Voice mashups - Notes on Alec's conference call today
I'll be speaking at Ingate's SIP Trunking Seminars at IT Expo in Miami next week

Skype says "No" to VoIP interoperability - *because customers aren't asking for it!* - Well, I am!

skype_logo.pngSo Skype says that they have no plans for interoperability with other VoIP systems because their customers aren't asking for it??

By way of Dameon Welch-Abernathy today I learned of Phil Wolff's post back in December about ZDNet's interview (Got all that? ;-) with Skype's VP of telecoms, Stefan Oberg. The article was primarily about Skype's London phone number debacle, but this was the part that most irritated me:

Another issue which may concern business users of VoIP is the Enum registry, which aims to unite not only the various VoIP providers — referred to by some as "islands" due to their lack of interconnection with each other — but the entire VoIP and traditional telephony worlds.

Asked whether Skype had considered opening up its famously closed communications protocols, Oberg claimed that there had been no customer demand for interconnection. "[Customers] are not saying they would love to call a VoIP provider on a different network," he said. "Customers are asking for better video and better conference calling. If it is something that customers really ask for, we would consider it, but it is very easy for anyone to get on the island."

Well, Mr. Oberg, here is one paying customer of Skype who can state unequivocally:

"I would love to call a VoIP provider on a different network!"

Here's the thing, Mr. Oberg. There are a whole lot of us out there who are looking to build the next voice communication network. We're looking forward to the day when today's PSTN is just some story that greybeards get together and reminisce about. ("Remember when we used to have to dial numbers? And wait for the connections? And remember how much we had to pay our phone companies for the privilege? And remember those 'busy signals'?") We're looking to make it simpler and easier and so that ultimately voice just smoothly fits in to our communication as one of the several different ways we communicate. (others being text/IM, video, etc.)

The funny thing is that many, if not most, of us experimenting with what will be next are Skype users. Probably in many cases paying Skype users since we have Skype Credit and SkypeIn numbers. Because, like you said, Skype makes it "very easy for anyone to get on the island." You do a lot of things right. You've got a very simple and easy-to-use client. Your directory is good. Your use of wideband audio usually gives outstanding audio quality. Your ability to work from very different network environments and through firewalls is great. Some of us love that everything you do is encrypted. You work across the major computing platforms. You make a great product and because you have hit a critical mass with so many of us there, we like to use your product.

But... with statements like this you're living in the same delusion that Facebook has been in until recently. You see, there's this wee tiny little problem:

You are NOT the only island!

Sure, you're probably the largest island with the most parties and easiest docks to land at. But there are a lot of other islands out there. Some of them are other services with whom you admittedly compete. Some are startups. Some of them are the traditional carriers now offering VoIP services to consumers and businesses. A lot of other islands are the companies and organizations now wiring themselves up with IP-PBXs or using back office software from Microsoft or IBM to "voice-enable" their infrastructure. Ditto for some cities and towns that are doing the same thing. In some cases, those islands are wiring voice so far into their business processes and systems that it's truly amazing.

Now some of us, seeing all these islands out there, say... "Hmmm... why don't we just connect the dots?" Let's build some bridges or high-speed ferries between those islands. Let's get them talking together. Let's interconnect the islands and build the new infrastructure. Let's bypass the old PSTN and build the new voice network entirely across the Internet. Let's forget all about those geographic boundaries... let's let voice flow to wherever wherever someone can get an IP address. Anywhere. Anytime. Let's interconnect business systems with other voice systems.

And you know what? We're doing it. Slowly. Very slowly at times. But we're doing it. We're using protocols like SIP and RTP and all the many others coming out of the IETF. We're creating "mashups" and using XML flavors like VoiceXML and CCXML to weave voice into the web. We're starting the interconnection. We're enabling businesses to connect to each other and dial each other directly. We're using SIP trunking to let local systems make and receive phone calls from other parts of the world. We're giving people their choice of endpoint... they can use a range of "hard" phones (traditional pieces of hardware) or "soft" phones (like you are). People can ring my deskphone simply by calling "[email protected]" using their SIP phone... goodbye hard-to-remember telephone numbers... hello user names.

Oh, sure, we've got lots of problems still to work out. Security is a huge one. You are extending your trust boundary out to include other networks. How do you know they won't send you tons of voice spam? Or abuse your network? Or run up bills on your dime? Privacy is another one. How do you show others only the information you want to? Sooner or later the various governments and tax authorities are going to wake up and realize how badly they are going to be screwed out of revenue by all we are doing - and we're going to have to deal with that. We've still got to agree on how to do certain features between systems. We've got a lot of work to do.

But we're doing it. We're rewiring the phone system. We're creating a new one, not shackled by its history.

The question for you all at Skype seems to be whether or not you want to help build that larger interconnected world. Or whether you want to just hang out on your island and hope that if you throw big enough parties and advertise "Free Beer" enough that everyone will forget about their own islands and just come over and join yours.

You know what? You'll get a lot of people to come on over. Today. And probably for some time. You've got a fun island to hang out on.

Meanwhile, the rest of us will keep on with our rewiring and remixing. We're building the fabric of what comes next. We're coding the DNA for the future of voice. We'd love it if you joined us. I'd love it. It would be great if I could call my colleagues on SIP extensions from directly within Skype. Not through some Skype-to-PBX gateway that really winds up running multiple instances of Skype... but through an actual SIP gateway. I'd love it if I could give them a SIP address like "[email protected]" that they could use to call me on my Skype client wherever I was. You know, I'd probably wind up using my Skype client more if I had that capability! You have a great UI. Why shouldn't I add my SIP contacts there, too?

What SIP contacts you say? Yes, clearly I'm an "early adopter"... one of those geeks who goes around chasing bright shiny objects. Guilty as charged. But each day what I do is becoming easier and easier for others to do. And you know what? If you supported SIP contacts, those of us who talk and write about topics like this would probably do a lot to evangelize you. We'd actually help you with your marketing.

Now you do make this excellent point:

"In order to provide richness, we have to create our own protocols," Oberg added. "SIP and the standard [VoIP] protocols simply can't do it."

You're right. Almost all the traditional vendors in the VoIP space do use their own proprietary protocols to give the rich communication experience people want. Cisco. Nortel. Avaya. Alcatel. Mitel. Others. But you know what? Their hold on the market is being disrupted. Lots of new players coming in. Big ones like Microsoft and IBM - who are interestingly supporting the open standards we're using. So the traditional vendors are evolving, too. They're supporting SIP for interconnection. Sure, they still have their parties on their islands and show people how great it is there, but they do allow bridges to be built. They understand the need to interconnect.

You're right, too, in that SIP only supports basic calls. We know that. We're working on it. So come join us. Join the IETF mailing lists. Send someone to IETF 71 in Philadelphia in March. Advocate for how we should interconnect to you. Building the Interconnect is long, often glacially-slow work, full of many people with different agendas, many of whom will all disagree. Join with us. You'll lose some battles and win others. But together we might just have a chance at making it all happen.

Or... just keep hanging out on your island throwing parties and trying to attract new people. Maybe it will work.

In the meantime, please don't say that customers aren't asking for interoperability. Count me as one who is:

"I would love to call a VoIP provider on a different network!"

I bet if I ask around, a few of the people I know would like that, too.

If you read this far, thank you for listening. You can now return to your island. Meanwhile, we've got some rewiring to do...

Technorati Tags: , , , ,