Unusually for me, I did not know Matt personally and had not even had a pre-interview call with him. So this was both a live interview and a first meeting. So it's a bit more two-way and with a lot of informal chit-chat, particularly during the first 8 minutes.
You can download it as a 96kbps MP3 here (19 meg, 50.05 minutes). Sorry for the slight clicking sound, it was entirely my fault due to playing with a software setting.
Additionally the full transcript is below. To distinguish between us I've indented Matt.
Transcript
...I just want to jump in and say thanks to RebelVox for sponsoring, either a lunch or a breakfast. I can't remember which. You probably know. Do you, or do we both not know? [laughter]
I believe it's a breakfast.
Okay, breakfast, thanks a lot for that because the ticket prices would be double. The venues, especially that one is astronomical. Nobody wants to be paying double the current ticket price. So thank you very much. Can I ask you why RebelVox sponsored?
Sure, I attended the first eComm in 2008. I was blown away. It was the conference that made the best use of my time in my career of attending conferences. The content was high. The schedule was rigid. People got to the point. It was a huge fire hose of great information. That is the world we are playing in, that we at RebelVox are playing in. We want to see more of that kind of thinking and the kind of people that go to eComm. Those are our people. We're happy to be a part of that.
Hey, that was absolutely fantastic, not even scripted [laughter]. Perfect answer [laughter], you've blown me away.
Is that going to be a sound bite and a little quote on your blog now? [laughter]
I think it will need to be. That was perfect [laughter]. I'm going to need to interview you more. [laughter]
Maybe next time I'll prepare. [laughter]
Yeah, prepare a two pager [laughter]. I should probably get into this, but because we've not spoken, you might find it could turn into a two-way instead of a one-way. For example, with Martin, because he is a close friend of mine, Martin Geddes that is, I was able to ask him questions I knew were good for him because we'd been chatting together for years. Wheras we're just getting to know each other here.
So anyway, I have to say that meeting with yourselves made me laugh because I posted on the blog, about the death of telephony and the you emailed me saying, "Hey, have you been influenced by RebelVox?" [laughter] And I had to reply to you and say, "No". I looked into it after you said that and I thought you guys were influenced by me. [laughter]
We all had a good laugh about that. [laughter]
So each end has been laughing about each other end, thinking the same things, and thinking each other had influenced each other. Then tonight, when I asked you by email what you wanted to speak about, it reminded me so much of a chunk of the opportunities that I helped Martin Geddes highlight, by means of a long interview with him nearly a year ago. I don't know if you read that interview I had with him?
Yes, I have.
You are a good eComm citizen. [laughter]
The overlap in thinking was just too great to ignore. It's really interesting stuff. I read it, start to finish.
Hey, that's pretty awesome. We should have spoken before but I think from now on, we're going to be doing a bit of speaking. It's a shame we haven't had a chance to meet up but I'm sure we will, a bit longer than thirty seconds in passing.
You know, we were in stealth mode for a long time. We weren't really talking to people about what we were doing. That explains a lot of why people such as you haven't ever really talked to us because there was really nothing to say, until very recently. Now, we're ready to tell the world what we're up to.
Maybe I'll say a little about myself here. I can go on forever so I'll keep it short. I I was away to say that you don't want pillow talk from me, that does sound a bit outlandish, but I'm allowed to say these things since it's a 2:00 a.m. here.
Anyway I think you've gathered, since you're clearly in the eComm community, that I have, or had, not sure which to say anymore, built my career in telephony. I'd been reading the British Telecom Engineer Journals, it was a general post office back then, since I was twelve. By the time I was fifteen, I was reading the CCITT signaling specs. But around four years ago, I realized that telephony was dead. It hit me like a ton of bricks because the thing I love most, I realized, was dead. It's dead, dead, dead, and dead. Yes, we're using it every day, or at least most people are. But it's categorically dead! There is going to be a massive wakeup to that!
Anyway, so, when I looked at what you had been saying, your proposed talk for eComm 2009, you are also saying that telephony is broken. I say it's dead because I know, in the long term - well I won't go into it all here, but it is certainly broken. In the short to medium interim, the money is in fixing it.
Can you elaborate on why you believe that telephony is broken?
Sure, it's funny that you think I'm not being strong enough by saying broken versus dead [laughter] . I thought I was being a little too strong by saying broken. I think that telephony, as we know it, will probably be around for quite some time. It's just in too many places. The bit about it, that I think is broken - it's not so much that telephony has really changed in some way that is now broken, it's just that society has sort of changed around this thing that is essentially unchanged since it was invented.
Now that information moves around very easily, and very inexpensively, we can communicate over great distances for the same cost as communicating over short distances. All of this inexpensive communication in our lives has put crunches on our time. As Martin points out quite succinctly, time is the new scarcity. That's exactly right.
The thing that's broken about telephony is it doesn't respect the time of the people who are talking through it. That's what we aim to fix.
Can you give me some examples? I tend to be a little bit harsh in my questions. It's because the whole thing interests and excites me. Can you elaborate on "it doesn't respect people's time"?
Sure, I hate to be regurgitating things I've just read on your blog, but as you point out in something you wrote, "Party A summons party B with a bell, and that's why we call it a 'phone call'". I want to talk to you, so I ring you. I wait for you to decide whether you want to talk to me or whether you want me to go away.
It puts you in this really uncomfortable position, where you might be talking to somebody in real life or might be doing something important. You might be driving a car, or whatever you might be doing. I'm saying I don't care. I demand your attention right now. There is nothing in between. It's [telephony] a continuum with only extremes. We're trying to open up some spaces in between the "you get to talk to me now," or "you don't get to talk to me now".
The way it doesn't respect your time is, sort of, like it wasn't designed to respect. This granularity of time wasn't even considered because of the other time it was saving, of travel time. It used to take a long time to get places. It used to be a big deal. You couldn't fly places. Calling was a lot cheaper than going someplace.
We have so many different ways of moving information around, now, that time does become more important. Telephony can respect people's time. At the moment, it's just ignorant of it.
I had to smile there because another little question came to my mind. Have you noticed how the B party still behaves, for the most part, as it had in the past? Because the A party was the one who bore most of the cost, i.e. because of the scarcity was the networking, they were paying large rental costs, i.e. call cost was higher than the human cost, then B parties feel obligated to answer. Yes, the older demographic is still up there, but it's still in the bulk of our society that you must answer a ringing phone, no matter how important the context that you're in. Do you notice this? [laughter]
Absolutely, the problem is on both ends. Not only do you sort of have to be interrupted, but then you think, "Geez, A party is already taking the time to call me. I might as well". They have sort of forced this obligation of your time and attention on you, out of nowhere. This is exactly what I mean when I say it doesn't respect people's time.
A lot of the problem is because you don't know what the value of the inbound call is. You don't know if it is critical to you or if it's something that would be better suited to another channel or another time. It's completely and utterly blind. Would you agree?
Yes, it's the same sort of live voice stream, whether it's just chatting, just calling to say hi, some short message, or some long message that's a conversation. The way you initiate it is the same, in the current model.
If you look into calls, there are so many different types of calls. For example, in his interview, Martin had been speaking about how a lot of calls are pure rendezvous as he calls it. They're actually, calls in order to organize the real call, like, "Where are you," "I'll be home in five minutes," "Okay, I'll call you then". It's kind of funny that we're using calls to do the rendezvous to have the proper calls.
Then, some calls are just information transfer, like to pass over credit cards. It's very strange that we're using the system to do all these things in such an inefficient way.
It's funny; I think the ultimate problem is that the only building block you have to work with is the call. Even after you have something like some other way to set up that call, let's say you get one of those; you can send a text message or you can do some other thing that helps you set up the call. At the end of the process, you've rendezvoused now. Now you have a call. I think even that is going to need to change for the brokenness to go away in telephony.
Could you expand upon that?
You and I are now talking on the phone. We're using fancy computers to do it. That's really neat, but here we are; we have a full duplex telephone call. If the call were to drop because the network screws up or if one of us needs to stop for a second and tend to another call, the things that each of us say just sort of go away. They're live only. Live is the building block for all of telephony.
The live, two-way call is the way it all works. I think that is actually broken. It should be more flexible in that. You should be able to have some time-shifting, if you like to call it that. You should be able to sort of listen whenever you want to listen, to whatever kind of media is headed your way.
So, calls take on some kind of permanence or semi-permanence, if you want?
That's another way to look at it, as a side effect of doing that this way; you end up having more of a context, more of a history of your interaction. I think we have to get rid of calls. I think that's ultimately the problem and why we find all of this so inconvenient. Once we've completed this rendezvous, we still have a call.
For some things, of course, a call is the right thing. For this conversation right now, this is the right medium. We need to have sort of a two-way...
It's handy. [laughter]
It works very well. I wouldn't suggest that we should do some sort of asynchronous, wildly delayed scheme, like a 187 milliseconds of latency is pretty nice, halfway around the world. The thing is; for a tremendous number of different usages of voice to communicate, it's actually not. But it's the only thing we have to work with if you want to talk. There's really nothing in between - of us talking, if you want to use voice. There's nothing in between us talking live and us going into each other's voicemail.
If you take the word telephony, generally telephony means a particular thing, what you're saying is - do you believe that the term telephony will disappear, will cease to have meaning, going forward?
I think that is one possible side effect. I think telephony is going to be one of these words that ends up being meaningless, like broadband. Broadband used to mean one thing and now it means fast Internet, I think. It will take on some sort of different term, or not, or maybe we'll always use it to refer to this kind of old-style of phone calling that we used to do before we figured out something better.
If I just jump back to this rendezvous thing, just looking around the Web here, I have ADHD [not officially, just get bored easy] so I'm looking at a couple of blogs here. There has been quite a lot of buzz in the blogosphere lately around VoIP is dead, or VoIP is not dead. I kind had to chuckle to myself because they're arguing about the technique to transmit the media itself. In the 80s, we went from analog to digital. In 1990, we went to ATM. In the early 1990's...
Yeah, it's the plumbing.
It most certainly is. I was away to say that Idon't remember people getting excited about digital but I did get greatly excited. Digital actually had a lot of huge promises when ISDN was wrapped in, but operators unfortunately failed [laugher] to realize what could have been the Internet back then. The same actually happened in the early 1990's [ATM].
They're getting exciting about or arguing about the plumbing. I don't normally share too many personal opinions, believe it or not. I've been fairly quiet the last four years, but I'll say this; the real revolution is going to be around what I call "person-to-person signaling". The revolution isn't in the media transport, but actually the signaling.
A sub-set of that is what Martin had called the "rendezvous process".
Anyway, you said to me, in the email earlier, "People have attention synchronization challenges in their lives". Can you tell me how you plan to solve these challenges, or how you are solving these challenges? I'm sorry. I don't know if you're already selling [solutions]?
You can't go to the website and buy our software or anything, but yes, this is something we are doing. What I mean by that is that the synchronization of attention is really the critical aspect or barrier to communication happening. Sure, one way you can synchronize attention is with these improved signaling schemes and then you end up with this rendezvous. And then, you've got full attention. I think it needs to be a broader spectrum than that. Yes, I completely agree with that and I'm totally on board with your improved, human-to-human signaling, but I think that the signaling actually is the media. It's not necessarily just in service of getting some live call going. It is the content. If that content is best relayed in a live way then that's how the people will use it. They should be able to get what they want. The signaling, in and of itself, has value as content.
Actually, that's a topic we should really focus on if we do another interview before this conference. That is a very exciting topic. It takes a lot of time to go into that. I'm sure you'll agree because [laughter] obviously, you know what I'm talking about. We really should focus on that for another conference interview. It's a whole talk in itself. It's a terribly exciting area. You've spurred me on with some excitement by bringing that up [laughter]. It's one of these other things where I kind of thought I was on my own, or fairly on my own. That will be a lot of fun, to discuss that, because I've been doing a lot of thinking about it.
I would love to do that.
Do you want to say how you're solving those challenges?
I'm sort of dancing around the issue a bit, but the signaling, the messages, or the content you can deliver to somebody else, becoming the purpose of what you're doing, runs into a problem with voice. Because the ways in which voice is currently delivered are pretty awkward. You have to get a circuit, call people, allocate a bunch of resources, get your QOS going; or get your time slices [TDM, or whatever it is that you need to so you can talk to somebody, even if you ultimately end up with voice mail. That's still what you need.
What we're doing is throwing away the requirements for circuits, reserved resources, and all of those sorts of things. We're pushing the complexity of managing that kind of stuff all the way out to the edges. You have these powerful devices now, your iPhones and what have you, with relatively speaking, very powerful computers on board. They have multitasking operating systems and advanced networking capabilities. None of these things is really being used to kind of change the way that voice works; they're being used to run applications. That's neat and that's great, but we're going to harness the power of these smart devices to process your voice, to deliver communications with your voice, in a way that is kind of separate from the requirements imposed by the network.
Once we're free of that, we're free of requirements, strictly, of people's dedicated attention. You can have a more arbitrarily time-shifted way to listen to incoming voice or video. You can send things without waiting for the person on the other end to acknowledge, basically without waiting for the network, without waiting for anything. We're respecting the sender's time by letting them talk immediately, and respecting the receiver's time by not necessarily interrupting them, allowing them whatever level of participation they want, at that time.
Even though it's 2:00 a.m. here, and I'm fairly tired, as you might hear in my voice [laughter]; I don't have the usually spritely thing on. But I have to say I'm really glad you came along to the 2008 eComm conference [laughter].
Why is that?
You're very much in tune. You're certainly thinking ahead. I know the direction, from what I'm hearing. I mean I know these areas well and we should really do some more chatting again, and record it, but maybe some offline? I tell you; it's terribly exciting things. It is so good to begin feeling that other people have noticed the same things. Because the industry has been so stagnant for so long. I'll be honest with you; I don't read that much in comms, blogs, and news just because I'm bored stiff. That's not going to win me many friends I guess, so maybe I shouldn't have said that? [laughter[
I'm with you [laughter].
I'm very very bored with comms [publications]. It's been rehashing the same old stuff for so long, and yet, the opportunity has never been bigger. Maybe I'm just really fortunate; I was lucky enough that Cisco said they would pay for an engineering doctorate for me. It was around the time I realized that telephony was dead. So I then spent seven days a week, twelve hours a day, fourteen hours a day, contemplating the future of telephony. I know I spent three years at it. I must admit; it might sound a bit extreme, but it actually destroyed my life [laughter]. I can say I destroyed my life for three years [laughter], thinking about the future of telephony. It was head breaking.
I've not been public, or so public at least about the conclusions I've came to. I've only kind of dabbled with mentioning things publicly, for various reasons. It would actually take so much time to get some of this stuff out. It was radical conclusions I came to. It's really good to hear others who are getting into the same conclusions. I would actually like, at some point, to begin saying where I see things longer term.
But anyway let me get a bit more down to Earth here. The economy is dire.
That's putting it mildly.
I'm being very polite tonight [laughter]. Communications and transport are the two backbones of an economy, and yet, the communications we have today, predominately telephony, is so inefficient. It's incredible; we can drive massively increased efficiency. That's why it will happen. That's why telephony has to die. Efficiency will always win if you believe that globalization will keep on trundling on.
Today, for example, I wanted to post speakers [of the hifi kind]. I'm in the U.K., so I called Parcel Force, which is a courier.
[laughter] By "post", you mean ship physically? That very doesn't mean anything to Americans.
Ship, yeah I'm sorry [laughter]. It's like the old car hire and car rental. It confuses the hell out of people. I've started saying motorway or freeway because if you say dual carriageway, people start off, "Where are the horses?" [laughter] At least I'm not calling taxis hackney carriages. [laughter]
I do like how you used "outwith". Is that where the accent goes? Is it on "out", or is it on "with"?
I'm confused.
In your blog post, you used a word that I've never seen before, which is apparently a Scottishism, you jammed the words out and with together, as one word.
I actually thought I'd wiped all Scottishness out of my speaking. Not only working in the States, I had a couple of ladies who were Americans. I soon realized that they didn't understand a lot of the things I said, not because they didn't hear the words but because the words are not in an English dictionary [laughter]. I wasn't aware of that one. I need to look into it.
I thought for sure it was a typo, but Google to me the truth, which was, nope - Scottishism.
Yeah, yet another one. There are also three hundred words for rain [laughter]. I can tell you exactly what - if I could use some Scottish words for rain, you will know exactly what type of rain it is. You will know the speed, how quickly you'll get wet, just by one word. In Scots, you can also say a word that begins with "f" and ends with "k" and you can say that in three hundred different ways, in order to have whole conversations with that one word.
That's weird; I have no idea what that word could be, but I would love it if you would tell me about your experience with Parcel Force.
Sure. I called them after being at the website. I got a deep, IVR tree that took me three minutes to traverse, including the ringing time. At the end of the three minutes, and this happens often, it told me to dial another number (surprise). But it only told me once. I didn't have a pen. I wasn't at a PC at that point. Guess what; I had to redial, traverse the same old IVR tree, to hear that number again. I noted it down, dialed a new number, got to large IVR trees, then I got hold music. I had no idea how long I was going to wait, so I just hung up. I actually didn't bother getting a price from that company. I went to a company I already know because it's a lot quicker time-wise. That was me trying to establish a session about trying to obtain a piece of information, via telephony. It was burning up so much of my time that it was cheaper for me to over spend on a company I already knew, than to engage a new company.
Are you able to highlight any inefficiency which you can see, other than kind of what you mentioned earlier?
Sure, in your world, what would have been the ideal interaction on your phone, with Parcel Force?
I dial a number and somehow magically I'm through to the right person who can actually answer the question I had, without playing about, without putting me on hold, and without passing me around or making me push buttons - speed of information.
For example, one way that your interaction with Parcel Force could have gone is that you could have found them. You could have gotten your device or whatever it is that you talk telephonically into, and you could've said, "I want to post some speakers. Here is where they're going. How much is that?" You could go do something else. You could do whatever else it is that you wanted to do [laughter[. Then that communication could get routed through the Parcel Force whatever, call center or however it is that they do this; somebody will answer that. Maybe they have some way to scan it for key words, and more efficiently get it to the right person on their end, but they can give you an answer.
The problem is that you have made a live phone call into some system. The only way they can interact with you is "live". You would like to get to a person, but what they would like you to do is hang up because people are really expensive. So that's a big inefficiency that we want to fix.
I like that. I would have liked to have been able to inject voice into that company; I ask a question within thirty seconds and I don't care if it takes an hour for them to come back to me with the answer. In what you mentioned, would you be suggesting going voice to text, then process the text for semantic meaning, to help in the routing process?
[laughter[ I'm definitely suggesting that. Computers can parse text and act on text very well. Even if they don't get the full meaning, chances are they can at least get your message into the right bin, wherever it's supposed to go, oops I can't use "bin" as that means trash, right? They get it to wherever it's supposed to go having converted to text, and still maintaining the original voice, to the extent that it's useful.
Not only that, but even if they get it in the wrong pigeon hole [laughter], whoever receives it can say, "This is in the wrong pigeon hole; send it here," and the machine can learn.
Yes, but you can't really do that if you have a phone call. Someone dials in, they have this long - they say calls are recorded. But, the whole call may be recorded but how do you separate out the part when you're talking about this, versus the part when you're talking about that? As you put it, injecting voice into that company with a very specific request, encapsulated within the boundaries of that voice injection. You allow it to be converted to text in a very meaningful way.
We've actually highlighted two areas tonight, which I want to say are exceptionally exciting. I might not sound exceptionally excited [laughter[ because of the time here and because I need to catch up with this much later on.
The first one is that there is a revolution around the signaling. It's not to do signaling on IP [laughter], the change of transmission of signaling. It's the fact that you go person-to-person. It's taken the signaling from a circuit level to a sociological level. It's actually, where sociology intersects with computer science. The whole rendezvous thing is a subset.
Secondly, what you've raised is absolutely fantastic which is you cannot process the media in a call. It's fairly unprocessible, as you say, they record it, but you can't understand it, you can't parse it. Once you parse it, can understand it, it becomes terribly exciting input to play with, to build cool apps, better things, and make things more efficient.
Is that a direction that RebelVox is looking at?
Yes, we're looking at it in terms of enabling those sorts of things to be built. We're not trying to define all of the different, actual uses of something like that. We're trying to say, "You can put your voice into the system, in a way that makes it much more useful than the current system you have. You can map this onto your business process in the ways you, the business owner, know best. We're going to give you the tools, this platform that can make all these kinds of things possible.
You mentioned business processes. I can say this; eComm 2009 will have a lot of talk about making business process more effective with communications. That's a really nice area to be looking into. You're really going to love the next eComm because I know the talks that are lined up.
I'm looking forward to it.
If I think about this, and I look at what you guys have been doing, you have also been looking at emergency services, in particular, tactical radio. Is that correct?
That is correct. The origins of RebelVox sort of come from the tactical world. My business partner was a Special Forces communications sergeant in the U.S. Army Special Forces. He served in Afghanistan. He ran into a number of very challenging communications problems that were the origins of what we're doing here. They sort of boil down to, you can't be on more than one channel at a time. We thought about that and how could you fix that? That's where we realized what the problem was. Of course, you could build a radio that is literally on more than one channel at a time, but that's not a very useful thing to have because you only have one brain. You can't give live attention to multiple, other people, at the same time. That's where we went with it, this sort of attention synchronization.
It turns out that most of the tactical radio communications don't have to be strictly live. They're transfers of information. The way those systems are used is almost always an information transfer. You press the button; you transfer some information to somebody else. You're not even expecting it to be full duplex communication because most of them aren't.
The problem of synchronizing your attention across more than one channel or more than one conversation at a time led us to realize this problem in the tactical communication space because it's dangerous. People's lives hang in the balance of communications being effective. Everybody has these same sorts of attention synchronization problems. You and I both have this problem; we just don't notice it. The consequences are not so dire for failing. No one dies if I don't answer my phone. It's sort of inspired by tactical communications but we think it's applicable to anybody who talks into a device to somebody else on the other end.
I totally get what you're saying. It's applying what you learned from looking at telephony to the tactical radio side of the house?
When we talked to people in the tactical communication space, they always say, "Wow, when can I have this on my cell phone?" [laughter] It's pretty interesting. That's part of the unique perspective that we're coming at this from. It's understanding that tactical communications mindset and bringing some of those elements into commercial telephony in a way that saves everybody time.
We've been on this call for forty-five minutes. We were actually meant to do fifteen. Because it's the early hours here and I need to get up for a plane also, I'm going to need to shoot off this call. I can say this; it was really fantastic having this chat with you. If it's okay with you, what I will do is I will not edit it, possibly me even describing words in Scots you can say [laughter], and just put it out there. I think there has been valuable content on our meeting call, our first call together. I think it will serve as a basis to link to. You and I should really do a little bit of speaking offline and probably do this again once you and I are more in sync. So far, it's been fantastic.
I totally enjoyed this conversation. I just want to say thanks for staying up late [laughter] . I appreciate you willing to be a little flexible in your schedule. It made it possible to actually talk. Thanks.
It's no problem. I tend to find that most of the people I speak to are in the PST time zone so I often find I'm up in the early hours. I really appreciate the enthusiasm I got with the speaking proposal from RebelVox. I thought that was fantastic. I don't want to go back in this whole interview again, but I have to say it to you; how long have you guys been researching it? It sounds like you guys have been doing serious research.
Two of us have been thinking about this problem for almost five years, now.
Hey you were doing it the same time as me [laughter] and also Martin. We've been like these nodes around the world thinking about similar things. This is so great.
That's really very funny. We've been talking about this for about five years. It sort of came together as a company about a year and a half ago. We built up a team and started working on building it. I've been sort of poking around, kind of as you describe, but not as anguished [laughter]. My investigation was more fun. I didn't have a lot invested in the telephony world, but I was really curious about how to save time. I realized that every time somebody called me on the phone I just felt as if I had travelled back in time to when it was novel that people could call me on the phone. I've just been trying to figure out how we could make this better.
A good word used there was "anguished". Imagine being me for a little while; you grew up as a kid, your love is the telephone and telephony, the heart of telephony, the intelligence, the nervous system is Signaling System No.7 (SS7), so you decide you want to become the world's top expert in it. You believe you've become that [laughter]. You think you've gotten where you wanted to be and then as soon as you get there, it suddenly slaps you right in the face that that very thing is actually dead. Anguish is a good word! But actually, the opportunity I learned from those three years which must have overlapped what you were speaking about. The opportunities I see are absolutely tremendous. This is just a part of it. So I really do look forward to chatting with you a bit offline. On that, I better get going. Thanks again, for your time.
Sure thing, I really enjoyed it. Talk to you later.
Thank you.
Subscribe to this blog's feed
Circuit switching is on the way out (is already out) but I don't quite understand why voice mail doesn't solve most of the attention synchronization problems.