media-3

Transcriber:	Bobbe Allen (bomallen @ cc.usu.edu)
Brief Bio:
Date finished:	November 1, 2004
Proofreader:	Nety Chen
Brief Bio:
Date finished:	June 9, 2005

Hi, my name is Bob Metcalf and I’m here to talk today to talk about the next big thing, the video internet.  Ah, now I’m not sure of what kind of video we’re talking about and I’m not sure if it will actually be the internet when we are all done but nonetheless I am going to call this the next big thing, the video internet.

And one of  the questions which I hope to get to in a few minutes   is whether in driving the proliferation of the video internet whether that will be driven by communication, by or content or commerce.  Let’s begin with a short history of networking.

In the beginning there was mail, the US mail, say, and it is still with us and what came later in the 19th century was telex, the carrying of mail, certain forms of mail electronically to the telex network and then telex was generally overtaken by a new electronic network called the telephone network, which of course, is still with us.  The telephone being, as you well know, a two way communication system generally paid for by subscriptions and its principle use being communication among people. The next network after that was the Television network which arose during my life time. Ah, and television as you know is a one way and generally entertainment oriented and paid for through advertising, although increasingly by subscriptions now with Cable television.  And finally the current big thing, the internet which came to us starting in 1969.  Now the internet itself has been a series of applications since 1969.  The people who first started building the internet ah thought that it would be used with  dumb terminals logging in to remote computers across the country and that was the first application built to run on the early internet ah and shortly after that it was realized that in order to run remote computer applications in large computers across the country you would have to transfer your data there, you would have to transfer files, so the file transfer protocol arose. And then shortly after that a certain kind of file became very important and that file was the electronic message.  And so suddenly, in about 1971, we had email and then email was pretty much the major use of the internet for the next 20 years.  Now during that time ah a new kind of structure arose called newsgroups that is where computers on the net would be used to collect emails and keep them so that ah other people might view a conversation conducted in email and these would ah be categorized by topics. So remote log in and file transfer and  electronic mail and newsgroups were pretty much what the internet  was used for until the 1990’s and starting in about 1990 we learned about the world wide web. And the first form of that in the early 90’s was the arrival of web publishing, the use of the internet for publishing information and ah A stunning conclusion from that arrival was the recognition among journalists, publishers that there was a new kind of content suddenly.  Prior to web publishing, generally speaking, there were two kinds of content.  There was editorial and there was advertising. And the whole publishing industry was very careful, (some were very careful) to keep those two separate in how they would be managed.  It was called the separation of church and state.  But when we began publishing on the internet we discovered a new kind of content which I’ll call community content, that is content which was not produced by the editors of the publication and not produced by the advertisers of the publication but produced by the readers of the publications.  So now on the internet I think there are three kinds of content and they are all important and community content is the new one.  Toward the mid 90’s the web then went from publishing into commerce and what we saw there was the dot com bubble which has only recently burst, not that web commerce is going away but that there has been a recent correction from which we will soon recover.  So what’s the next big thing after the dot com bubble has burst, after web commerce and mind you, email is not going away and FTP is not going away and web publishing is not going away and web commerce is not going away.  They are all with us like the US mail.  But what’s the next big thing, the next thing that will drive the proliferation of the internet, electronic networking and I think it’s the convergence of the web with the pre-existing voice and video networks, the convergence of the internet, the telephone and the television networks.  So let’s talk about convergence, the convergence of the internet, telephone, and television networks.  This is not an entirely new process, it has been going on for some years. In fact we have some information which I am about to share with you about the recent history of convergence.  So we have these three communities of interest:  we have the internet people, my people, we have the telephone people and we have the television people all of whom saw convergence, the notion that the internet, and telephone, and the television networks and telephone networks might converge and they all approached this convergence with certain goals in mind and what I’d like to now summarize is, my view of what they wanted when they approached convergence and what recently they got which is quite different from what they wanted.

Let’s begin with the internet people.  The internet people wanted convergence with the telephone and later television networks because they wanted ubiquity.  They had built this wonderful thing, the internet, and they believed, as I do, that we wanted everyone to have it, to use it and convergence was a way of getting there because the telephone network is ubiquitous and the television network is ubiquitous so if only the internet could be put on the telephone and television networks, this great thing that we, internet people, brought to the world will then it would be available to everyone.  What a great goal.  And so we began moving the internet initially onto the telephone network and in fact today most people today, more than 90% of the people who use the internet, use it through a dial up telephone modem.  But the internet people got something different, something  unexpected that they didn’t quite want.  For example when you let a lot of people log onto the network who are not university Phd’s is you get what is called newbies.  These are people who weren’t brought up properly in the ways of the internet.  They had different interests and different values when they came and they didn’t know how to use the internet all that well.  This was infuriating to the internet intelligentsia.  The second thing they got was slows.  The telephone network was not designed to carry high speed data.  It was designed to carry voice communications. And so all these people who were accessing the internet through their dial up telephone modems are doing so slowly and this is really painful and so we are all in a really big rush to solve that problem.  The telephone people saw great promise in convergence too.  They were kind of interested in convergence because what they heard was the internet is a world wide network and people use it a lot so the telephone people said we’re going to and get a lot of long distance minutes out of this.  People are going to be dialing into that internet around the world and we’re going to bill them minute by minute for there their use of the internet and what a great new business opportunity this is they thought.  What they got, the telephone people, they got something completely different.  They got a lot of short distance hours.  That is instead of long distance minutes they got short distance hours.  People, instead of dialing around the world to access the internet, just dialed an internet access port in their hometown on the same exchange, and instead of being on there for a minute or two or three which is how the telephone network was designed, they use it for hours, if not twenty four hours a day.  The telephone company’s companies were technically unprepared to take on this load and they had no way of billing for it because for years and years and years they had been organized around billing for long distance minutes not short distance hours.  so this was a bit of a jolt and surprise and it made them angry about this convergence. Then something even worse happened  people started talking about using the internet to carry voice conversations for free and the telephone companies in the united states generate about $200 billion dollars a year in carrying long distance telephone conversations and the thought that these conversations might be carried for free over the internet was quite shocking to them.  Now of course free is a very low price and they are not likely to be free but the threat of voice over the internet has been very motivating to the telephone companies.  They are not sure they want to do convergence all that much anymore.  The television community, the television networking people also saw promise in convergence. For years and years and years  the telephone industry has been trying to branch out into interactive television and there have been many trials going back 10 and 20 and 30 years, 40 years perhaps to develop two way or interactive television and they saw in the internet which is two way interactive a chance to finally get at the many benefits that they saw in the interactive television so the television industry began  approaching the internet and convergence and they got something quite different from what they were expecting.  On the internet the tradition is that content and carriage are separate that is there is the infrastructure of the network which is owned and operated and run by one group of people and then there is all this content which is carried on it and they are generally not owned by the same people, well in the television world there is a great overlap between those who own the carriage and those who own the content.  So the business model of the television industry is different from the fundamental business model or evolving business models of the internet.  So this was a problem and the television people were roundly criticized for linking content and carriage in their various schemes for driving revenue from the internet.  It was even a worse development, however for the television people because television people for the most part are used to having five channels or ten channels or twenty channels or a restricted number of channels in which the attention of their viewers is channeled and so the competition among these channels is under their control pretty much but on the internet there is not ten channels or 100 channels or even 500 channels, on the internet there is a million channels.Every user of the internet can be a generator of content and the competition among users of the internet that is providers of content in the  internet, the competition among them is much more fierce because the real scarce, the real scarcity on the internet is not broadcast frequencies and its not channels on the network it’s the attention of readers, and so the TV people confronted a business model that was alien to their experience.  But still television has been the convergence of the internet and television has been the most effective to date.  Although when I say that you have to understand there are three kinds of television internet convergence of increasing importance. The first kind of television internet convergence is the use of televisions as cheap computer monitors and you saw this, and you see this with such systems as web tv where really what you are doing is using the telephone network to connect to the internet but you are using the television as the monitor because televisions   are generally cheaper than computer monitors.  This is really low grade convergence between tv and the internet and it hardly counts.  The second more vital convergence is the use of cable modems to carry the internet over the television network and it turns out that cable modems have been the most effective, the most widely deployed method of delivering internet services at higher speeds to consumers. But still cable modems  is a weaker form of convergence between the television and internet. The real convergence is when you can carry television over the internet which is I believe the intent of what you are watching now. This is suppose to be coming to you over the internet so that’s the convergence that I think we are after:  The ability to carry the video over the internet. The trouble is the standard for carrying  video over the internet is what I would call the jerky postage stamp and you have all seen them when you have tried to watch video, perhaps you are watching me now as a jerky postage stamp  A little tiny little square on your screen with me going like this  of course you can’t tell that I am doing this because I am probably already doing this in front of you right now.  So I claim that convergence  between television and the internet is the next big thing.  But in saying so there is some serious work ahead.  There’s work ahead in the proliferation of the video internet and among the work to be done is the deployment of what is called broadband to homes.  As I mentioned earlier, most people access the internet through a dial up telephone modem at a pitiful 56 kbs per second or less but in order to put video on the internet we are going to have to carry bits quite a bit faster than that.  A funny problem has emerged.  We have lots of bandwidth in the long haul on the internet thanks to amazing technologies like wave division multiplexing where we are carrying now approaching a terabit per second on a single fiber.  We have plenty of bandwidth in the long distance and thanks to local area networks like my own baby Ethernet, we have plenty of bandwidth in the local area, that is within the inside of buildings but there is a gap there’s a gap between the LAN  (local area network) and the WAN, (wide area network) and we’ll call that access bandwidth, getting from the LAN to the WAN.   And there is a chasm currently bridged for most people by this low speed dial up telephone  modem.  So part of the work is getting that chasm bridged and there is a variety of ways of doing it, some better than others.  Most of the so called broadband methods of access many of them are through an old fashioned system that is called ISDN and believe me it would do you no good for me to explain what ISDN actually means, suffice it to say it’s a way of accessing, going from the LAN to the WAN at up to about 128 kilobits per second which isn’t much faster than a 56 kbs.  And it’s a sad comment in our time that most market analysts include ISDN under the rubric broadband when really it’s much too slow to be considered broadband.  The next method of bridging the bandwidth chasm here is DSL (digital subscriber lines) that is using the copper that was installed for the telephone network to carry data at high speeds at 384 kbs per second at 1.5 mbs per second - DSL.  And DSL is being deployed, over, now, as we speak, over the copper network by a variety of players, including, mostly, the telephone copper monopolies. Another method of bridging the chasm is cable modems as I mentioned earlier or the use of the cable television distribution network which reaches about 65 million homes in the United States out of 110 million and cable modems are the leading method of bridging the chasm of the local area to wide area.  And then there is some other technologies which we hope will come along to add to the alternatives and create more competition in solving that problem.  Now, in 2001, the cable modems are winning.  That is there are more homes connected at higher speeds through cable modems than through digital subscriber lines than through DSL and that can be explained very simply the cable monopolies have weaker monopolies than the telephone monopolies.  The telephone companies are huge, and they have air-tight monopolies on their geographies so their motivation for installing digital subscriber lines is much lower.  They have no one pressuring them, very few people pressuring to do so.  An industry did rise up in recent years to bring DSL to homes in large numbers in competition with the entrenched, incumbent local exchange carriers as they are called. But most of that competition has been throttled through the use of the monopoly power of the telephone companies.  The cable modems on the other hand have been proliferated by the cable companies, which are themselves monopolies, but they are much weaker monopolies they have much more competition from say, satellite for example and plus they are more entrepreneurial and energetic and they have been proliferating these cable modems and we should be rooting them on as the only way of stirring the telephone companies to movement.  So we have had this recent bad news with the bursting of the dot com bubble that is the telephone monopolies  are more entrenched than ever in providing broadband and they will be slow to provide broadband, but the broadband is  being deployed and they are and they will eventually get around to it and eventually broadband will begin to arrive in further and further  and more and more places and faster and faster speeds and in that context we need to judge where to put video, what applications of video over what populations, over what businesses, small, large businesses, homes and so on are a very complicated strategy problem.  So one of the problems, or another of the problems is the proliferation of the internet video is the plumbing and I am a plumber, an internet plumber, so let me talk a bit about the problems presented by the internets plumbing.  The term to describe the heart and soul of the internet plumbing is TCPIP which is the name of the protocols that have been, I guess they were installed in the internet in 1983, they were probably invented in 1973 and these protocols present problems for  the carriage of video over the internet.  When you, in today’s world,  when you try to carry video over the internet, its, the plumbing is slow, and its lossy, that is information is lost and it’s delayed, jittery, jittery, and delayed when it gets there.  And these present problems for television.  Television is much higher, uses many more bits per second, than telephone or the web.  It relies on the data getting through lively so the whole picture gets painted not just little corners of the picture.  And it’s real time, you know moving in real time and if the packets are delayed and I stop momentarily that’s disruptive to what you expect from video.  Now there have been some efforts to solve the problems of the internet as plumbing for the video and one of those is broadband deployment that I’ve just touched on that is increasing the speed of all the circuits so they can better carry video.  But then there are protocol problems with the internet.  The internet was never designed, it was designed in 1973, I know, I attended the meetings and no one was talking about video in 1973 and not even in 1983 when it was finally deployed in the world wide internet.  But recently as we are trying to get video on the internet there have been various proposals to fix TCPIP.  One of those proposals is to replace TCP with various protocols one that comes to mind is RTP the real time protocol to make up for some of the deficiencies of TCP.  Another solution has been the development of what’s called MPLS (multiple protocol labeling system) actually I’m not sure that’s what it means. But that’s a lower level fix of the plumbing whose purpose is to speed the packets through the network to reduce the losses that they encounter and to get them  through in a shorter time and with less jitter.  While all of this is going on there have been great advances in the plumbing that underlies TCPIP.  We’ve seen the development of wave division multiplexing which is the use of optical fibers  to  carry on different colors of light carrying very high speed data.  So you might have an optical fiber with a thousand different,  we’re not quite there yet, but very shortly a thousand different lights each carrying ten or twenty or thirty or forty gigabits per second so we’re talking about fibers carrying terabits per second and those speeds are very promising for carrying video for example are much larger than 1.5 gigabits per second which is how many bits come out of a tv carrying high speed definition television camera.  So underneath it all, underneath TCPIP is the development of this lambda switching network, the ability to carry lamdas that is wave lengths of light over these fibers  very high speed and at very low cost.  So what is likely to happen, and I’m about to say something very controversial, what’s about to happen is that the internet which is trying to fix the TCPIP plumbing up here, to better carry video, is going to be overlaid, or I guess the right word would be under laid by a lambdas switching network which is not TCPIP based and that

22.02 these two networks won’t fight it out somehow, but they will be layered.  Now we see some of that even today, I mean there is  precedent for this,  that is on the cable networks when you use cable modems.  The cable modems carry the TCPIP network on the same cable that television is carried.  That is there is already an overlay underlay  structure in the cable television network. I’m thinking that in order for us to get to wide scale high definition video on the internet we are going to have to do a similar thing with an underlying lambda switching network in the overlying TCPIP internet network.   On top of this plumbing, problematic plumbing for the video internet, are what you might call the applications, video applications.  And I talk about video as if  I know what those applications are but I don’t.  And no one does so we’re talking about various uses of the network that involves moving pictures and sound like you are watching now and the big question is I mean as this as the infrastructure is being deployed, is what will these applications be, what will drive deployment and how will this video be used for actual delivering valuable services. So there is a growing list of forms of video.  The big question is will one of these forms or yet an undiscovered form of video rise up to be what we call in the computer industry the killer app for the video internet.  One long sought use of the networks for video is called video on demand.  It used to be that you would go to the movies to see a movie and then it got to be that you would walk down to the video rental store to get a video, or you’d watch it on pay per view or cable television there’s some there’s some notion that people want to see video when they want to see it and they don’t want to wait.  So being able to select a video and watch it right then when you want to watch it, it’s called video on demand.  So will we be using the internet for video on demand?  Will that would be the killer app?  Will millions of people wanting to watch this video right when they want to watch it?  Will that demand drive the proliferation of video internet? I don’t know.  In fact I have a theory that it won’t be. Another, and let’s call it an opposite kind of use of video, would be the video phone.  For decades we have been trying to get to the video, right now  telephone as you well know is just voice and wouldn’t it be nice to be able to see the person you were talking to.  I think so.  But video phone has been very hard to achieve.  It’s either been too expensive or too crummy.  Can we use the internet to finally introduce video phones?  Now this is different from video on demand.  Video on demand has to do with the delivery of what they call content whereas the video phone is a deliveer of person-to-person and communication these are not nearly the same but they both involve video.  Then there is video conferencing and there is a lot of video conferencing going on in the world.  You know, you go to a special room with a television and a camera and then groups of people can meet at great distances.

Perhaps video conferencing moved over on to the video internet will be the killer app.  And then there is email. Now email for twenty years and probably still is the killer app for the internet.  What about video email? What about instead of typing in your message, your short message, what if you just turned on the camera briefly to send a video email?  So that was a very short lesson.  I think there will be many nuances of video, and one of those or two of those will be the killer app driving the proliferation of the video internet. I don’t think we know what it is.  Now there are some people already starting businesses based on internet video.  I’m on the board of a one small company called AVISTAR and based on the state of the infrastructure today they are focused on delivering in essence video conferencing for companies, enterprises.  Of course they choose companies because homes don’t have broadband sufficient to carry video whereas big companies do so there’s infrastructure there.  Avistar has also decided that human beings have seen enough television that they won’t really appreciate video until it’s at least of television quality not jerky postage stamps. So their whole system is focused around delivering at least television quality if not high definition later television quality of a certain standard that is normal television and up not something down.  But after that it becomes unclear as to what the, the real applications are.  As this little company introduces new features, they find that some features don’t affect whether people use their network for desk to desk video conferencing and other features have a dramatic effect on usage.  My theory is that in this case this is a communication use of video it’s the and I have invented a word for this, if you’ll excuse me, the consumateability when you make a video call what is the probability that you get the job done?  Voice mail, for example, increases the consumatability of a telephone call when you call someone and you get no answer you have not consummated your telephone call but if you get an answering machine well then you might be able to consummate the purpose of the call even if they are not there.  I think what we are finding in the video world is that the consumateability of video calls is important. If you try to make a video call and the person is not there and if you have the opportunity to leave a video mail instead that increases the consumateability of the call, and therefore its usage.  So the picture I would like to leave you with on the emergence of the video internet is one of a ten or twenty year process unfolding in unpredictable ways with a lot of experimentation and discovery going on in the meantime.  And here are some of the phases that I anticipate will occur in this proliferation.  Often in the proliferation of a new technology or a new infrastructure there is an arbitrage factor, that is the old way is expensive for artificial reasons (for example regulation of the telephone network) and the new technology arrives without the burden of that regulation and is able to just be cheaper, better, faster by version of arbitrage against the tariffs that are not really related to costs or underlying factors.  The arbitrage phase if is frequently followed thereafter by actual cost savings that is the old way is more expensive than the new way intrinsically and so the new way begins to dominate.  Another phase, often the third phase is where the new technology brings with it new applications that were previously not possible with the old technology and that’s attractive to people to do entirely new things.  Not to do the old things better, cheaper, faster but to do new things.  Then after that you begin to see the convergence of the old and the new, the integration of the old and the new.  So for example video, pure video, like you’re watching now is probably not the right answer ultimately.  It will probably be the connection of video with powerpoint, video with telephone, video with conferencing that is an integration across applications that were previously separate. 

And then other factors which pace the proliferation of the infrastructure has to do with the making of standards.  These technologies must be standardized over time.  That is communication and content delivery infrastructures require standardization because they require vast investments in infrastructure and no one is willing to make infrastructural investments without standards.  If they are willing, they are unwise.  Because when you invest a lot of money you want to be sure to have standards so that you have multiple suppliers that you can set against each other in a competitive environment.

A question which arises in this unfolding is, as I mentioned, which will be the killer apps and we have these broad categories that we’ve seen emerge on the prevideo internet.  We had the emergence as I mentioned of the publishing web, web publishing where that was content delivery in a new way.  Then we had communication.  And in the history of the internet, communication has been the more vital use rather than delivery of content.  In the early days of the internet there was Compuserve and Prodigy that were content oriented.  And along came America On Line which was very communication oriented and we know which ones won.  That is communication proved to be the, at least in the short term, the more desirable use, the more driving use.  And then came commerce. We had communication, content, communication and commerce and we can wonder and guess whether that same evolution will recapitulate itself in the video internet.  Now video has in the past been driven by entertainment television is mostly entertainment and its an advertising business model for the most part although increasingly there are subscriptions but I think its best to think of it as an advertising entertainment driven medium.  How will that be affected by integration with the internet. I think internet, entertainment will drive adoptance of the, adoption of the video internet and that has some good, will have some good implications for education, we talk about education on the internet and the use of the internet in education and I think as video and entertainment get into the internet this will, bodes well for education.  Cause right now education competes with entertainment on an uneven footing.  The entertainment model allows for high production values which are attractive to the a, to the kids, whereas the the education business is not, does not have the same access to production values so there is kind of a, I mean thank goodness for the discovery channel which does have high production value so there you see an example where if you raise the production values of education it can compete on an even footing with entertainment  So as we get video on the internet, and the general facilities for production values on the internet as the infrastructure improves for delivering those production values is, I think this will have a related value benefit to the use of the internet for education,.   It’s getting to the point now where we can begin to think not so much of putting the internet in the schools, but instead using the internet instead of schools that is if we begin to have a very high bandwidth and ubiquitous internet with video more and more it can be used for education from the home instead of busing people off to remote places to be taught with chalk on slate.

I’d like to close with a short story of a recent interaction that I had with video over the internet only this wasn’t just normal video this was high definition television, a coming thing in video and this wasn’t just the internet, this was internet 2, the new version  much higher speed version of the internet people are working on and I was asked to do a short video presentation not unlike this one now from here, from MIT to be delivered on a huge screen at a conference center in Las Vegas in fact it was part of a keynote address from of Craig BarrEt from INTEL so I arrived at this room over here at MIT across campus where there was a high definition camera and out of that camera came 1.5 gigabytes a billion bits per second of video, a much higher quality version of me that went into these boxes there were people sitting around by the boxes running them and those boxes would take this 1.5 gigabits and crunch it down compress it down to 20 megabits per second that was then carried over Ethernet, a hundred megabit per second Ethernet cable into Internet 2 there in Cambridge, here in Cambridge and then it went several hops through the southern United States to arrive in Las Vegas at 2.4 gigabits per second.  That’s the speed of the circuits being used.  So I showed up to give this presentation.  It was very exciting because this was the latest technology, the best of video, and the best of the internet assembled for this short video presentation and I delivered the presentation and listened to what was going on around me.  Unfortunately my presentation fell at a time of day when the traffic on Internet 2 was a bit higher than usual and so the broadcast wasn’t getting through reliably to Las Vegas.  There was a bit of breaking up, as of course, in the internet, the way the internet works using TCPIP, when packets accidentally arrive in large numbers at a single place, some of them get lost.  They are thrown away.  And this led to break ups in the video.  Fortunately the video had been recorded in Las Vegas just in case the internet didn’t work during this particularly busy time of day.  So I’m thinking, oh oh, is this good plumbing for the video internet?  It didn’t seem to be good plumbing.  But there was a worse problem.  Even when the internet wasn’t crowded, and  therefore failing to reliably deliver this video stream, there was another problem.  is that it took time to compress this video from the 1.5 gigs down to the 20 megs and it took time for it to wind its way through the internet routers that we have known so well over the last 30 years to get all the way to Las Vegas so then in the end it took 4 seconds for my picture to leave Cambridge and arrive in Las Vegas.  Now it was a pretty picture to be sure, but it was not an interactive picture.  Craig Barret in Las Vegas and I could not really have much of a conversation because there is a 4 second delay (that would be 4 seconds this way and then 4 seconds back) that would be an 8 second round trip delay.  It is generally believed that in order for that to work you need round trip delays in the milliseconds,  hundred milliseconds or two hundred milliseconds not 8 seconds.  So I took this little interaction as a sort of a progress milestone in the proliferation of internet video.  And my conclusion, which you have already heard a bit of, is that the internet is not the right plumbing for the video internet.  We will call it the video internet, we will call it the internet, but it’s not going to be the internet we know and love today, I suspect.  That the time tested TCPIP structure of the internet will have to be changed.  And as I mentioned earlier, perhaps in this overlay fashion underlying dense wave division optical switch network underlying the TCPIP packet switch network in order to deliver high definition video and I also saw that this a one way application was not nearly as, it would have been much more interesting I think to the audience if Craig Barret and I were to argue about something but we couldn’t argue about something instead the audience had to suffer a one way stream of my words instead of interaction between Craig and I.  So that the applications for video as I mentioned, will be considerable, exciting experimentation ahead as to what kind of video based applications or better yet where to put video in other applications to make them interesting and to compel the deployment of video infrastructure. so in closing its gonna take time, that little experiment between Cambridge and Las Vegas indicated the video internet is not, it may be the next big things but it is not going to come tomorrow its gonna take, as I mentioned earlier, probably ten to twenty years and these years will be very interesting and I look forward to them.  Thank you