OR- when someone (usually my kids) talks to you during the recording and you miss some options (most likely the one that you needed!), so you have to listen to the entire announcement again.  This, my friends, is IVR hell.  Well I just stumbled upon a company that is addressing IVR hell head on, and with a whole bunch of ingenuity mind you…

Fonolo has created a novel concept.  They have transcribed the menu prompts of the major destinations – meaning they created a visual tree of the IVR menu structure, and you can navigate it online via clicking through the options.  Then you input all the information necessary.  Once all the inputs are registered by Fonolo from you, then Fonolo will place the call and automatically get you to the point that you need to be in the IVR menu.  This point may end up being on hold for a human, but just think of all the time you have saved just getting to that point.


Adam "voiploser" Uzelac

DISCLAIMER: The comments here are mine only. They don't necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

Click here for the SIP 4.0 IETF Draft!

Be sure to note date of posting...

Adam "voiploser" Uzelac

DISCLAIMER: The comments here are mine only. They don't necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

Of the more interesting developments that are emerging as SIP  technologies and products mature is the changing context that Enterprises’ IT organizations operate in.  Historically, telecommunications has been defined and provided to the marketplace by Service Providers like Global Crossing, Level 3, BT, etc.  The requirements needed to formulate an offering were provided by SPs to the manufacturers of the platforms (Cisco, Juniper, etc).  Things are changing rapidly on this front though.   Enterprises in some cases are bigger in many aspects of the definition than some smaller traditionally defined Service Providers.  In my humble opinion, this warrants of new definition of the Enterprise.  The Enterprise is the newest category of a Service Provider. 

With regards to SIP, there is ample evidence of its adoption in the Enterprise marketplace and vendors that have catered directly to Enterprises have taken heed.  As examples, see Microsoft’s, Siemens’, Cisco’s, Avaya’s adoption of SIP in their product portfolios.

The impact of this is an overall decentralization of the Enterprise communications environment that leads to a more collaborative workforce.  If you need some examples, think of the IM communications in support of standard telephony and email infrastructures.  Nowadays in our MS OCS/Exchange environment at Global Crossing, when I read an email – I not only have the ability to reply, but IM or click to call in response. This is all handled via a single user interface.  This form of collaboration is _NOT_ dependant on the traditional services that Global Crossing provides. (with the exception of the underlying IP transport), but rather the infrastructure is supported by our internal IT department.  By the way, this is all decentralization is enabled with SIP.

Adam “voiploser” Uzelac

DISCLAIMER: The comments here are mine only. They don’t necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

The impact of the Open Source movement on VoIP is strong and continues....

Some of the tools that we use and/or tested with here at GC are included, for example SER, Asterisk and one that's not on the list SIPP.

Top 50 Open Source VoIP Apps

Adam "voiploser" Uzelac

DISCLAIMER: The comments here are mine only. They don't necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

This one is a beauty - on Jan. 10th of this month, the ACLU issued a statement that reported that a FBI wiretap was "unplugged" due to a lack of payment.  The ACLU is quick to point out that this action was taken from the same telecoms that permitted the tap without the proper approvals...

From Michael German, ACLU National Security Policy Counsel: "It seems the telecoms, who are claiming they were just being "good patriots" when they allowed the government to spy on us without warrants, are more than willing to pull the plug on national security investigations when the government falls behind on its bills."

Adam "voiploser" Uzelac
DISCLAIMER: The comments here are mine only. They don't necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

Femtocells the Answer?

There have been some very interesting developments in the wireless world as of late.  Femtocells are basically Access Point Base Stations that permit wireless operators to extend coverage in places where “dead spots” are problematic.  This sounds like a great idea for those that have experienced the issues where a wireless phone call drops due to one moving into an area where coverage is “shady” at best – for instance an elevator or a remote location in a campus or building.  But with this concept, comes some problems that need to be resolved, like E911, Lawful Intercept and other such governmental obligations.  

First note the informal poll below from www.cellcoverege.com - this is a problem the industry wants to address.

Femtocells have been designed to use licensed and unlicensed wireless spectrum.  In the licensed scenarios, there are concerns around Interference with the already established towers that provide subscriber access.  There are limits to the number of adjacencies that mean special attention needs to be spent with regards to the placement of the femtocells. This concern stems from marketing such solutions direct to the consumer base, and this means a “willy nilly” approach to spectrum access for subscribers causes confusion with regards to E911 and Lawful Intercept requirements that mobile network operators much meet.

Though femtocells are gaining momentum as an alternative, there are concerns that first need to be addressed before an expectation of widespread use can be realized.

DISCLAIMER: The comments here are mine only. They don’t necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.

Google’s Impact on the Wireless Internet

In case you missed the announcement last week, Google has tossed it’s hat into the ring for the last remaining commercially viable wireless spectrum in the US.  The 700MHz spectrum was used to provide analog TV service and becomes available to the highest bidder on January 24th, 2008.  The impact of Google’s announcement is yet unknown, but should prove to be very important in the long run.  

Some interesting datapoints:

•    From Wikipedia - Several countries, including the UK, now have more mobile phones than people. There are over five hundred million active mobile phone accounts in China, as of 2007. Luxembourg has the highest mobile phone penetration rate in the world, at 164% in December 2001. In Hong Kong the penetration rate reached 139.8% of the population in July 2007. The total number of mobile phone subscribers in the world was estimated at 2.14 billion in 2005. The subscriber count reached 2.7 billion by end of 2006 according to Informa[citation needed], and 3.3 billion by November, 2007, thus reaching an equivalent of over half the planet's population. Around 80% of the world's population enjoys mobile phone coverage as of 2006. This figure is expected to increase to 90% by the year 2010.

http://www.internetworldstats.com/stats.htm

Now allow me to pluck the germane data from the above for my point.  Phone = 3.3 BILLION -  Computers =  1.25 Billion.  Google already enjoys dominance for eyes in the computer space, and if they are as successful on wireless phones as they are with PCs, watch out world!

The problem Google currently has with wireless devices, especially in the US, is the locked-down nature of the device to the wireless network operator.  It’s actually a sad commentary that Google has to forge ahead with buying spectrum, but it makes all the sense in the world.   Boy is it interesting watching the impacts to the industry.  Check out Verizon Wireless’ announcement about opening up their network to devices.  Don’t think for a second that these two announcements are unrelated.  They couldn’t be more tied to the hip.

All this guy has to say is, “Good on ya, Google!”


Adam “voiploser” Uzelac
DISCLAIMER: The comments here are mine only. They don’t necessarily reflect intelligence, refined thoughts, or anything that the reader should take too seriously. Should the reader expect a polished thought process in the content addressed here, then a strong dose of medication should be prescribed to address that misconception.