By silicon.com, 28 June 2004 16:45
The market for voice over IP - the routing of phone calls over data networks - is finally taking hold after many years of talk thanks in part to the emergence of a standard known as Session Initiation Protocol, or SIP. Or so say many industry experts.
SIP enables communications between multiple users involving voice, video, instant messaging and other multimedia elements. At least that's the best we can explain it in plain English. Everyone we talk to tells us it's quite complex.
The good thing about SIP, say the experts, is that it allows for interoperability. It provides a way for companies to know whether the software they buy from Vendor A will work with what they already have from Vendor B, and so on, and thus makes them more inclined to switch to VoIP than they were in the past. Hence the recent uptick in the use of the technology and the bullish predictions for the future.
The bad thing about SIP is that it is such a complex standard. Because it's complex, it's very expensive to implement and support into a VoIP product. This naturally creates a high barrier to entry for smaller, cash-strapped players or even large players who don't want to invest huge sums to get in on VoIP.
Going with SIP - instead of some other standard - means, in the end, there will be less competition in the VoIP arena because only a few can afford to compete.
Big companies won't mind - we're not saying they're rubbing their hands and cackling but they should be pleased with fewer upstarts getting a piece of the pie. No, the casualties are the little guys in the market that can't stick around and even more importantly the potential upstarts with great ideas that would have become involved in VoIP but now can't, at least not on a commercial level.
Less competition is bad for companies using the technology because it means less innovation and likely higher prices. It could also mean poor service because an IT director has few competitors to go to should his vendor disappoint him.
This isn't the first time smaller players have been elbowed out of a business sector once the resource-rich big guys came to play, surely. And it's not likely to be considered 'anticompetitive' in the legal sense.
But that still doesn't mean it's a good thing.
Comments
There are 6 comments. Join the discussion
1. JC Francois
What sort of kindergarten analysis is this!
SIP is complex = SIP is expensive therefore the little guys with no money will get crushed by the big guys with the bucks. Duh.
Facts? Examples?
Have you been on sabbatical leave on the desert island for a few years or what?
If you think SIP is complex, have a look at the numerous flavours of H.323. But that's not the point. More importantly, there has never been so much competition and new companies in the telecommunications industry as today. Some will get big and rich, some will be taken over, some will disappear but that's the same story in every industry. One thing is certain: it's benefits all the way for customers.
2. Goten Xiao
I've yet to see benefits from Microsoft. They elbow people out all the time, simply because "Everyone uses Microsoft, we have to use Microsoft to be compatible."
Interoperability isn't the key.
A common protocol that everyone uses and implements the same way is.
Example: HTML. It's a standard, but there are many different implementations of it. Microsoft have one, Netscape another, etc etc.
TCP/IP is more like it; everyone uses it, everyone uses the same protocol, everyone implements it in the same way (circuitry doesn't count as long as the end result is the same).
Now, VoIP is a standard, correct? So all we need is a common implementation from Microsoft, Sun, HP, Cisco, whoever.
It's like saying "We can't use bolt X because nut Y has one thread at the end that is 2mm out." If everyone made bolts and nuts to a standard, and used the same implementation, there wouldn't be a problem because everything would work together without any modification.
3. Alun Lewis
Strongly inclined to agree with the last poster - especially as someone who's been following SIP for the last few years and was even involved in launching a magazine on the subject a couple of years ago....
Telecoms has been a closed camp for small companies until recently - anyone remember Intelligent Networks ?
SIP - and related technologies like OSA/Parlay - are truly disruptive.
SIP/VoIP issues are more to do with security, power supplies and the general challenge to the telecoms status quo than company size.
Post-modernist telecoms rules - eventually !
4. Michael Crown
I find this analysis baffling. If SIP is more expensive to develop and implement, how does the writer explain the recent proliferation of sub-$100 SIP devices and why they don't say Cisco on them? If the big players are the net beneficiaries, why have they been so slow to embrace SIP? Why are the vast majority of the new consumer VoIP services using SIP? Here's the real "Dirty Little Secret": The big established telecom players sell proprietary systems, and have little incentive to do more than pay lip service to SIP. I have heard every excuse there is; "It's too complicated", "It's not really a standard", "It doesn't support the features people want". You name it, I've heard it. Meanwhile, SIP, Asterisk and other open standards have allowed a number of small upstart companies to focus on creating best-of-breed components, and these components are being used by a legion of clever integrators to engineer world class systems that do more for less. I've seen this dynamic many times in the past and on every occasion it marked a changing of the guard. Big fish will always eat little fish, but SIP has created an incentive to enter the market, not a barrier. Microsoft might eventually use it's muscle to try and squeeze others out, but two things lead me to believe it won't happen; First, MS Messenger is a welcome but largely irrelevent player in the emerging telecom landscape. Their influence over the standard is overestimated. And second, their place at the table is by no means secure. If they overplay their hand, they may find they are holding Aces and Eights.
5. Christian Stredicke
The author should make a better proposal - and take it through the standardization process. SIP has surely its little flaws here and there, but its by far the best standard for real time communication today and problably for a long time.
Is anyone out there who wants to reinvent http (because of the little flaws)?
6. Simon Lewis
Are you kidding ? I find it hard to reconcile the author's comments with reality. SIP is a prevalent, established framework which requires similar levels of investment to adopt as other protocol standards. I find it baffling to read such an analysis and merely makes me wonder about the motivations behind it.
FOr the launch of our Babble product, BON.net adopted a standards based approach. Admittedly, the development of a presence engine requires extensions to SIP to provide higher level services but we are developing an API in order to promote interconnection with other providers. This most certainly does not mean that customers require any additional investment or development to take advantage of our service's benefits, when compared with any other. Surely networks such as Skype, where a proprietary approach not only limits, but precludes, competition and interconnection, are more properly the target for such comments. Try connecting your Cisco gateway to a peer-to-peer service !