Over the years, the Public Broadcasting Service has stayed on the leading edge of technologies that cut costs and enhance programming services.

Among other things, it claims the first satellite distribution network (1978), the first all-digital TV-broadcast facility (1996) and the first national HD broadcast (Ken Burns's Frank Lloyd Wright in 1998).

Promising to keep the tradition of innovation alive these days is John McCoskey, who joined PBS as chief technology officer in April 2007.

In this interview with TVNewsday, McCoskey says he is wrestling with the same DTV-transition problems as his commercial counterparts but, at the same time, is pushing ahead into file-based program distribution, mobile broadcasting (real and non-real time) and multiplatform services.

An edited transcript:

PBS is overseeing the DTV transition of its member stations. Are the PBS member stations ready to make the switch to DTV next February?

 If you think about it, this has been an effort that has been ongoing for about 10 years, so stations are kind of at the last stages of doing the things that they have to do to meet the FCC mandate for transition. What we're also seeing is stations that are really starting to grow into their new digital infrastructure.

What I mean by that is, they're a lot more interested in HD and they're a lot more interested in multicasting, which is something that public broadcasters have embraced a whole lot more than their commercial counterparts.

How many noncommercial stations are still gearing up for the switch?

I would say there are about a dozen stations that have significant effort before them. When I say "significant effort," I mean things like transmitters and a lot of the heavy iron. It's a very small minority of public stations. So, in general, we're in good shape.

I know that some noncommercial stations have been experimenting with distributed transmission systems as a way of improving coverage. Are you a proponent of that?

I would say that public broadcasters have taken two approaches to that. One is a more traditional translator approach, common in the mountain west, where we deploy hundreds of translators. Now, it's interesting that those sites are not required to transition to digital as part of the FCC mandate. So they fall under the same category as low-power broadcasters. The large bulk of them will stay analog after the transition.

We've also been looking at the same-frequency networks, interestingly enough, through our involvement with the Open Mobile Video Coalition and the ATSC. They have been looking at taking the DTV broadcast spectrum and using it for mobile services to handhelds and to vehicles.

And you think this single-frequency network is a better way to get the propagation you need for mobile.

Yeah. The trials that they've done on mobile highlighted what you would expect to find in some urban environments, where you have canyons and holes in coverage. So what we've been looking at are same-frequency networks for filling those canyons.

Are you also involved in helping stations get their digital signal to cable headends?

Yep. In fact, we've got an arrangement involving PBS, the NCTA, member stations and APTS, our legislative-affairs group. As part of that agreement, we have worked closely with the stations and their liaisons with the cable operators. So, in terms of timing, we're pretty confident that everything is on track.

It's really a matter of when are the cable operators going to switch to the digital lineup and, more important, how are they going to deal with HD from stations that actually specify how they want each operator to deal with the downconversion of their 16:9 content to 4:3.

Most cable operators will carry an analog downconverted 4:3 version of the PBS stations as well as the HD version. For the analog version, the stations have to work out with the cable operators whether they want to do a center cut of the picture or provide a letterbox display.

Well, that's more of a negotiation than it is a technical issue, isn't it?

Today it is, although we're looking forward to something called AFD, or Active Format Description. Basically, that allows us to tell the receivers at a cable headend what the content format should be. That's something we've been working on, along with stations, the cable operators and cable-equipment suppliers.

There seems to be growing concern that the big problem with the DTV transition will be not with converter boxes but with the antennas. Is that your impression?

There's a lot of anecdotal evidence that viewers are having trouble with that, and I think we can kind of categorize those in a couple of ways. One is an awful lot of stations, both public and commercial, are not running full power on their digital channels. They're doing that to save money and to save on their transmitter costs while they have an analog channel up. So a lot of viewers out there are getting a weak signal because of that.

The other part is what people call the cliff effect, where there's not much of a zone for degraded operations. It's pretty much you have it or you don't. People who are now getting a snowy, ghosty analog picture may be just on the other side of the digital cliff. They're not able to reliably decode digital. Outdoor antennas for a lot of folks will help certainly, but it's going to be a troublesome part of the transition.

Do you think that digital heralds a rebirth of over-the-air television? Do you think more people will opt to receive TV off-air?

I don't think we've seen any evidence of that. Now, you know, one of the upsides to this is that, with over-the-air digital, you do have HD and you do have multicast channel capabilities, so the number of free over-the-air options will be increasing.

Some viewers will probably recognize that and probably take advantage of it, but we're certainly not seeing a big trend of people saying that the experience is that different that they're dropping cable and satellite services.

How much of the PBS HD service is true HD, not upconverted SD?

I'm guessing that it's probably a small majority. It's probably 50 to 60 percent. Today, there's a lot of legacy stuff out there that is being upconverted. One of the interesting things for us is, we have a lot of kids content that, because it's animated, upconverts very well. It really looks good on the viewer end, and it's not unlike what you're seeing on most other HD networks, with the exception of Discovery and HBO, which have made commitments to do only native HD on their HD channels. Everybody else is pretty much mixing native and upconverted.

Is there a point where PBS says, we are only interested in native HD programming?

There will be a period of a few more years where there's a mix, and, of course, there is always going to be some legacy content that deserves to be on the air. But as we talk to our big producers, they're all transitioning to HD. The NewsHour [With Jim Lehrer] went HD a few months ago. That was a big step for the public-television community. So it's happening along with the rest of the world.

PBS distributes some SD multicast channels, and individual stations program some of their own. Do you have standards that limit how much stations can compress your HD service for broadcast to make room for multiple SD channels?

We really don't. It's a station choice. We like to give best practices and help educate stations about those decisions on bit rates and compressions, but it really is a station choice.

Now, what about mobile? You mentioned before that you are involved with the ATSC or the Open Mobile Video Coalition.

We're interested because we think there is potential for stations from both a public-service model and a commercial exploitation of part of their bandwidth. So, for us, it's an interesting technology that we feel could have positive implications for stations. We think it's a little bit early to know for sure what those are, but it's interesting enough that we want to stay very close to it.

One of things that is potentially very interesting and valuable to public television is the concept of taking some of the kids content and essentially downloading that into handheld and mobile receivers during off hours.

Mobile video doesn't have to be real-time. It can be distributed in a non-real-time trickle. Stations that might have excess capacity overnight may be able to take a selection of kids programming and download it into the backseat of cars and into handhelds in shopping carts. That's kind of an interesting opportunity.

As I understand it, you assemble and get your programming ready for distribution at your media operations center in Alexandria, Va. Is that ready for the world of new media, or is that still a work in progress?

Do you mean from an infrastructure standpoint? I would say it was a work in progress two or three years ago. We've built an infrastructure that's brought us to a point of file- based workflow. Now the bulk of our efforts is scaling that for high-definition. As you might expect, high-definition puts a lot more stress on storage and on the size of data-communications pipes and those kinds of things. We are also moving the architecture into non-broadcast multiplatform applications.

We're taking our content and making to available in a way that stations can provide access to PBS shows on their Internet pages. We're delivering shows to iTunes and to VOD on cable systems and to third-party syndicators like Hulu, which is a great example of where you can see high quality online. So we're really seeing a big shift in using that common infrastructure to drive lots of new publishing outputs for PBS content.

You distribute content via satellite over what you call the Next Generation Interconnect System. What's its status?

NGIS has a couple of phases to it. We're well into the second phase. The first phase is an industry-standard satellite-distribution mechanism. We went with a standard that's known as DVB satellite transmission. The idea there was to get away from single-vendor support and go to a multi-vendor satellite-distribution approach. Today, we put up a number of linear feeds, both high-definition and standard-definition. It's a traditional approach to television distribution.

What the next phase of NGIS does is take the majority of those linear feeds and change them to a file-transfer mechanism. So, if you take the example of Nova, we bring that in, process it and ultimately put it over the satellite on a linear feed. We'll also take that file and push a copy out to each station. The stations assemble the files that they want to put into their local broadcast and drive the local broadcast that way. So it pushes the file-based infrastructure beyond PBS headquarters out to each station.

Do you foresee PBS distribution remaining all satellite?

This is a system that will be in operation through 2016. When we look to replace that, my guess is that high-speed terrestrial circuits will be a viable alternative for a large number of stations. We'll still have some stations that are going to be most cost-effectively reached through satellite. So I would anticipate a hybrid at that point.

HD is a bandwidth hog. Do you have enough satellite capacity today to accommodate all your HD needs going forward?

Yeah. We've had to make some changes in the way we do distribution. Traditionally, in standard definition, we've carried four different time zone feeds — Eastern, Central, Mountain, Pacific. As we go to HD, we don't have the capacity to maintain that number of time zone feeds, so we're dropping down to an Eastern and Pacific feed, similar to what you see most cable networks do these days.

How do the members feel about that?

It's problematic for members in the Mountain time zones. The majority of stations that are in the Central time zone run off of the East Coast timing anyway. But those in the Mountain time zone are going to have to time-delay their channels, and, for some stations, that's challenging.

When the NGIS becomes file-based and stations are much less dependent on linear feeds, the files are actually staged at stations a number of days before they're due to broadcast. It kind of takes care of that problem. So it's really a transitional problem.