In the small pond of existing voice-over-IP (VOIP) equipment installations, Cisco Systems Inc. (Nasdaq: CSCO) is unquestionably one of the big fish. But even Cisco executives admit that it's going to take some time for the market to sustain a catch worth talking about.
At a press briefing held Tuesday at Cisco's San Jose campus, Cisco executives said that expanding the arena for VOIP equipment is going to be a slow, tough process, one that the biggest potential customers -- service providers -- may have to be dragged into, sometimes against their will.
Attractive to end-user customers as well as investors like Intel Corp. (Nasdaq: INTC) (see Intel Snatches VOIP Startup for $550M) for its potential cost-cutting and open-architecture attributes, VOIP is an ugly acronym to some service providers for the very same reasons, especially given the current whipsaw market conditions facing most sellers of telecom services.
Though some service providers (see C&W Bets $1.4 Billion on VOIP) and large-enterprise customers may be attracted to VOIP platforms as a way to increase functionality and reduce prices, even Cisco knows that large service providers (especially those with lucrative TDM businesses) are loath to spend money on new equipment that will ultimately drive down the tariffs they can charge their customers.
"Proceed with caution" is how most RBOCs and PTTs approach VOIP, says Mike Volpi, Cisco's chief strategy officer.
"Like any market, the incumbents will not be the ones that move first," says Volpi. According to Cisco, most of its early VOIP wins are coming in places like China and Latin America -- not North America, where the RBOCs reign.
"Our strategy is to [sell] to the new players first, then drag the bigger guys along the way," Volpi says.
Drag is the operative word. According to Volpi, the whole idea behind VOIP support is to increase sales of traditional Cisco routing and switching platforms, many of which are being VOIP-enabled.
"It's about the infrastructure," says Volpi, who estimates that VOIP-related equipment sales currently represent at best a single-digit percentage of Cisco's revenues.
"We don't expect it to be huge right now," says Volpi. "This is a long-term play for us."
According to Alistair Woodman, director of marketing for Cisco's service provider line of business, the company has already sunk $1 billion into VOIP-related development and acquisitions (see Cisco Turns Up Voice Signal) over the past five years, and currently has approximately 1,000 engineers dedicated to VOIP development. In order to prime the market, Cisco even manufactures IP telephones, of which it claims to have sold 280,000.
Cisco, which claims that 80 percent of its product line has some form of VOIP support, also says it has sold equipment that accounts for 1 million VOIP ports, a number it embraces as a sign of market leadership. Still, that's somewhat like bragging about leading a marathon after the first 100 meters -- a reservation from which Cisco doesn't hide.
And even though Cisco enthusiastically demonstrates cool VOIP applications -- like Dialpad.com's free Internet phone service -- Woodman knows that many service providers are going to cringe, not cheer, when they hear about the competitive possibilities that Cisco says VOIP technology could allow, such as Web-based dialtone portals, where customers could pick and choose among voice service providers.
Woodman, who brought up the service-portal idea during his presentation, jokes: "I don't show that slide to service providers, because I don't want to make them uncomfortable."
But to make the VOIP market succeed, Woodman says Cisco and others will have to develop new applications to convince service providers to embrace VOIP (and they'll have to be more entertaining than the in-house app that delivers Cisco stock quotes, which, when demonstrated Tuesday, returned a slightly embarrassing answer). Otherwise, VOIP will remain but a small stream in the overall river of telephony opportunity.
I cannot believe that this stupid thread is still going strong. Give it up! Umm - just in case nobody told you, the company closed 3 years ago. I thought people were supposed to mature with age - it seems that stage in life has skipped you half wits by.
Although - while I rallied for Tachion and found all of this stupidituy offensive 3 years ago - now it's hysterical that you just haven't moved on.
Don't lump any of these companies into the same sphere as Tachion. The level and depth of the complete deception practiced at Tachion simply cannot be re-created anywhere else. Tachion was never anything more than a series of powerpoints, a scheme hatched by one individual to create a pump 'n' dump play, and an exercise in PURE egotism.
The entire story may one day be told. It could be a drama, except that it's just too damn funny.
lread: "Rapid5 will be last to the CLEC media gateway market, but could very well be first to the ILEC media gateway market, which is a huge market."
I hear other vendors are also trying to get into ILEC COs. Telica may be on the same track. Whoever gets there first will have a nice feather in their cap. Why do you think Rapid5 will get their first?
mu-law: "Don't get me wrong, I'm no cisco loyalist either, but at $10k list per OC3 port or so (JNPR is similar) this is a far cry from what you pay to switch 84 toll t1's."
Agreed, the switching costs of packet vs. TDM networks offer significant economies. And I know of a PTT in Europe that has ATM switches @ $1500 per OC-3, which could reduce the cost by nearly another order of magnitude.
"P.S. Its about time this board got back to business. Thanks for your refreshing contribution."
I agree, there is too much whining and vendors trashing each other on these boards. but there is also much valuable information being exchanged. Thanks for your technical insights.
Certainly, VoIP will take-off inside of Enterprises first. However, I thought this discussion was more related to VoIP inside a given carrier's network. VoIP doesn't force the idea of VoInternet. To make it work across a heterogeneous, multi-carrier environment, you would:
a) Need to establish QOS relationships with other carriers with VoIP backbones b) Need to handoff to traditional TDM/circuit-switched systems via trunking gateways.
The VoIP packets do not need to be part of the Internet, just part of a packet network.
The tests I was referring to were part of a production VoIP network (it was not part of the Internet).
Although TDM networks themselves don't experience congestion, the Voice system running over the TDM network most certainly does:
(tri-tone) I'm sorry, all circuits are busy...
It is a capacity planning excersise to ensure that you don't get too many of these messages. It is also a capacity planning excersise to ensure that you do not congest a VoIP network and get degraded sound quality. In either case, if you do not deploy enough capacity, your service sucks.
You make some good points, but you gloss over the fact that VoIP isn't at this point comparable in quality to TDM over heterogeneous networks due to the fact that there is no QoS standardized for IP traffic across vendors. I suspect that the first place that VoIP will take off will be in the enterprise where a single vendor will provide the VoIP capabilities for the internal network and a gateway to the TDM (public) network.
>The idea that VoIP sounds worse than POTS is simply wrong.
In the lab or over the internet? In the lab you can control things such that you get fine performance, but in the "cloud" you don't have control over whose boxes you traverse, etc. In the presence of congestion, there is currently no agreed upon mechanism to provide the continuity of service (over IP) that voice traffic requires. Note that there is no such thing as congestion in the TDM network. If there is no capacity, the call doesn't go through. You don't get degraded service. When every IP box on the public network agrees on and implements the sort of QoS that voice traffic requires, then we will see the replacement of the TDM network. Not before.
My point was that "quality" isn't all-important. People will use services if they find some aspect of it useful (even if "quality" diminishes). Obviously, VoIP networks don't sound anything like the wireless phones. VoIP's sound quality is indistinguishable from a traditional phone and with different encoders sounds far better than traditional phones. Bandwidth for voice is essentially a non-issue unless you are a carrier that is somehow bandwidth poor. Voice bandwidth is becoming an ever decreasing part of the driver for network buildout. The volume is being driven by data. So, who cares if VoIP takes a little more bandwidth to sound better? In the near term, this voice bandwidth will be in the noise (pun intended).
Bottom line: the QOS offered by VoIP is MORE THAN SUFFICIENT for the vast majority (99+%) of the population. Think about the sound quality on a scale of 1 to 10, where:
1 - Unintelligible 10 - sounds like a high-end home theater system
Wireless => 3 POTS (compressed) => 4.5 POTS (uncompressed) => 5 VoIP (standard) => 5 VoIP (premium encoders) => 7
The idea that VoIP sounds worse than POTS is simply wrong.
Huh? Sonus is in the CLEC market? Huh? XO is a clec? Global Crossing is a CLEC? What is your definition. Qwest is a CLEC? Are you nuts? Doesn't the letter L in clec mean LOCAL? So Global Crossing is now in the local class 5 market? ok there friend....go back to Marketing class at Rapid5 and make some more marketing slicks.... Is Rapid5 a spin off of Tachion?
I'm not sure you're entirely correct. While 64Kbps voice is the standard, particularly for interconnect, it is v.common for carriers to "compress" it down to 16K without any loss of quality (although they are all cagey about admitting it publicly, in my experience).
The thing is they do this within their own transport domain, where they have sufficient control to assure themselves they won't pee their customers off by so doing. And these conversation do not contend for the same bandwidth, they are still separated by tdm mechanisms.
Many 8K compression systems exist and some carriers even use them.
As has been suggested, it's not so much that more bw is required to maintain a specific quality metric, its that the controls have to be applied end-to-end in order to ensure a service that people will pay a premium (over and above best efforts) for.
fiber_r_us: So voice has to have strict TDM-like QOS before it is useful? Seems the Cellular industry would disagree!
fk: What people will put up with WRT service from their wireless phone and what they will accept from their wired phone are two different things. ---------------------
While the radio interface uses low bit rate vocoders (~13kbps) per voice channel in common mobile nets, these channels are not shared, ie 1 conversation uses the whole channel. The core of the vast majority of mobile nets are based on circuit switches using tdm transport.
Ever stop to wonder why?
They are just now beginning to think about moving to packet cores. Guess what their major concern is?
And one of the reason why mobile has pushed voice quality down is that they NEED to fit more conversations in the same spectrum just to stay in business.
As rk implies, it is mobility which is the major characteristic valued by the customer. e.g. you don't see many call centres using mobiles, since the quality is insufficient for most enterprises to be prepared to foist it on their customers.
The blogs and comments are the opinions only of the writers and do not reflect the views of Light Reading. They are no substitute for your own research and should not be relied upon for trading or any other purpose.
To save this item to your list of favorite Light Reading content so you can find it later in your Profile page, click the "Save It" button next to the item.
If you found this interesting or useful, please use the links to the services below to share it with other readers. You will need a free account with each service to share an item via that service.
Digg
Del.icio.us
Reddit
