Flutterby™! : A bet 2001-01-19 19:25:18+00

A bet

2001-01-19 19:25:18+00 by Dan Lyke 11 comments

Public notice: I (Dan) bet Todd one bottle of scotch of the winner's choice that on January 1, 2020, the majority (50%+1) of consumers will be getting their entertainment content through some sort of broadcast type network rather than point-to-point network. My sense of security in making this bet comes from the vested interest of the entertainment businesses in keeping their product on a producer-consumer basis, and that the O(N²) basis of network expansion isn't economically viable if there are substantial peak times ("Prime time"). My worry on this, and something Todd and I will have to refine, would be the existence of neighborhood caching systems that work like the Tivo except on a slightly broader scale, so that one could choose to see "this week's 'Friends'" downloaded from a neighborhood server. Todd, you wanna refine this in the comments?

I think ya'll are going to have to work on the specs a bit more. I don't think it will be either/or. I think that in 2020 you will have 50%+1 getting entertainment from both broadcast and peer-to-peer networks. Right now I watch television network programming and rent videos. Why would that change twenty years from now? The only difference is that I'll rent the video off a server instead of having to drive to the video rental store. I still see a big draw for broadcast programming, also. There's something communal and comforting about scheduled broadcast. Even though I'm all alone watching the currently popular video cocaine I'm doing so while millions of others are doing the same thing. We're in this together. ;-)

right - the premise is bogus. "Getting their entertainment content"? Where do we get our entertainment content from today? Newspapers, magazines, books = point to point. Movies = distributed multicast (individual prints sent point to point to the projection room, 300 viewers see that copy at once). TV, radio = broadcast. Internet = mostly p2p unless you have multicast connectivity. In 2020 you can bet that people will "get their entertainment content" from a mix of media and topologies, maybe not the identical mix to today's, but some kind of mix. Also, you should make it 2019, i.e. 18 years from now, because that's how long the Macallan ages. :)

Well, my call would probably be for a Ardbeg 17 or another of those dark peaty Islay scotches, although my tastes are mellowing a bit toward some of the subtler sherry aged ones. Todd's a fan of the Glenfiddich Reserve. I'm actualy cocky enough to think that we don't need much refining. What portion of the entertainment market television doesn't own, radio (in the form of people who leave the radio on at work) does. Books aren't even a possum in the road, unless those average American adults who buy 3 books a year are taking person-months to read 'em (and that means that for every me there's 35-45 who aren't buying any), movies are a factor only in the effects of video, the 7 billion or so a year that box offices take is actually slightly less than book grosses, and as Larry pointed out they're essentially broadcast. My prediction comes from two places, actually: First, for all of the hype of "good shows", when Phil and I were investigating better TV social trends, looking at ways to build a better TV and enhance the TV viewing experience, we talked to a lot of people, and most TV watchers watch TV, not a specific show. Second, networks don't scale well, O(N²) is not a good way to deliver entertainment cheaply, and with the mergers between the network delivery folks and the entertainment providers there's all the incentive in the world to not deliver video that's truly on demand.

Yuppers it's an interesting bet(amazing what happens when you play hookey from the churchillclub). It was Dan's idea to put it up here on Flutterby and this is a very cool way to record a bet (especially such a long one). We started with Woz style language "within my lifetime etc etc" and I think we have it nearly enforceable. #1 It's by january 1st 2020 (New Years at my place?) #2 I believe we were refering to electronic(generally streaming) type of entertainment (movies,TV,counter strike) but the lines get a little blurry already and may get worse in the future. Do we need to define this? #3 The bet (in my eyes) is more of a cultural adoption. The technology will be there a la Glider's Law (bandwidth will increase 3 times faster than CPU power (doubling every 18 months Moore's law)). The infrastructure rollout will most likely happen with fixed wireless(cheap and easy). so I think O(N2) doesn't hold water anymore (certainly not 19 years from now) but will the people adopt this change??? #4 I don't think we defined the consumer base. I'd kinda like it to be the United States or at least 1st world countries. If it's the entire planet Idunno...

If we are willing to (re)define video rental/purchase, where a physical video is in my hands as I walk from Tower to my apartment, as "essentially broadcast"(!) then the bet might as well be declared already won since we have loosened the meaning of "broadcast" to include practically anything in use now or in 2020. To take a specific future example, if the viewer of 2020 wants to watch an oldie like MATRIX VII: THE REVENGE OF THE KID WITH THE SPOON, and they download it PPV over shared broadband to an in-set cache, as I think likely, we can surely shoehorn this, too, into the "essentially broadcast" category, because it's no different from renting my video today except it comes through the coax or fiber instead of the mitts of some point-of-sale slacker. I still think that "get their entertainment content" is a slippery premise that's likely to allow both sides to claim victory on judgment day. If this is really about some overly specific hardon for particular technologies and protocols, e.g. wireless, then I submit that in 2020 it's going to be superseded by developments we haven't thought of yet.

Oh, no, an individual download is the same thing as a video rental, which is not broadcast. Going to a movie theatre is broadcast because it's a shared experience, and the limitation is you can't send everybody an individual print and the technology to watch it. My argument goes further because I claim that in truth people don't really want

that capability. We do need some language clarifying this, because the spirit of the bet is that I believe that all of these people talking about broadband and unlimited bandwidth are full of hoohey, and that we will neither have the technology nor the social change to have "video on demand". Instead we will continue to have "shows" which, because of limitations both of technology and consumer desire, will have to be presented to large audiences at the same time.

Cable companies. Consider this: cable modems are the fastest expanding broadband network service providers (at least in the southeast). I predict that TV will be closer to a blend of web and streaming video - kind of like a CNN with HREFs. But the real issue here isn't the nature of the technology - it's the nature of the consumers. I agree with Dan because I think that the majority of Americans actually PREFER broadcast, spoon-fed entertainment. Even if Bubba Joe is watching CNN with HREFs, he'll probably be too busy cleaning his shotgun to actually click a link.

This is a mixture of consumer demands and technology. On the one hand I'm hearing Dan state that unlimited bandwidth is "hooey" while on the other hand I'm hearing that the demand won't be there for truely unlimited bandwidth. If technology gives us just enough bandwidth to meet our needs with a little left over for the occassional surge of demand, how does that differ, in actual use, from unlimited bandwidth? Sure, Dan, we will continue to have "shows". People like shows. People also like video rentals. People will continue to show and interest in broadcast entertainment along with a lesser interest, but still an interest, in video on demand. I think what needs to be done is to pick a ratio instead of just "either/or" on the bet. Oh, Eric, in 2020 "Bubba" will probably be watching a turkey hunt video while he's cleaning his shotgun and he will click on the link to purchase the newest turkey call.

Movies are not "broadcast" from a topological or systems standpoint, because what actually happens with a movie is that a physical PRINT, one of a limited set, is manufactured by the studio for the distributor and sent to your local movie theater, where if you happen to catch a show on Wednesday night or Thursday afternoon, you may see it sitting waiting outside the manager's or projectionist's office in its thickly labelled octagonal hardcase. That print is then shown to a succession of audiences of 1-1,000 people each (never zero, they typically don't waste projector bulb time if there are no patrons, which might be for several daytime showings in a multiplex on a weekday). Then it is returned, or in some cases sent directly to a second run house. The times chosen to show this print do not necessarily coincide with the times it's being shown in any other theater, in or out of your home town. You cannot call up a friend in Salt Lake City from your sixth row movie seat in Chicago and say "wow did you see that!" - the friend won't know what you're talking about. You cannot see the movie at all if the house capacity is exceeded for a given showing. You cannot see the movie at all if it has not been released in your area. From a network topology standpoint we would describe movies as a point-distributed cache with asynchronous local broadcast. With digital distribution we may soon describe it as a multicast cache with asynchronous local broadcast. The physical transport of prints will cease, but the timed delivery of unwieldy large datasets will continue. Actually if digital projection precedes digital distribution, the present model may remain with DVD's or digital tapes replacing the cans of celluloid.

As this discussion's gone on I've realized that, as accused, we didn't define the terms well enough. Yes, movies aren't strictly broadcast, but as you say they are cached with asynchronous broadcast, and the viewing of them is a shared experience. I posit that the net will never be reliable enough that they'll be viewable as synchronous broadcast, and that people will always be interested in the shared part of the viewing. I think Todd and I need to discuss this one further.

Agreed. I think Lunch is called for and drinking of beer and ..no and drinking of scotch and we'll pen this bet on a pilot and post here on Flutterby.... BTW this is a rather cool way to record a bet