Conversation With Alan MacLeod of Ramp
In this Executive Insights podcast, BCStrategies' Jim Burton welcomes Alan MacLeod, VP Product of Ramp. Ramp specializes in optimizing the distribution of video behind the firewall to significantly reduce bandwidth consumption and alleviate network congestion for the best quality viewing experience. BCStrategies Experts participating in the conversation include Phil Edholm, Michael Finneran, Kevin Kieller and Francisco Kattan.
Jim Burton: Welcome to BCStrategies Executive Insights, this is Jim Burton, and I’m joined today by a group of our Experts. We are here on a call with Alan MacLeod from Ramp. Alan has been a friend of all of ours for quite some time and we are excited to hear the latest about what’s going on with Ramp and have him explain to us some things that probably a lot of us don’t know. So Alan, I’m just going to turn it over to you. You can give us some background and some information, then we’re going to go our group of Experts and start asking you some questions. So, Alan, take it away.
Alan MacLeod: Thanks, Jim. So yeah, some background on who Ramp is and what are the problems we’re trying to fix for customers. As we move through the industry picking up more and more the use of things like video and high-quality communications between individuals and companies and between executives to staff, etc., we’re starting to see a significant drive in the industry towards more and more and more corporate communications taking place over streaming video. And many companies, companies like Brightcove and Kaltura, have been in this industry for a long time and have been solving this problem for companies. But they’re seeing the demand grow continuously. More video events, more people connecting to video events.
And what that’s driving, given that most of these platforms are up on the internet, is a challenge of how to get all of the people who are sitting on the corporate network onto that video event. And so the problem seems fairly straightforward; they’re just going to connect and download a fairly slow speed, two or three megabits per second video. But of course that scales. When you have a thousand people sitting at a single corporate location, that’s not a three megabit per second video now, that’s a three gigabit per second video. And when that scales up to maybe 10,000 people at a campus, that’s a 30 gigabits per second video that has to be now pulled down over the internet. And when it’s a live event, particularly in that scenario those bandwidth requirements are very high.
And so, the role that Ramp plays in this is that we provide the ability for those corporates to solve that internet connection problem. We can take a single video event, bring that down onto the corporate network, and then basically fan it out to multiple users on the network. And so, the scenario would be that a very major corporation, maybe 100,000 users, could now pull down a single instance of the video and all 100,000 users can watch that same video at the same time. And so in a company where the CEO is trying to reach every person in the organization to explain what’s happening particularly in real time, and again it’s very much about that highly connected video experience and the highly sort-of personal expression to see someone who’s talking, see their face as their talking, and the credibility that brings. As well as having content supported around that conversation. How to get that to all the people at the same time means you have a big network challenge. And it’s not a challenge you fix every day, therefore it’s a challenge you need to fix elegantly. So Ramp is the way to fix that elegantly. And we do that by protecting corporate traffic which is the business traffic whilst maintaining high quality video.
So, at a 50,000-foot view, Jim, that’s kind of who we are and what we do and the problem we’re trying to fix for customers today.
Jim Burton: Great, well thanks Alan. So we’re going to go through, have a group of our experts ask you some questions to really get to what we hope are the kind of things that people that would be looking at your product and your technology would want to know about. And nothing more appropriate than to turn it over to our in-house technical guru, Phil Edholm. Phil?
Phil Edholm: Hi Alan. Well so I guess it’s a lot less technical, and a lot more really around what percentage of traffic do you really see yourselves being involved in? And then kind of the second half of that is internal traffic versus external traffic. So, if the CEO is addressing the employee base and the majority of the employee base is inside on the WAN, are you also looking at how you manage that across sites? Or is this more focused on external traffic coming into the camp or into the corporate environment? So I can kind of get an idea about how you see this fitting into the overall video management and the overall problem.
Alan MacLeod: Yes, so one size still doesn’t fit all, unfortunately. The industry continues to be addressing customers who refuse to become homogenous and identical. So, it’s hard to give one answer to the question you just asked.
What we are seeing though, particularly around video communications, it falls into two camps: live events as well as video on demand for things like training, or for corporate marcom communications, things like that. So what we’re seeing is more and more companies put that content out into the cloud. And the benefit they have there is that there’s no infrastructure for them to manage. The people who create the content can instantly push it up onto a portal to present it very effectively to their users. They don’t have to worry about storage, they don’t have to worry about maintaining systems. All the normal arguments we see for why cloud systems are so powerful and so effective.
And so for those companies, it’s very much can you solve how do I get across that internet connection without buying an internet connection for the one-off event? And so again, you know with a 100,000-employee company for a live event driven by the CEO you’d expect 50 plus percent connectivity or 50 plus percent engagement on the live event. And in that situation, you have to size your internet connection to, example 50,000 times two gigabits per second.
When you move to companies where they have tried to solve this as an on-premises problem and then many companies have solved it this way. And there are many elegant ways of doing this and many elegant solutions for on-premise solutions.
The second part of the question becomes, okay my WAN is sized appropriately for my corporate business but have I sized it again for that live event. And so in that respect, we do fix for companies a lot of internal distribution problems as well. Historically, large enterprises have moved very slowly in terms of refreshing internal infrastructure. So, moving from 100 meg internet or 100 megabit to the desktop connections to gigabits with desktop connections. Driving on to things like Wi-Fi. All of which mean that there is a network challenge that’s internal as well as external. But superficially our first answer to most customers is we fix your internet connectivity problems, that’s very simple and very obvious when that’s broken. The second part is, and again many of our customers who choose multi-cast recognize this in their issue, how do we fix our internal communication to make it efficient across the internal side of things. I don’t have numbers unfortunately for you that really nicely would answer your question but anecdotally, our customers tell us we fix both of these problems for them.
Phil Edholm: OK, that’s a good answer. It very much seems that this applies both to internal traffic that starts in the network, inside the corporation, and also external off of the internet. Regardless of whether the traffic is immediately required is a stream that’s a multicast to lots of folks, or if its cached. You kind of solve all those video problems simultaneously, it sounds like.
Alan MacLeod: Yeah that’s correct. So oftentimes it’s driven on the type of communications, so live from the cloud would require multicast and the obvious way to do that for a very large scale. But, equally you could use caching in other environments and we have the ability to very elegantly cache live video events as well very efficiently across the network. But typically when we’re talking about things like corporate training, oftentimes when it’s a mandated cause that everyone has to sit by a certain time, that’s going to be pushed onto the premise overnight. It’s like the off hours of electricity concept. We can push the video down overnight, have that available, when the users think they’re connecting to the internet to pull it down from the video platform, it actually is directly served from our cache directly to the user. So, they get a high-quality video experience but they feel they’re directly interacting with the video platform itself. So yeah, both of those two problems are real customer issues and yes we’ve helped them with both of them.
Phil Edholm: Excellent. Thanks – I appreciate the info. Back to you, Jim.
Jim Burton: Great, well thanks. Let’s move over to Kevin Kieller. Kevin, I know you have some questions.
Kevin Kieller: Thanks, Jim, and thanks for taking the time to kind of answer our questions, Alan. So I’m going to talk a little bit about kind of in the Microsoft ecosystem because I’ve certainly lived with the large enterprise customers that my team works with. Situations where organizations have to ask people to group into meeting rooms because they’re very concerned that if everybody just stays at their desk and has thousands of video streams, it’s going to be a problem.
So, obviously in the Microsoft ecosystem for live events you’ve got the older Skype meeting broadcast and Teams live meetings. Right now those can each scale to like 10,000 people and up to four-hour events. So, talk a little bit about how you, I guess, confirm for people that are listening that the Ramp solution works with both of those. And then maybe talk a little bit about, you know I go into my Teams admin center and is this hard to do? Easy to do? How do I set things up to have Ramp help me?
Alan MacLeod: Yes, so let me start at the end of that and go back to the beginning of your question, if you don’t mind. One thing that Microsoft continues to do really, really, well is make it easy. So typically configuring a Microsoft event is little more than deciding when’s the event, creating the actual event itself, and publishing that to your audience via an Outlook meeting or something like that.
So for us working with Microsoft there are three ways we go to market with them. There’s the Teams Live Meetings you described, and there’s the Skype Meeting broadcast again as you described. Both of those two use the Azure infrastructure that, directly coupled to those two business units, and they drive decent size scale event. Very much centered around the experience you have with those two platforms. So, Skype Meeting Broadcast is very Skype for Business-like. And Teams Live Meetings are very Teams-like.
The other play that Microsoft has going which I think they are telling us and we believe equally is the strategic long-term direction is Stream. And we’re seeing Stream really starting to change the industry. It’s serving two purposes: one is it’s a classic Microsoft foray into a new business. They’re bringing in as part of your Office 365 package. Which means that you’ve effectively already paid for it before you got started. It already you know has a lot of capabilities in terms of hosting lots of videos. But equally, its major advantage is it’s totally validating this marketplace. People really are demanding high quality video streaming. And so what we’re seeing from the Skype Meeting Broadcast Group, or Skype for Business Group and the Teams Group is they’re moving over to this platform as well.
So, with Teams today you have two ways you can have a live event: one is euphemously called Quick Start. And that’s Team to Teams. And in that scenario, we definitely support that with our caching solutions. When used, and typically this is for larger scale events, potentially with higher quality external encoders, they again have a euphemistic name of External Encoder. You’re guided to the answer with the question if you know what I mean from Microsoft. The External Encoder system goes via Stream.
And so with Stream, they have significantly greater scale capabilities there. And we’re able to work with Stream and deliver over multicast or over omni cache, live events as well as events that are video on demand. And again, Microsoft focuses on ease of use. One of the very nice things about their platform is that as soon as a live event is completed, it becomes a video on demand event and the same URL is applicable. So, if you’re one of the modern work force, like most of us, and getting to the live events on time can be a challenge due to the other pressures of work, you don’t have to worry about how do I find it after the event. It’s the same link, you go straight there.
And for us working with Microsoft, we’ll just move you to the right platform to give you the best possible performance. Whether it’s live or on demand, etc. So yeah, we absolutely work with Microsoft. We see Microsoft as a significant validator of where the industry is going as well as having a good strategic direction. And we’re finding a lot of customers are resonating with their messaging.
Kevin Kieller: That’s great information. Let me just follow up on one piece because I think Stream, which is effectively Microsoft’s internal answer to YouTube and also when you have a Teams meeting that’s where it gets recorded. I find a lot of organizations don’t even know sometimes that they have those licenses. But the ones that are using Stream and recording training videos and then Stream does some great automatic transcription. So, it’s not a live event but I’m recording training and I’m putting up an increasing number of videos from my organization up on Stream. I think you said this but I just want to get clarity. So even if it’s not a live event and it’s this on demand, is the Ramp platform going to figure out, like based on who’s viewing what and do some, optimize some caching and then pull down things that are in high demand? Talk a little bit about kind of the logic behind the scenes.
Alan MacLeod: Yeah, absolutely. We basically work two methods on our caching and all of it is transparent to the end user. So, for the end user, going to grab a video will start caching that automatically, working on the principal that if somebody else grabs, we’ll serve it from cache. And if nobody else grabs it, we’ll age it out based on either the time that Microsoft set for it or that the customer set for that video itself. Or because we have something come in which is a higher demand that multiple people are fetching. And so, in that respect we will optimistically cache as needed, or opportunistically I should say cache as needed. And thus, make sure we get the best possible performance. And of course the challenge is when somebody asks for something, it’s hard to know whether ten more people are going to ask for it immediately. So we have to take an opportunistic strategy there.
The second approach is one that we allow the IT people to drive. And so again, this goes to the idea that we know something’s going to happen. So, maybe some executive has done a pre-recorded meeting or pre-recorded message they want their staff to get the following morning or directly after lunch or something like that. We allow them to do that recording and then push that out to the caches across the whole corporate network. And again, so then the users just jump on they think they’re fetching it directly from Stream but in actual fact it’s coming from us and it’s completely transparent to them that they have no need to worry about it other than the high performance will feel like it’s a little better. It’ll be a little faster coming straight to them. So in that respect, yeah, we can fix both of those two problems.
And the most important thing, and again, Kevin, you and I have talked about this over the years on things like Skype for Business. It’s all about ease of use. The more transparent we are, the easier we are to use, the happier the customers are.
Kevin Kieller: No, that’s great detail and you’re absolutely right. Certainly the fact that it just works better to the end user, that’s the magic that we look for from technology. So thanks for that insight and that information. Back over to you, Jim.
Jim Burton: Great – thanks, Kevin. Next up is Michael Finneran. Michael is one of those people who is very deep in technology but also very concerned about user experience. So, it’ll be interesting to hear your questions, Michael. Over to you.
Michael Finneran: Thanks, Jim. Alan, let’s go with the practical first. When somebody buys your product, what exactly do they get? Is this a gadget, is it something that hooks onto the LAN? And related to that, is this going to be compatible with whatever LAN infrastructure the customer has in place? What am I buying here?
Alan MacLeod: So we sell software, first and foremost. Some customers want to run it directly on a given server. More predominantly we’re seeing now people running it inside virtual machines and deploying it directly in their data centers. By providing it in a software format, they can size, to some extent, the hardware it runs on. They can choose their vendor of preference so we don’t have to go through certifying hardware for them and put their own IT solutions through that problem as well. And so in that respect, it’s much easier providing software. Not the least of which is shipping is remarkably free across the internet.
So yeah, so that’s the first part. The second is in terms of compatibility. We run on Linux, we run on Windows, whatever your preferred typical operating system is going to be there. So that’s not a concern. And in terms of interacting with the network infrastructure, our caching solution of course it’s just packets moving backwards or forwards across the network so it looks just like normal IP.
For the multicast solution, multicast does take some configuration. But the reality is virtually all of the enterprise-quality networking equipment supports multicast. And so, that’s mostly a configuration exercise. Typically, when we’re talking to companies who are interested in multicast they’re typically larger though not always. They normally have a pretty sophisticated understanding of multicast. And nine times out of ten, they’re replacing an old Flash or a Silverlight-based multicast video solution. And they’re looking to how do I do that with modern platforms. And that’s where we come in to help them. So, we let them literally take the infrastructure they’ve already invested in and just apply a modern technology directly on top of it.
Michael Finneran: Great. Kevin Kieller asked you about compatibility with Microsoft Teams. Is this Teams exclusive or does it work with Webex? Zoom? We’ve got lots of video options today.
Alan MacLeod: Yes. So you’ve put me on the spot to try and remember all of the companies we partner with. Ramp supports dozens of platforms, and you can see a list on our website. We’re very heavily focused on streaming solutions rather than being explicitly focused on UC vendors who are expanding into more streaming live platforms. And so in the case of Microsoft with Teams and with Skype Meeting Broadcast, we do work with them. They actually, the back end looks very much like a streaming solution there.
And so, we partner with Microsoft, with IBM on their Watson Media Solution, with Kaltura, with companies like Brightcove, Intrado. I’m going to have to apologize to all of the other partners whose names aren’t coming to my mind as I need them. Typically, I think 90 percent, if not more of the streaming platforms we absolutely work with. With the UC vendors, of which Zoom is similar, they’re coming into this space. I had a few brief conversations with Zoom about their long-term plans. At the moment, one of the things that works very well for them is that they interface into platforms like Microsoft Stream. And then we use that.
Michael Finneran: You had mentioned the magic word before Alan, which is transparent. Does the user have to do anything different if they’re going to be participating in this video through your solution as opposed to just logging on as they normally would?
Alan MacLeod: No.
Michael Finneran: Good.
Alan MacLeod: I can expand on that if you’d like, but basically the answer is no.
Michael Finneran: Good. Well, that was my list, Jim, so back to you.
Jim Burton: Okay, thanks a lot, Michael. Now I’m going to turn it over to Francisco Kattan, who’s going to talk about things from the consultant level and what he would look for in helping his clients. So Francisco, over to you.
Francisco Kattan: Thank you, Jim. Yes, Alan I have a little bit of a higher-level question. Which is your go-to-market model. How do you get to market, do you typically sell directly to the enterprises, or do you go through a channel? And a second related question, a lot of the video conferencing systems, the way they handle these large company meetings or big live training events, is typically there’s only very few speakers in these kinds of meetings and there’s always a very large audience listening. So typically they just delay the video stream to support thousands and thousands of viewers. So how does your solution fit in? Is it a competing solution with what is already offered through the likes of Zoom, Google Hangouts Meet, or even video APIs such as Nexmo’s TokBox API? Or is your solution embedded in those kinds of solutions? Can you shed some light on that?
Alan MacLeod: Yes, so again let me go backwards through your questions starting with the last one and go back to the first one. So, obviously we’re better than any other solution that any other company could think of. So I just want to get that on the table up front. In all seriousness, the analogy I think best describes us is we’re kind of the roads and the video platform of the cars that go along the road. So, in that respect, a company like Zoom with a UC-type solution is not really competitive to us. They’re much more likely to partner with us. Whereas, a company like Zoom would be competitive with a Stream-type solution. Albeit they actually work well with that as well.
And so for us, we don’t really compete with any of these platforms and the thought process behind these platforms. What we do is enable them. And so, when we look at the streaming media platforms that we partner with, very much it’s about how do we make you more efficient in how you communicate? How do we make you more efficient communicating across the corporate infrastructure to get to all the end users?
And so, going back to the previous part of that which is the number of speakers, going to a larger part of the audience, and the immediacy of it… I would have said three or four years ago that the level of delay that you could put in there was fairly high and people were pretty tolerant. And high varies. So, high could be a matter of hours. Or it could be a matter of seconds. What we’re seeing is a desire to get that time down. And a lot of what’s driving that is a desire for interactive videos.
So this is not a UC-type solution where we’re all sitting around the table, a virtual table, and communicating as peers across the table. This is still very much the idea of there are a limited number of speakers and a very large amount of audience. But what we’re now starting to see are audience participation-type measures. Polls are being pushed out, slides are being flipped, and people have been asked to comment on slides; people making interactive comments about what’s going on to ask questions. So there’s a desire to drive some immediacy into that interaction. And in that, a lot of the streaming protocols that we’re using, which they’re not designed to be immediate. They are designed to leverage the fact that latency is actually a valuable tool if you want to drive high quality and high bandwidth, and you don’t need instantaneous feedback, we’re starting to see some evolutions in those.
So, both Apple and Microsoft have been working on low-latency streaming. So, we’ve seen that being driven through the Mpeg Dash Group, which is the working body that drives the Dash standards, which is one of the ways people stream video. And Apple of course owns HLS which is their own propriety solution which is now ratified and standardized and many, many, companies do that. And they’re both trying to work. How do we get the, effectively that latency increment, which historically has been in multiple seconds down to milliseconds or hundreds of milliseconds, to drive a much more immediate interaction.
And again, to me this is all part of what was a problem that was fixed in a very simplistic way originally is now becoming a fundamental business requirement and when things become fundamental business requirements, compromises are much less well tolerated. And you have to get very much down to the audience of today expects immediacy. I was reading an article on LinkedIn that was saying that people today complain about videos taking seconds to start on Netflix. Whereas only 20 years ago we were happy if we could dial into the internet in tens of seconds.
And I think we’re seeing something similar now with live streaming video in the sense that the expectation that I get on immediately, when the person is speaking, that I am within a few seconds of them speaking so my comment back doesn’t have to have huge amounts of context, I can be very, very, appropriate or specific. And it can be understood as it’s received back at the far end. So, immediacy is becoming an increasing requirement. And again, for us, some of that is transparent to us. We’re able to transmit it quite simply. Other parts of it will be working with our partners fine tuning and get that experience up.
To your first question, my last answer to your questions, in terms of go-to-market. We typically partner with our partners, which sounds a strange answer. Most customers are not buying an ECDN solution, which is what we are, as a separate acquisition. They’re typically looking at their video platform. Oftentimes it’s been Adobe-based and their desire to move away from Flash, particularly as it’s end-of-lifing next year is quite high. And so in that they’re looking for a video partner to help them solve that problem. And that’s the real business need. How do I solve my video problem? As a part of that solution, most of the video partners are saying,
“okay you need to think about how do we move traffic across your network?” And many customers already understand that up front. And so in that situation they’ll come to us and say, “hey, can you partner with us on this account?” And depending on the partner, either we will let the video platform lead the whole deal and we’ll sell through them. Or if the customer wants or the partner prefers, we’ll sell directly to the customer. But 99 percent of the time we work in collaboration with a partner for a deal.
Francisco Kattan: Thank you, Alan. Quick follow up from your first answer to my question. So, in these situations where there is a very large meeting with thousands of attendees, even though I think if I interpreted this correctly, even though typically the viewers are not going to be speaking, there is some interactivity that is required because they may be commenting on slides or chatting via the chat channel. And so your solution enables a reduction in the latency to make that interaction more engaging. Is that correct?
Alan MacLeod: So, yeah. So today we are seeing a drive to reduce that latency to allow that interaction. We’re only just starting to see the interaction become a customer requirement. If you go today and you ask for, for example we were talking about Stream earlier. So, Stream is trying to drive up that experience. And again, very much Microsoft way of thinking about the world. And so in that, they’re increasing the poll capabilities, the feedback capabilities. And we see that from other companies – Intrado, Kaltura all have the ability to allow the end users to watch the video and interact back to the platform.
As we get that latency down and again as I mentioned earlier the protocols themselves are inherent adder of latency. And they’ve done that for a lot of good reasons. But we’re seeing the standards to try to get to lower latency environments. And part of that we’ll enable. The problem we really, really fix for customers is not about latency, it’s about bandwidth. And so, as we fix the bandwidth problem, we don’t particularly change the latency issue. As the video platforms and it’s very much the way they encode the traffic that’s going to fix the latency problem, as they fix that by moving to these newer variants on the encoding techniques, we’ll enable that same low latency experience at the same time.
I think it’s like all things real time. You can’t take out latency. Once you’ve got it you’ve got it. So we have to get it down at the encoder level to get it down at the platform level, to get it down when it reaches ours. And at that point, we can start looking at sub second or maybe one-second two-second type latencies or lower. And that’s really… in these interaction environments where it’s one to many, a couple of seconds is normally perfectly fine. That’s normally going to be good enough for what people are trying to achieve. We’ll see. Constantly throughout this industry we’ve seen requirements go up to get things better.
Phil Edholm: Alan, this is Phil. So just one side comment. I don’t believe there’s a huge impact on latency because a lot of latency comes from areas that area outside of the control of this kind of an added system in today’s environment. But, one major impact of latency in video is the 30 frames per second. Which means essentially that you’re only sending 30 packets per second. So, if you went to a higher video rate, you could actually at 60 frames per second, potentially have a much higher quality video. But that would also double the bandwidth. So in that case again the caching environment or the multicasting might be very important as we go to higher frame rates. Which actually would give you, like I said, significantly better quality.
Alan MacLeod: Yes. So the streaming protocols HLS and Dash are so far away from frame rates as being part of the challenge. You can see this when you’re watching Netflix. This is the best example. That first part of the movie you’re watching, that first five or six seconds, seems quite low res. And then it gets quickly much, much, much better. Then all of a sudden it will drop off and go horrible for a second. And that’s just when another member of your household has started playing a videogame and taking all of your video bandwidth. And so what we see is that segment size, that four- or five-second segment size as being the big latency measure. And so in that, you have encoded it probably adds 20 to 30 seconds of delay from the time someone says something into the camera to it being processed, turned into mp4, then chunked into segments and then passed on the platform to distribute out to us.
By the time you’ve gone through all that you’ve lost so much time already. And a lot of that’s because these segments are quite long. So, in the same way, exactly the point you’re making earlier on about voice. If you’re going to send a slice of voice that’s 20 milliseconds long, it takes 20 milliseconds to get that slice together. And so in video we’re sending typically a six-second segment, it takes six seconds to get it together.
So, in today’s protocols, six seconds is a given latency and it only goes up from there. And most of that up is in the encoder. These new low latency variants are saying, “okay, let’s stick with that six seconds because that works very well for us across the internet and it’s very robust for video on demand.” But for live, can we send little components of that, little fragments or chunks or partial segments depending on which of the two standards you’re following? And so those partial segments are now trying to get down to 500 milliseconds, 200 milliseconds. So again, still very large compared to frame rates.
Phil Edholm: Right, and compared to real time traffic.
Alan MacLeod: Absolutely.
Phil Edholm: One of the things that’s really important is, for the folks that are looking at this from the video conferencing real-time perspective, when you talk about latency, you’re talking about very different level of latency that is impacting these type of interactions.
Alan MacLeod: Right. And yeah it’s not, “you finished talking now it’s my turn to talk.” It’s not that level of latency at all that we’re worried about. It’s more about I’m describing sales are going to be like an Asia-Pac and you have a comment about Asia-Pac sales and your comment comes in 30 seconds later because that’s how much latency there was. And I’m off talking about something totally different and it’s out of context. The goal is to try and get down to I’m talking about A-Pac sales and within three or four seconds of me making the comment about A-Pac sales your valid comment comes in. And it’s in context based on time.
Or it’s maybe a case of, “okay guys, we’re going to enforce red shoe wearing on your left foot on Thursdays. What do you think about that? Let’s have a quick poll.” And trying to get out to the audience should we be wearing red shoes on our left foot on Thursdays and getting a quick poll response from the audience. Again, when you’ve got 30 seconds of latency, that’s an inordinate amount of time for everyone to be sitting around twiddling their thumbs waiting for people to answer the poll. But when you’ve got two or three seconds, that’s a completely reasonable time. Most of us can fill in that time with some small talk.
So the other problems we’re trying to fix, it’s really quite different to interactive video. Even though we’re trying to get to semi real time, even though we’re trying to have human interaction, it’s not about trying to get it down to I finish speaking and you seamlessly step in.
Phil Edholm: I think what you’re saying is that the events you’re focused on are not laissez-faire video conferences but rather structured and moderated events that have a structure and have different experiences potentially for a presenter or a moderator versus a participant. So it’s those kinds of events, which there are a lot of those in the corporate world.
Alan MacLeod: Yes, it’s town halls, there’s actually a report out, it’s stakeholder meetings or shareholder meetings. It’s those sorts of things where it’s very much a report out without a high expectation of feedback in real time. But the desire for feedback in real time is going up. I think, if the world is teaching us nothing at the moment it’s that we all have a voice and we will all be heard no matter what we have to say.
Phil Edholm: Cool thanks.
Jim Burton: Well this has been great, Alan. Thank you very much for your time today. It’s been very interesting and very enlightening and with this podcast we will post more information about the products we’ve just discussed to make it easier for anyone that wants to dig deep into the technology and learn more about Ramp. So Alan, thanks much.
Alan MacLeod: Thank you all. Thank you very much. I really appreciate it.