« A wee drink for thirsty astronauts | Main | Net Neutrality - Money Talks »

March 21, 2007


Hamish Harvey

Hi Earl,

But what's a podcast? Two uses of the delivery mechanism that I make use of:

BBC Radio 4's "In Our Time". I never catch it when it's broadcast, but now I can accumulate and listen until my ears bleed on long car journeys.

IT Conversations: recordings of presentations at events that are happening anyway, but that I can't go to.

I guess you're talking about the more "audio web log" end of podcasting?


Earl Mardle

Hi Hamish.

Then you use podcasts, its interesting that you don't know the term, which undercuts the model even further since the story talks about the growth in name recognition while actual use lags far behind.

You are also partly responsible for their problems; you have a couple of sources that you like or trust, you use them regularly, but you accumulate and time shift them, which makes any announcement, especially advertising, moot.

If I have a website, or an RSS feed, I know not only WHERE you viewed the ad, and the IP address, and maybe the user, and sometimes what other ads, pages, content etc you view, I can also substitute it in a heartbeat so that the very next viewer sees a different ad.

In fact, using those data above, I can customise the ads according to your interests and preferences, dynamically. But once the ad is in the audio stream I lose everything. I have no idea when it will be heard, or by whom. I lose every advantage that the connected web gives me. Even if a page is ten years old, I can serve other content into that page that is 10 seconds old, customised to the user.

Now if podcasts were networked it would be possible to serve updated inserts into the stream, but since most users are by, definition, disconnected when they hear the programme, the best I can do is generic, brand-support stuff. At least radio is real time, but podcasts are the worst of all the bargains.

And the other point is that, even though you are a solid podcast user, you contribute nothing to the growth; you have two regular sources, you use them all the time, but, at least by your report, you don't sample others, add them to your pod-stream etc.

And there's a good reason for that, even your long journeys have a limit; there are only so many seconds in a lifetime and using them for one podcast at a time is the best you can do.

My aggregator can sumamrise 100 RSS text feeds in a single web page, I can flick through the 10 new items in seconds and open the 3 or 4 that I find interesting and open them, and them alone.

Text is random access, audio is linear, I don't even know what is in IT Conversations until I listen to the intro, or remember Phil's text email that tells me what is in the piece but is not connected in any way to it. I can't just click to the two items that I find interesting, I either have to listen to the whole thing, or risk an accident by spooling through to the bit I want.

Or something like that. Thanks for the nudge.

Jon Husband

I agree with this post.

I don't want to pretend I am smart, but I have never thought podcasts would displace radio, nor scale well, and I agree that they are not an efficient way to connect and build relationships. They are good for people who like to play at radio, they may be good for a small percentage of people who like to listen to someone else (or themselves) while walking, or driving, etc. They are also another vehicle for people to use to promote themselves (which I think too many people do too much of already, probably including myself).

I do think podcasts have use in the domain of education, or in training or certain forms of internal communications in enterprises.

And yes they are not inherently flexible nor are they easy to pound into little bits and remix into something else.

[EM] Cheque's in the mail.


Agreed, they just don't have what it takes in a networked world, at least I can quote, comment and forward to a mailserv an email I find interesting.

Hamish Harvey


Many fair comments.

I have heard the term, I was really questioning what qualified, in your view, as a podcast. Your post seemed to be saying that the primary problems with podcasts was performance (which I took to mean production). Perhaps that production problem isn't so bad if either a) it has already been paid for (In Our Time, which is broadcast as a regular radio programme) or b) the listener is prepared to accept some rough edges (IT Conversations).

Then there's the listener end in time-shifted broadcast mode. I use google reader to filter the IT Conversations feed. Each is available with a short description, and anything that strikes me as interesting I tag "listen". Google reader exports a feed of everything I've marked "listen", which I then subscribe to in a ipodder.

Sure, I can't tell if the podcast will really be interesting, but I don't have to listen to get some idea of the content. I'm slowly learning to be more selective, too, because as you note, my listening time is limited.

And sure, I'm using a bunch of tools in there which aren't immediately obvious to the average potential user.

A third issue you raise is that of conversations and the fact that podcasts play badly in the networked world. This is true, and it seems reasonable to assume they will always be harder to deal with than text. But the problems could be alleviated with the right tools. Jon Udell has written a fair bit about this issue. Audio and video players are all designed on the assumption of linear listening/viewing of whole files; there are no readily accessible facilities for linking to segments. Udell managed to work out how to link to a point in a real audio stream, but it really took some doing.

I don't see podcasting exploding as a conversational _medium_ like web logs (and my own web log writing has run into the ground because the production costs are already too high relative to other demands on my time at present ;). But there's a raft of material which is becoming available, and I can forsee that material being the _subject_ of conversation _if_ it becomes possible to link within files/streams.

Every medium has its strengths and weaknesses. Some of the weaknesses are inherent, while some are to do with the surrounding technology. Essence and accident. Some are to do with my convenience, some with business models.

(On that subject: I do fast forward past the ads on IT Conversations. I have a visa card already ... I have donated, though, and will do so as long as I continue using the service. And I pay for the BBC already!)

Services like CastingWords (http://castingwords.com/) are a potentially interesting addition to the mix.


Earl Mardle


I can't disagree with most of that. I suppose my main issue is with the boosters of podcasting as the next big thing which I don't buy, and with the numbers now coming out which confirm my suspicions that this is not going to provide potential advertisers with the kind of measurability nor audience reach that will be needed for it to be more than an also ran business model.

Your point about the BBC recycling programmes is spot on, it makes huge sense for any broadcaster to do just that, both in audio and video, because the marginal cost of turning a mass audience product into a podcast is near zero. But creating niche market audio is another kettle of fish, for that to be worth doing it has to be COMPELLING for its market to make it worth funding.

And its not that I'm down on audio, it was my bread and butter for 23 years, there's a lot about it that I like a lot, I wish like hell that it would network.

Which is why I would LOVE it if Udell could find way to link into and out of, and give me random access to parts of audio streams and tools for marking entries and exits the way we could with the old Mini Discs.

And for your sake, i hope they keep it up, I just think you wont have an increasing amount of company, that podcasting has probably peaked and the networking tools that might have saved it will be applied to video instead.

The comments to this entry are closed.