« What is Google doing at FMC? | Main | The Second half of the year Master List »
July 05, 2007
Can VoIP Be Sexy
I’ve asked Jay Batson to be part of the unconference 5th track discussion at Fall VON. As always Jay has insight that brings some interesting points to light. Here was the start of the thread.
So here's the thing: VoIP pioneers built a lot of cool stuff. But we had all these visions about new stuff users could do. And, IMHO, most of those haven't appeared yet. What went wrong?One could make the case that:
- VoIP got co-opted by traditional and new communications service providers, who have mostly given us POTS over a new medium with a few new features, all at a new price;
- SIP suffered from this co-opting, and grew and grew and grew as it tried to replicate all the old world for the providers who were writing big checks;
- Skype showed us that making something "just work" and provide a good UI, and _give it away_ can build the biggest businesses fast;
- Open source shows us -- in _other_ domains - that it can change markets. But VoIP open source has some problems.--- Asterisk is fine, but as far as end users are concerned, it is a thing whose operation is controlled by an IT guy; it's not end-user software.
--- Microsoft hasn't really innovated in a dramatic way.
--- There is some good infrastructure open source code for SIP, but much less on the client side
--- There _is_ some good client and server code for XMPP/Jabber/Jingle.I've tried to add VoIP to my product (http://www.plumcanary.com), and even though I'm a SIP guy, it's just too hard. I've seriously considered XMPP/Jingle....
So, what happened? Did VoIP get co-opted by big-check-writing service providers, who distracted the industry away from new innovation by paying for us to re-implement the old world over IP? And if so, should we care? Should we do anything about it? If so, who will pay for doing it?
Is this a topic that is even worth talking about?
My reply....
Lets start at a basic question.Is it possible to make voip sexy, or is transport just a function in a world of communication that can't be spiced up.
An analogy would be Cars and Road. You made in the past Jay the Maserti of SIP phones. But making the call is a road. Roads are not terribly sexy. Sometimes you can admire the engineering, but rarely do you say this road is cool. Its just the pavement.
Look at this iPhone nonsense. All this noise about what you can do with it, but they never make a call in the commercial which is probably good since recent studies indicate they can't make it in real life either.
Even Skype not that the bloom is off the rose, looks like transport with a nicer interface and good engineered (transport/road). The skype devices are iPhone equivalents.
So with all that said, should we really be surprised that Telco 2.0 is not as inspiring as Web 2.0?
Any more thoughts?
Jay’s Reply
You're thinking in terms of handsets, Carl.I think what is sad to me is that VoIP hasn't simply become widely integrated into desktop computer applications.
You know how in various applications (e.g. web browsers, MS Word, etc.) URL's automatically get highlighted, and are clickable? Well, I'd like everybody's email address to be _not_ just clickable to trigger an email being sent; I'd like the click to raise a context
menu that says "Send email" or "Voice chat". In tons of applications.For this to happen:
- There needs to be code available either in open source, or built into the OS, for developers to use;
- Developers shouldn't have to build server infrastructure to supply this feature - it should just work;
- It must work through NATs without user intervention;
- It should be free for everybody to use, the way Skype is "free" (free for on-net calls, costs money for PSTN gatewayed calls);But on every point above, we (the industry) failed.
We have nothing "interesting" in VoIP because it's not accessible to the people who write software that normal people use. Why aren't 95% of people in SecondLife using VoIP to talk instead of typing? Why doesn't Basecamp have voice chat in it's collaboration portal?
Where's the ad-hoc voice chat rooms? And I keep speaking of voice - but I _should_ be saying video for all of the above.
So, I don't want voice Sexy - I just want it pervasive. And it isn't because this new idea got co-opted by the old guard. We (myself included) chased revenue from traditional telecom sources because we needed to. This meant we got co-opted, and left us in the non-VoIP-enabled software world we now live in.
Posted by carl at July 5, 2007 12:34 PM
Comments
I installed Asterisk on my server at home recently. It was definitely not something Joe Average could have done; that's clearly a problem. And now that I have it installed, I'm not really using it at all.
Why? No clients. All of our computers at home are Macs. Theoretically there are clients, like OpenWengo. I actually found about five or six clients that claimed to work. But none actually work - OpenWengo just crashes on startup, for example. A lot of the clients that supposedly work don't actually even build on OSX.
From what I can tell it seems like all of the client work that's being done is funded by people who are trying to use the client they're developing to sell a service. They're not developing a client that will talk to your asterisk server - they're developing a client that will talk to their SIP<->PoTS gateway. And any functionality that's not solving that problem is at least a low priority, and possibly a negative priority.
From my side I really don't know where to get started - it seems like the right thing to do is write a client. It seems like the libraries the client needs are there, and at least in theory can work. But I don't really grok enough of the system to just sit down and hack one out - I need to climb the learning curve first.
Do you know of any good resources for doing that, aside from reading all the RFCs and trying to mentally integrate them?
Posted by: Ted Lemon at July 5, 2007 03:36 PM