Last week Pete Cashmore blogged about a new podcast transcription service called Trancribr. I wanted to test it out so I got out my Paypal wallett and splurged $11 for 11 mintues of a podcast I recorded last week. It was dead simple to sign up and I got the results back this weekend (it has been sitting in my inbox, but was about a three day turnaround).
I've not edited the results (below), just republished what they sent - I thought you'd be interested in seeing the raw results. I'll make edits to the transcript and post separately later.
OPML and Attention
The audio for this podcast can be downloaded at http://raq.fig.com/~lisatmh/podcasts/alex_barnett/opml_attention_alexbarnett_nov20_2005.mp3
This morning I woke up to another little podcast going on about Attention and OPML. It seems that Attention and OPML and podcasts seems to be the known relationship there. Podcasts about OPML and Attention than anything else.
I listened to Nick Bradbury with Steve Gilmore and a chap I’m not too familiar with, and I’m sure he’d be delighted that I just said that, called I think it’s Mike, and they were talking (That’s unfair. I think I’m going to edit that out. I can’t just leave that. Sorry Mike, if you’re listening.). What they were talking about was OPML and the support for named spaces and how that is essentially opening the road up for OPML to become the attention data format.
1:03 - The people there, with Steve Gilmore himself as kind of essentially the chief pusher of attention.xml, behind Attention Trust and Nick Bradbury’s Exec Director, I think, of the Attention Trust so I think these guys’ views clearly matter on this case. What they were talking about was a number of things around who’s going to be using this Attention data and how it might be applied. Nick’s view, and he admitted this, he said this often himself, that he was looking it from an aggregator’s perspective. He’s built an aggregator feed demon as part of the Newsgator family of products and so really his perspective on OPML and attention data is coming from that perspective and what I’d say is that that’s one – and like I said, he’s totally admissive of this – it’s one perspective. It’s one view and I’ve bandied on about OPML in the past and how it’s got a huge potential, I think, beyond the aggregator. How exactly is going to be figured out. I’m not entirely clear.
But there’s a few things that I wanted to say and this is rambling here and just kind of making my podcast for the sake of it here. There’s a few things I wanted to say about this. It’s weird doing this on my own by the way, speaking to a vacuum.
2:36 - One is that picking up on Nick’s proposal around Brank, what he was talking about there was a way to be able to express how a feed is of value to you – its low value, high value – and definitely that kind of information inside of Attention data is going to be of use, and I’m supporting that for what it’s worth, not that anybody’s going to care but I personally am supporting them.
3:01 - The second part to this is – and that’s at the feed level – so what I’m proposing is something that works actually at the item level, so we can get to another level of granularity on this. Now, I wrote a post up a couple of days ago and it was clear that I hadn’t made myself clear. Some of the comments I was getting back were saying things like in the quotes which is, “Ya’ know, isn’t the fact that you’re already subscribing to an RSS feed essentially a vote for in the first place?” Absolutely. Yes it is, but that’s not what I’m talking about. I’m talking about at the next level down, beyond the feed, below the feed, which is the actual items themselves. I’m talking about URLs to posts. I know that’s my reader telling me I’ve got more things to pay attention to. Classic. And posts, web pages, could be podcasts, could be videos themselves. I’m talking about the actual items themselves that the feeds would be pointing to, and that they have a URL. That level of information is what I’m talking about.
4:23 - I know there’s been clicks has been talked about, especially around the Attention.xml schema and I, again, haven’t have necessarily a problem with clicks. I think it works and the rest of it but I’m not entirely sure that all of the value can be extracted out of clicks compared to something different which is me explicitly saying that I think this item is of interest to me or "this podcast, I listened to it and I vote for it," is all I’m saying.
Nick did actually point out in his comment to me that this already exists in a way inside of the attention.xml schema and is in fact a micro format and it’s called Vote Link. Vote Link is, as the proposal stands today, is a way for you to say, “Hey yeah, I’m gonna vote for this.” You can also have a vote against this or another vote which is abstain. So there’s kind of three states in that. I think that actually that’s probably what I am talking about, and so I think I must have missed that in the first time and the good news is that there’s something there already that’s kind of in the attention.xml schema that I think we should leverage.
5:44 - The value that I see in that, the reason I’m pushing for that at least to be within the OPML attention data is that I think that data can be incredibly valuable both to the user and to the services that would be leveraging that data. I put out along with this post just a scenario that describes it but it’s very straightforward I think. The idea is that I would be one day going to amazon.com and today at amazon.com, I am buying something like 5 books a year and the reason I’m buying it from Amazon is because they’ve got nice relevancy, nice recommendations; but all of that data that they’re using essentially is based upon the data that is collected while I’m at the amazon.com site which, as I said before, is probably like 0.01% of my total surf time, if that.
6:49 - The question is how, outside of my amazon.com experience, can I get my attention data, either things that I’m interested in, into Amazon so Amazon knows that and so that Amazon can provide an even more relevant experience so that to the point where instead of me buying 5 books in a year, I’d buy maybe 10, or more books than 5 in a year because the thing is more relevant. So then the question is, what data would Amazon need in order to be able to do that in a format that makes sense to it and in a format that is portable from my point of view, and that I own and I manage.
So that’s the idea behind the Attention Trust which is my data and I’ll give my data along to people, services, applications as and when I want to and that I’m in control of that. So the idea would be a kind of this service where I’m on Amazon and kind of an initial, kind of basic, straightforward idea is that I could say, “Oh yeah, I’m going to import my OPML file now into Amazon,” and now Amazon looks at that OPML and there’s two things that it looks at – one is the feeds and the right against those feeds as Nick was proposing and I think that that would definitely be of value. There’s stuff to do with that. No question about it.
8:25 - Then there’s something at the lower level than that which is URLs, posts, podcasts, videos, URLs of other books or whatever that I have previously stated: “Yes, I’m interested in these things,” and I own that data. Now Amazon looks at that and the idea is that it looks at the URL or one of the URLs that I’m pointing to that I voted to, it could be a web page. If it’s a web page, it looks at the web page and it has a look and analyzes what kind of content that is through its own analysis of key words and whatever and then it can match that against, or import that kind of data into its recommendation and then include that as part of its relevancy, or as part of its recommendation engine. Therefore, that’s why it would be more relevant to me because it knows this data about me.
9:16 - In that example, I think I’ve given there as a scenario that is outside of the aggregator space, traditional aggregator space, OPML – that’s one of the big drivers for around the aggregator. I guess what I’m saying is that there’s a world outside of the aggregators and if these include the applications, recommendation engines, e-commerce engines, whatever you want to call it and that by having a standardized way for us as users to be able to express our interest and manage that through its kind of format, that’s what I’m getting at here. I think that’s why beyond feeds, there’s value in having that.
10:01- I mentioned already the vote link could well do that actual job. The great thing is it’s being specified already today and that we could just kind of include that as part of the OPML or at least one of the named spaces inside of that. I just wanted to clarify. That’s what I wanted to do for this podcast. I just wanted to make that clear and I wanted to verbally explain why I think that this is important, and I look forward to hearing from you.
enabled by transcribr Get your own podcast transcribed at www.transcribr.com