JMS

We caught up with Rich Dean from Austin’s fabulous KUT today on day 6 of the Digital Distribution Consortium working group sessions and got his take on the pros and cons of the aggregated approach and the priority needs for stations like KUT, much of which Rich outlined in this prescient email below back in May.

From: Rich Dean
Subject: Re: Taking the Next Step:
Date: May 4, 2006 3:06:51 PM EDT
To: Tim Olson, Mike Bettison, Jake Shapiro

Hey Guys,

Hope you all had fun in DC. Alas, my travel budget for the year is spent, so
I couldn’t go, but Stewart was there.

Tim, I think you’re right. Far too many folks are looking for a “thing” or
system to solve the challenges we face, rather than a solution that might be
squishier, but more effective (standards, cooperation, unified marketing,
content sharing).

I do have some thoughts about next steps, some of which will be repetitive
with what I just emailed, but rather than retype it, I offer the
comments/next steps below (Jake, you’ve seen this mostly):

—–

My comments are predicated on these goals:

1. Bring a greater degree of cohesion between station and network web sites
2. Find new sources of revenue to support our activities
3. Reduce redundancy whenever possible (not at the risk of local innovation)
4. Strengthen local stations (and continue to support the network sites)

In a “new reality,” the biggest threat is at the local level. With “bypass”
likely and increased ways for local listeners to find national shows, we
need to find ways to strengthen our stations.

To that end I propose that we focus on these areas:

1. The podcast project — or more accurately, time shifting.

Downloading quality audio files is an inherently flawed approach. It is
costly and places the burden squarely on the broadcaster. Compare this to
time-shifting television, which is increasingly done through DVRs (Tivos).
RadioTime.com (and others?) enable users to record local station signals. It
is podcasting without the bandwidth. And it is a locally-focused approach.

Of course Tivo-like solutions only cover broadcast content, but NPR and
others should be working closely with these time shifting companies to bring
these tools to our listeners.

As for podcasts, it would be interesting to continue the project, but rather
than focus on building a large backend system with a new staff we should be
focused on how local stations can insert content (news, stories,
underwriting) into national podcasts. This more pressing issue is cheaper to
solve and results in more dollars to the local station.

2. Federated Search

No single project would do more to bring cohesion to the system’s web sites
than federated search. While technically this project is not trivial, it is
also well within Google’s (or other search engines) capabilities. PBS has
been working on this project as well with their stations — an alliance with
PBS.org would be a tremendous advance for public broadcasting users.

Sponsored results and/or banner ads on result pages are among the easy
opportunities for increased revenue.

3. Local Content on Network Sites and in National Podcasts

At NPR.org we can identify local users (at least those who have chosen to
localize or been ‘forced’ through links from local stations). We should be
able to serve local newsfeeds as well as banner ads to station visitors to
NPR.org. Station sites already include network content on the local site,
NPR (and APM) should offer the same opportunity.

Again this advances the cause of integration, while strengthening the local
station. And from a technical standpoint this too is relatively simple.

It is already possible to “stitch” audio files together to create a single
podcast. Localized versions of national podcasts could include local
newscasts as well as local underwriting. This revenue opportunity also
strengthens the local brand and mitigates “bypass” concerns.

4. Banner Ads

This is a relatively simple exercise technically, but has enormous revenue
potential. There are several aspects: aggregated inventory, local banners on
NPR.org (see above), and shared standards.

KQED, PRP, and NPB have already started projects around aggregated
inventory, but some national standards and leadership from the networks
would help make this accessible to everyone.

As for local banners on NPR.org, we already know how many localized page
views there are for each station. Allowing a local rep to sell that
inventory shouldn’t be rocket science.

5. Shared/Common Metrics

This is an old project iTAP, that needs to get back on track. I don’t think
I need to say any more about this.

Thoughts on Common Backend Systems

Building a shared backend system for content delivery is, to put it mildly,
a massive undertaking. Look at Content Depot, which took years to build (and is not up yet, really) and millions of dollars in investments.

The goals of a shared backend system are poorly defined at this point, as
are the benefits to the local station. While there are benefits to group
purchasing of bandwidth or other services, this does not then argue for
shared hardware or software. Every station has different needs and being
hamstrung by lowest-common-denominator approaches ultimately hurts the
system. In short, a shared backend limits our ability to innovate at the
local level.

If a shared backend must be built, it must be managed outside of NPR. NPR
cannot be the central location for all network and station content.

In the end, spending the next couple years and millions on a shared backend
serves us less well than aggressively addressing the 5 issues above. And if
we address the 5 issues over the next year or 2 we will have a track record
of working together as a system — and better understand what our needs may
be for centralized capabilities.

Simply put, there is no pressing need for a centralized system yet.

If the goals we set out are…

1. Bring a greater degree of cohesion between station and network web sites
2. Find new sources of revenue to support our activities
3. Reduce redundancy whenever possible
4. Strengthen local stations (and continue to support the network sites)

…then our best solution is to work together on the issues laid out above.

How should we best work together to make this happen? Your comments welcome!