Re: [backstage] BBC Touch
Chris Riley wrote, reordered slightly: In particular I think its useful for highlighting issues the public care more about. For instance a couple of says ago whilst Pakistan was the headline, most of us were reading the climate change story. Are you sure Pakistan was the headline? The climate change story became a subheading just after midnight on the 29th, became the main headline around 07:50am, and stayed there, as far as I can tell, right through the 29th and 30th October, until 04:05 on the 31st October when the main headline became the Prince Charles/Pakistan story for around 15 minutes (data from my front page archive: http://www.bbc.co.uk/homearchive/ and my news archive). On the web page you'll see subjects they want us to read about vs. what we're actually reading about for the past 24 hours, and past 2 weeks. they want us to read? That's not the point of the editorial (by which I mean the ordering of stories on the front page) at all, in my view. I have BBC news in my RSS reader, so that gives me the latest news. I click on the ones I want to read, but that shouldn't affect in any way which ones the BBC decide are important. They hopefully weight stories by more than popularity, otherwise all the stories would be about celebrities and kittens? :-) What your site measures (presuming the popular feed goes on page views, which seems likely) is which stories have been clicked on, not read. I frequently click a headline if it sounds interesting, read the first paragraph, decide it isn't or I already know the story, and close the page. If lots of people are like that, then that makes that story a popular story even though it isn't at all. So what you're actually measuring is how good BBC headlines are at getting people to click through. Similarly, if a BBC post gets linked to from Slashdot or Boing Boing, it will almost certainly become a most popular link. But that doesn't mean it is most popular in terms of the what we're actually reading about, just that lots of people read those sites and click links, realise the first paragraph tells them all they need to know, and that's it. Most emailed would perhaps be a better XML feed to use than Most popular, as then at least people have gone out of their way to send the story to someone else. But that doesn't change my first point - the story that is most emailed will be the one about a man marrying a goat or somesuch, which I wouldn't think should be a top story anywhere, no more than the also in the news bit of the front page for humour value. Lastly, surely the headline stories on the front page are, quite possibly, new news. The stories that are most popular are going to be those that have been most widely distributed, by email, IM, RSS, whatever, and so will almost certainly be a few hours behind. So I wouldn't expect the headlines to match the most popular? I don't have any desire to highlight any hidden agendas the BBC's editorial staff might have (although I guess it can), but more from an interest in how in touch are the BBC with what the public actually reads and cares about compared to what they think we do. Why would the BBC want to be in touch with spammers? I say this because of the story some months back of the BBC's MSN charging story from 2001 suddenly becoming very popular - On Sunday it was the most-read business story - http://news.bbc.co.uk/2/hi/business/4778046.stm - a complete irrelevancy from the headlines point of view. :) -- ATB, Matthew | http://www.dracos.co.uk/ - Sent via the backstage.bbc.co.uk discussion group. To unsubscribe, please visit http://backstage.bbc.co.uk/archives/2005/01/mailing_list.html. Unofficial list archive: http://www.mail-archive.com/backstage@lists.bbc.co.uk/
Re: [backstage] BBC Touch
Hi Matthew,Thanks for your feedback. On the Pakistan thing, I'm not sure, it was just when I glanced at it when writing the mail that stuck in my mind. So you're probably right :o)On your second point about the editorial, completely agree that they shouldn't be ordering stories by popularity (I think I mention in the about page that they shouldn't be striving for 100% or something), but I do still think it is fair to say stories they want us to read about, a story is more likely to be read if the editor has chosen to make it a headline on the news homepage wouldn't you say? As for the popular feed and the data it provides being accumulated with a certain degree of lag between new news and what we've been reading over lunch, yeah, of course you're always going to get that issue, and I completely understand where you're coming from. Similarly saying that its the BBC's ability to write a good headline that's being measured is a fair comment, but I don't believe it provides an unfair popular bias to a story. In part that's why I introduced the automatic subject extraction, to try and help get through that issue and delve into the actual subjects we're choosing to click through (and so show some interest in) vs. those we don't. That way even if you don't read the full story, clicking through registers your interest in the subject. You'll never be able to get away from the fact the journalists are going to write captivating headlines, but if they are consistently doing that, with the main subject in the headline, I'm happy with that. Ultimately its an experiment with the data, and if someone can make use of the data in a new way thanks to my efforts, I'm happy with that.Cheers,ChrisOn 11/1/06, Matthew Somerville [EMAIL PROTECTED] wrote: Chris Riley wrote, reordered slightly: In particular I think its useful for highlighting issues the public care more about. For instance a couple of says ago whilst Pakistan was the headline, most of us were reading the climate change story. Are you sure Pakistan was the headline? The climate change story became asubheading just after midnight on the 29th, became the main headline around07:50am, and stayed there, as far as I can tell, right through the 29th and 30th October, until 04:05 on the 31st October when the main headline becamethe Prince Charles/Pakistan story for around 15 minutes (data from my frontpage archive: http://www.bbc.co.uk/homearchive/ and my news archive). On the web page you'll see subjects they want us to read about vs. what we're actually reading about for the past 24 hours, and past 2 weeks.they want us to read? That's not the point of the editorial (by which I mean the ordering of stories on the front page) at all, in my view. I haveBBC news in my RSS reader, so that gives me the latest news. I click on theones I want to read, but that shouldn't affect in any way which ones the BBC decide are important. They hopefully weight stories by more than popularity,otherwise all the stories would be about celebrities and kittens? :-)What your site measures (presuming the popular feed goes on page views, which seems likely) is which stories have been clicked on, not read. Ifrequently click a headline if it sounds interesting, read the firstparagraph, decide it isn't or I already know the story, and close the page. If lots of people are like that, then that makes that story a popular storyeven though it isn't at all. So what you're actually measuring is how goodBBC headlines are at getting people to click through. Similarly, if a BBC post gets linked to from Slashdot or Boing Boing, itwill almost certainly become a most popular link. But that doesn't mean itis most popular in terms of the what we're actually reading about, just that lots of people read those sites and click links, realise the firstparagraph tells them all they need to know, and that's it.Most emailed would perhaps be a better XML feed to use than Most popular, as then at least people have gone out of their way to send the story to someoneelse. But that doesn't change my first point - the story that is mostemailed will be the one about a man marrying a goat or somesuch, which I wouldn't think should be a top story anywhere, no more than the also inthe news bit of the front page for humour value.Lastly, surely the headline stories on the front page are, quite possibly, new news. The stories that are most popular are going to be those that havebeen most widely distributed, by email, IM, RSS, whatever, and so willalmost certainly be a few hours behind. So I wouldn't expect the headlines to match the most popular? I don't have any desire to highlight any hidden agendas the BBC's editorial staff might have (although I guess it can), but more from an interest in how in touch are the BBC with what the public actually reads and cares about compared to what they think we do.Why would the BBC want to be in touch with spammers? I say this because ofthe story some months back of the BBC's MSN charging
Re: [backstage] BBC Touch
personally, I think it's a fabulous experiment in data. from a news angle? vast amounts of news consumption is people who don't click a link or read a single story - they go to the homepage to check if anything 'important' has happened (usually not). That's editorship, which is different from journalism. But then again, that caveat only really applies to the homepage, so I think Chris' ideas could be more appropriate to a particular index (business, politics etc) where (I'd guess...) far fewer users adopt this nothing big happened - great, I can bugger off qick scanning behaviour. one tip: you can get far better term extraction if you run two or more term extractors (I like Yahoo!'s - and it's an API) over the same story copy and then only use those terms with two or more matches. You get far less noise for not much more effort (Chris Sizemore's idea... not mine...) On 01/11/06, Chris Riley [EMAIL PROTECTED] wrote: Hi Matthew, Thanks for your feedback. On the Pakistan thing, I'm not sure, it was just when I glanced at it when writing the mail that stuck in my mind. So you're probably right :o) On your second point about the editorial, completely agree that they shouldn't be ordering stories by popularity (I think I mention in the about page that they shouldn't be striving for 100% or something), but I do still think it is fair to say stories they want us to read about, a story is more likely to be read if the editor has chosen to make it a headline on the news homepage wouldn't you say? As for the popular feed and the data it provides being accumulated with a certain degree of lag between new news and what we've been reading over lunch, yeah, of course you're always going to get that issue, and I completely understand where you're coming from. Similarly saying that its the BBC's ability to write a good headline that's being measured is a fair comment, but I don't believe it provides an unfair popular bias to a story. In part that's why I introduced the automatic subject extraction, to try and help get through that issue and delve into the actual subjects we're choosing to click through (and so show some interest in) vs. those we don't. That way even if you don't read the full story, clicking through registers your interest in the subject. You'll never be able to get away from the fact the journalists are going to write captivating headlines, but if they are consistently doing that, with the main subject in the headline, I'm happy with that. Ultimately its an experiment with the data, and if someone can make use of the data in a new way thanks to my efforts, I'm happy with that. Cheers, Chris On 11/1/06, Matthew Somerville [EMAIL PROTECTED] wrote: Chris Riley wrote, reordered slightly: In particular I think its useful for highlighting issues the public care more about. For instance a couple of says ago whilst Pakistan was the headline, most of us were reading the climate change story. Are you sure Pakistan was the headline? The climate change story became a subheading just after midnight on the 29th, became the main headline around 07:50am, and stayed there, as far as I can tell, right through the 29th and 30th October, until 04:05 on the 31st October when the main headline became the Prince Charles/Pakistan story for around 15 minutes (data from my front page archive: http://www.bbc.co.uk/homearchive/ and my news archive). On the web page you'll see subjects they want us to read about vs. what we're actually reading about for the past 24 hours, and past 2 weeks. they want us to read? That's not the point of the editorial (by which I mean the ordering of stories on the front page) at all, in my view. I have BBC news in my RSS reader, so that gives me the latest news. I click on the ones I want to read, but that shouldn't affect in any way which ones the BBC decide are important. They hopefully weight stories by more than popularity, otherwise all the stories would be about celebrities and kittens? :-) What your site measures (presuming the popular feed goes on page views, which seems likely) is which stories have been clicked on, not read. I frequently click a headline if it sounds interesting, read the first paragraph, decide it isn't or I already know the story, and close the page. If lots of people are like that, then that makes that story a popular story even though it isn't at all. So what you're actually measuring is how good BBC headlines are at getting people to click through. Similarly, if a BBC post gets linked to from Slashdot or Boing Boing, it will almost certainly become a most popular link. But that doesn't mean it is most popular in terms of the what we're actually reading about, just that lots of people read those sites and click links, realise the first paragraph tells them all they need to know, and that's it. Most emailed would perhaps be a better XML feed to use than Most popular, as then at least people have gone out of
