A MODEL FOR AN ADVERTISER SPONSORED INFOBAHN - some interesting ideas. I wonder

[David Farber <farber () central cis upenn edu>]

From: Toby Braun <utobia () mcs com>
Date: 5 Sep 1994 19:57:33 GMT
Organization: Eisaman, Johns & Laws Advertising/Chicago




BREAKING THE BARRIERS TO THE
NATIONAL INFORMATION INFRASTRUCTURE


Wednesday, September 7, 1994 The Council on Competitiveness and the
Clinton Administration!s Information Infrastructure Task Force will meet
in Washington, D.C. to discuss the applications being developed to run on
the national information infrastructure.


This document proposes one possible model, which I call bartered
sponsorship, based on cutting-edge technology and a respect for the
personal and economic interests of this country!s citizens. I hope you
will take a few minutes to consider these ideas and will further discuss
them with your peers.


These words are mine, they do not necessarily represent the views of my
employer or my industry. They represent ideas and ideals gained from 15
years of experience with networked computing and 10 years of experience
working as an advertising art director.




Toby Braun
Senior Art Director, Macintosh Systems Administrator
Eisaman, Johns & Laws Advertising
401 N. Michigan Avenue, Suite 2900
Chicago, IL 60611
312/828-040   Fax 312/828-0748




---------------------------------------------------




A MODEL FOR AN ADVERTISER SPONSORED INFOBAHN




!1994 Toby Braun Senior Art Director, Macintosh Systems Administrator
Eisaman, Johns & Laws Advertising Chicago, Illinois




There seem to be two starkly different visions of what we!re aiming at
with media convergence. One says that we!re working towards a utopian
vision of enabling technology and free information. The other seems to be
that we!re creating the commercial opportunity of a lifetime. The actual
arguments are certainly more complicated than this, but that!s the
essence of the two positions.


Does it need be one way or the other? Are the two ideals really mutually
exclusive? I don!t think so, I!ll argue that the answer lies in something
more familiar; an advertiser sponsored network similar to the way
advertisers sponsor print, broadcast and cable network content today.


First a couple of assumptions: 1) Media convergence will result in an
interconnected network of consumer !boxes! of some sort that will combine
the functionality of today!s televisions, personal computers and
telephones. For this discussion we don!t really care what the boxes are
going to look like, how they!ll be wired or who!s going to make them; 2)
The United States of America (and many other countries around the world)
will still have a capitalistic economic system.


Now that we!ve settled that, we can begin asking a few questions.


WHERE IS THE INFORMATION CONTENT GOING TO COME FROM? You certainly can!t
have an information superhighway without information. While there is
definitely no shortage of information in this country, most of our
history exists in analog formats such as books, papers, magazines,
paintings, vinyl records, magnetic tapes and so forth. The infobahn is a
digital domain and for us to even represent a small fraction of the
wealth of human knowledge on an internet will require a significant
investment by someone to convert our analog history into digital bytes.
But that!s not even the hard part, the topic of information ownership is
much more difficult.


As an art director in an advertising agency, I have to think about
information ownership. I have to be aware of copyright laws regarding the
use of photographer!s and illustrator!s images every day. If I sign a
contract with an artist to buy an image to use in an ad, what I!m buying
is not the physical image itself, but merely the right to use that image
in a single advertisement. If I want to !own! the image itself I have to
pay an additional fee called a buyout.


Writers in all sorts of mediums work in a similar fashion, selling the
rights to reproduce their words, usually under limited conditions, to a
publisher. A writer might sell the North American rights to a book of
poems to one publisher in New York and the European rights to another
publisher in London, but limit the London publisher to only one edition
(meaning the European rights would revert back to the author after the
first edition of the book was published).


The point is that most information isn!t free. Copyright laws protect the
rights of the author and forbid the distribution of copyrighted materials
without expressed permission. An artist or author will usually ask to be
paid for the use of their information, which seems fair, since everyone
has to eat. But as long as this is the case, we will need a system for
the commercial distribution of information to ensure that copyright
holders are fairly reimbursed when their information is accessed on the
infobahn.


I think that we can learn a lot by looking at two powerful distribution
models: movie theaters and network television. Each method has its
particular benefits. Movie theaters are a social experience, the picture
and sound quality is superior and first-run theaters feature a product
that can!t be found in any other media. Television, by contrast, is often
a singular experience and the medium!s currently limited fidelity is far
outweighed by its diversity and relatively low production costs. And of
course we pay-as-we--go when we go to the movies and pay by watching
commercials when we watch network television.


For some, movie theaters are the model of interactive media delivery
(perhaps since you have to be active to transport yourself all the way to
a theater). When you go to a movie theater, you only pay for the movie
that you want to see. And there is a simple kind of logic to this that!s
alluring.


But for others, an advertiser sponsored media makes more sense. You don!t
pay for anything but the box that decodes the signal and the people who
want to sell you stuff pay the networks to play shows for free (so you!ll
stay tuned-in long enough to watch a few commercials). This also
interests me, since I!m a big fan of getting things for free.


I support a bartered sponsorship system in which users would choose
whether a downloaded package of information (a first-run movie, a
personal newspaper, a satellite weather map, etc.) would be sponsored or
not. This type of system would preserve the best of both models
(presuming we!re using higher fidelity displays like those promised by
HDTV and at least CD quality sound).


HOW WOULD A BARTERED SPONSORSHIP SYSTEM WORK? I have to make another
fundamental assumption before I can define how I think a sponsorship
system should work. My assumption is that agent programming will mature
to the point where it is economically feasible to have an agent in every
box that can watch and even anticipate your actions by learning about you
and your personal preferences. (Even if this takes some time to achieve,
much of what I!m suggesting can be accomplished using slightly more
clumsy manual and semi-automated methods.)


I suggest that users should be able to securely trade data collected by
their agent programs to agencies seeking their particular demographic
profile. In return for sharing personal data, users would get credited
for a certain amount of online time, measured in hours or perhaps even
days. Every time a user accessed an information service on the network he
or she would have the option of having that event sponsored by the
addition of a commercial message. I believe this will spawn new, more
compact, forms of advertising messages, since it makes little sense to
wait through a 30 second commercial just to look at a weather map for 5
seconds. If a user chooses to have the event sponsored, the requested
event will be free (if not the user!s account will be charged
automatically). Some events may require viewing several commercials to be
completely free. In fact, the ratio of minutes of commercials to minutes
of program will probably be something like what we currently see today
(although I!d vote for a lower ratio since it raises the value of a
viewer impression to meet with the expected rise in quality).


In this model a user who chooses a sponsored online session could also
have the choice of picking a sponsor, letting his agent pick a sponsor or
letting the network pick a sponsor. If he wanted to pick a sponsor, he
would be presented with a menu of advertisers who have chosen to to be
associated with this program or service. If nothing in the menu was
appealing, the user could also choose any other sponsor from a personal
list of favorites, but would be credited at a reduced rate. If you let
your agent program choose for you, it would cross-reference its database
with the menu of available commercial messages, rating the connections it
found. A user could instruct her agent to automatically select the top
rated selection or to read the top few items aloud. If a user decided to
let the network pick a sponsor then the selection would be randomly
chosen from the requested item!s sponsorship roster (advertisers who!ve
told the network that they are willing to sponsor a request for that
particular service).


Users could decide to have commercial messages delivered at the
beginning, end or dispersed throughout the content (dispersed messages
would not be practical for all offerings!no one would want ad slogans
mixed in with a downloaded encyclopedia reference). A user would also be
allowed to defer a limited amount of commercial messages for later
viewing and would be encouraged to spend time browsing through
advertising messages to build up a nice personal stash of online credits.
Security systems would no doubt need to be created to limit abuse of
advertiser sponsored credits (you wouldn!t be able to simply order 8
hours of commercials and then turn down the sound and go to sleep for the
night). Most likely a user would be limited to no more than 10 messages
or 10 minutes (whichever came first) of commercial messages at one time.
[Thanks to Pete McNamara of the University of Wisconsin for helping me
work out some of these details.]


Security systems will need to be developed to provide users with the
confidence that their most personal data won!t be used in ways they don!t
authorize. For this reason, I suggest that the personal data gathered
never need be stored on any computer except the user!s own box. When a
transaction is requested the agent program sends a coded request that
contains personal information (a standardized data structure of
demographic and preference information) to the host network!s command
program. The network would then match the service request with a
sponsorship roster and price list and return to the agent program the
cost of the request in both minutes of commercial viewing time and
dollars. If we think of an Internet model for a moment we realize that
the user!s host needn!t be the server for the commercials or the
programming, so really all the user!s host network needs to return is a
list of addresses that describe where the agent can access the requested
items (it costs too much bandwidth to actually send all the items and
there!s little chance that everything will run off of some single global
master server). This also means that neither the user!s network host nor
the advertiser needs (in a technical sense) to ever store the user!s
personal profile data. If every transaction is transient and huge
databases of profiles are avoided, then (and I suggest only then) will
consumers have enough confidence that Big Brother isn!t secretly running
the entire operation.


The agent!s interface will need to be flexible enough to accommodate
several users, large argumentative families and the occasional wild
party. The agent must be able to weight the profile data of all active
viewers (using the system!s defaults when not enough or conflicting data
is available) before sending a request.


Users should be able to send an anonymous request if they wish to keep
their personal data to themselves. An anonymous request would simply pass
on preset system defaults. Users should also be able to create different
aliases or personality sets so that they could, for example, receive
different options when at work then they did when in the mood for basic
entertainment.


The agent!s interface will also need to be simple and adaptive. Users
shouldn!t have to suffer through endless layers of decisions. Ordering
online content shouldn!t be any more complicated than choosing a channel
with a remote control or sticking a video tape into a VCR. You should be
able to speak directly to your agent using natural language like:


!Agent. Give me a quick look at today!s weather report and satellite
view.! or,


!Agent. Who do you think should sponsor the intermission of tonight!s
screening of Mary Poppins?!


In the first case the agent would have been previously instructed that
!quick! is your keyword for the sequence of commands: !choose sponsored;
choose agent!s top rated sponsor; defer playback until later.!


!Agent! is a keyword you choose that the box uses to recognize that you
are talking to it and not someone else (like yourself).


In the second command example the agent would automatically send the
commands to order a movie that would have a commercial break somewhere in
the middle of the movie (the exact moment that the movie breaks away
should be tagged in advance for the sake of continuity). The agent would
automatically choose a sponsor based on its cross-referencing of the
sponsor roster with the user!s preferences.


HOW WILL USERS BENEFIT FROM A BARTERED SPONSORSHIP SYSTEM? I can foresee
several key user benefits under a system like the one I support. The
fundamental benefit is that users have the right to choose their own
level of sponsored interactivity with the system. Users can choose to pay
for everything ! la carte and avoid advertiser!s messages altogether. If
a user is interested in seeing only messages from certain advertisers, or
in certain categories, then that user will be able to filter out all
other messages, resulting in a much more desireable signal to noise
ratio. And, of course, a user could choose to have everything they do on
the network sponsored in some fashion, so that they effectively !just
have to buy the box! resulting in a nearly free system similar to today!s
network television delivery model.


Like the movie theater model, users only access (and if they so choose,
pay) for only the things they want to access.


Because a user can choose to make their personal preferences a part of
the sponsorship equation they will be exposed to less irrelevant
commercial messages. (Why do I, a single male, need to be inundated with
ads for baby food and pantyhose?) A user could define the types of
messages he will see both positively and negatively, prompting the system
to show more of one type of offering and to never show others.


But perhaps the most structural benefit of a bartered sponsorship system
will be an economic incentive towards high quality content. Advertisers
are usually willing to pay more to sponsor an offering than a consumer
will be willing to pay ! la carte, because the sponsor can usually
benefit by association with a program or service that people want to
view. Since this will mean that the overall profits made by both the
network and the people who created the information (writers, artists,
producers, scientists, etc.) will be higher, I think one can presume that
this will translate into a higher quality of the information available to
consumers. (That the creative mind needs to suffer in poverty to do its
best work is a myth.) Internal ratings systems will also play a role in
maintaining the quality of content as sponsors will certainly not flock
to offerings that the system!s records show nobody ever orders.


This last point might make some folks cringe with fear. A quick look at
an your local television listings during !sweeps week! (the period when
television programs are assigned ratings by companies like Nielson and
Arbitron) might explain this reaction. During sweeps week the most
shocking and titillating programs are usually offered!not exactly what
most people would call high quality programming. But to counter this
notion I!ll point to a recent survey of the types of !programming!
content we!re actually talking about for the infobahn. From the October
1994 issue of Macworld magazine, here are the top ten types of online
events that people want from the infobahn, in order of interest:


        1) Voting in elections
        2) Searching reference books
        3) Participating in distance learning
        4) Obtaining local school information
        5) Searching library card catalogs
        6) Participating in opinion polls
        7) Obtaining tax/credit data
        8) Participating in electronic town hall meetings
        9) Accessing government information
        10) Watching video-on-demand


Now certainly there are several of these items that I think should be
immune to sponsorship, namely access to government and basic public
educational materials and services. But I picked this particular poll
because it seems so far out of sync with every other poll that!s been
offered by the industries that are currently investing in the technology
that will be able to offer us such things. In the words of Charles
Piller, the author of the Macworld article, !These are poignant findings
in a nation where anemic voting totals are the norm and many citizens
typically view civic affairs with the same enthusiasm they reserve for a
trip to the dentist.!


Video-on-demand, the closest thing to programming in either the network
television or movie theatre delivery models, ranks well below offerings
that seem to be much more mundane in terms of their entertainment value.
Video telephony/conferencing ranks nineteenth in the poll.
Sports-on-demand ranks twentysecond on the list. Clearly, even if the
poll is to be taken with a grain of salt, there is interest in
participating with information in ways that can!t be appropriately be
modeled on purely commercial delivery systems.


I believe that the infobahn should be constructed around a basic set of
core capabilities that include the first nine items out of the Macworld
survey and a couple of others like access to information about proposed
laws; access to federal, state and local records; communications with
public officials and agencies; and access to emergency and municipal
services. (The exception to this statement is all the information will
need to be in the public domain, which limits reference works to those
that are not protected by copyright law unless other provisions are
made.) These types of access should be free to all users and made
publicly available to any citizen via terminals in public buildings like
libraries and town halls.


I do not believe that we can fund the creation of the kind of infobahn
the Macworld poll respondents are asking for without the economic support
of commercial activities on the network. Either we allow sponsorship or
we raise taxes to support the building of the network and its
services!and I don!t think people want to raise taxes any higher than
they already are. To think that passing the costs of development along to
consumers in a purely pay-per-access system is genuinely na!ve given the
tremendous investment in both research and execution involved.


HOW WILL ADVERTISERS BENEFIT FROM A BARTERED SPONSORSHIP SYSTEM? The
benefits of a sponsorship system that gives advertisers access to
valuable personal data should be obvious. Advertisers can better tailor
their messages to individuals and better track the responses to their
messages since ads will most likely have built-in hypertext capabilities
for immediate consumer responses like !show me more! and !order one now.!
But beyond these well documented benefits is another, more subtle,
benefit.


Advertisers currently suffer from a bad reputation in the eyes of many
consumers. Although many recognize advertising as a !necessary evil! in a
modern, capitalistic society, others rally towards banning advertising
altogether on the infobahn without really considering the political and
economic implications of such a position. If advertisers can work
together with the policymakers of the infobahn then they can move towards
a position that may be perceived by the public as more philanthropical
than simply a matter of personal greed. It is important to remember that
just as we the people are the government of this country, we the people
are also the businesses and industries that need advertising to survive.
I believe that much of the public distrust and distaste regarding
advertising stems from the advertiser!s current desperation to create a
message that will somehow penetrate the enormous amount of clutter
(noise) that defines todays media. Any system that works to reduce the
clutter (especially one devised with the support of the advertising
community) will work to erode these negative feelings and reinforce the
positive benefits of advertising.


HOW WILL THE INFOBAHN ITSELF BENEFIT FROM A BARTERED SPONSORSHIP SYSTEM?
I made most of this argument earlier when I mentioned that I don!t think
we can build an infobahn with the depth of services we seem to want
without some kind of sponsorship system to help fund its construction. I
think a bartered system, where consumers trade personal information with
advertisers and get to choose whether an online request will be sponsored
or not, both helps stimulate demand and maximizes the profitability and
quality of the system!s content.


The early phases of a commercial infobahn are already taking place in the
form of online services like America Online and Compuserve and in tests
of movie-on-demand and Internet access systems sponsored by content
providers, computer companies, cable companies and telcos. If advertisers
are smart they will enter into systems that already exist and begin to
offer sponsorships of some of the basic services that can already be
delivered today. By interacting with their customers using these same
technologies, advertisers can gauge public responses and modify their
messages and behavior to better meet expectations. Advertisers will pass
along consumer demands to network operators and through their mutual
cooperation new services will be created that couldn!t have existed had
either party worked alone. This will accelerate the growth of both the
digital knowledge base (translating our analog history) and greatly
increase the depth of services available.


WHERE DO WE GO FROM HERE? Beyond the call to action above, I think we
need to consider the positive benefits a convergent, interactive media
offers our entire socioeconomic system. We often consider industry and
business and politics and family to be discretely separate entities. But
this is not the case. The construction of the infobahn or whatever its
going to be called must be a collaborative effort. It is not enough to
think that you!re going to be happy with whatever you get, because if
enough people think that way we won!t get much of anything. People of all
areas of interest must become involved in discussing and planning our
digital future!building a digital community has to be a community effort.
I believe a bartered sponsorship system that is structured around user
choice and advertiser interactivity will help us build a dynamic digital
domain that best integrates with our current socioeconomic reality.




[If you would like to direct comments to the author please feel respond
via e-mail to utobia () mcs com or by telephone at 312/828-0400 ext. 151.]