news with the added flavor of time, and the spicy kick of impact
In a similar vein to yesterday's post, an interesting question to ask is how do news publications impact events? The story that spurred this thought was Steve Jobs' denouncement of DRM last February (NPR story: "Apple's Jobs Argues Against Song Protections"). At the time, all of the four major record labels (Sony BMG, Warner Music Group, EMI, and Universal Music Group -- Wikipedia: for when the brain just can't remember) were resolute about keeping DRM on their digital downloads, say, from Apple's iTunes Music Store. As of the past few months, all four have gone DRM-free, either through Amazon's mp3 store or the iTunes Music Store or some other online retailer (stories confirming these somewhere out there, but I remember clearly hearing about this on TWiT and Buzz Out Loud).
The point is, many people did not (and, I guess, still don't) know what DRM was. Tech-oriented people who knew about it and dealt with it almost unanimously didn't like it (and, I think, at best, were against it but didn't really care because, for example, while they can't only play their purchased downloads from the iTunes Music Store in iTunes on a limited number of machines and on synced iPods, that's all they do so they're essentially unchained). Some of these people spoke out against DRM, and their advocates, like Jobs eventually, demanded that the record labels go DRM-free. Now, assuredly there was news coverage on DRM in the popular places (not just special-interest sources like many of the tech shows/podcasts I listen to), and this had to have had some effect on the people listening. At best, it would have educated them about DRM and probably made them not like it (or, perhaps, if they didn't do much with downloaded music, simply shrug their shoulders and ask, "who cares?"). Perhaps it made some of them speak out, put pressure on the record labels to get rid of DRM. Maybe because of this fire, fanned at least a little by news sources, the labels gave up DRM quicker.
Now, is this a common trend? Does the presence of a news story -- that is, essentially getting the word out to many more people at a time -- impact future events? It must, I think, because people talk about it in the political sphere, how news programs show only a certain side to a story and cause those who watch it to become one-sided in their opinions, perhaps causing things such as the President's embarrassingly low acceptance rate. Though those discussions are usually relegated to political topics. I'm more interested in all topics. The above story related to tech. What about science and economy and medicine?
Would it be possible to track what stories are in the news, be it in print newspapers, their online counterparts, web-only sources (CNET, etc.), personal blogs (as people certainly voice their opinions on news stories and simply publicize the news there)? Could we see if there is a definite action correlating to the presence of news on the topic? It would be hard to test this, though, as you can't have repeated trials (we can't put back DRM and tell everyone to forget what happened, and this time don't put it in the news at all, and then see how long it takes, if ever, for the labels to put out DRM-free music). We would need to be able to compare different stories and events to find ones that are similarly interesting or have the same expected outcome, the action believed to be coming will occur in the same time frame.
I'm not sure this is possible, as there are many, many factors. One would be the topic -- pop culture is more, well, popular than technology news (unless it's about a phone from Apple, *ahem*), another would be its "difficulty" -- what percent of the population even understand the story and its implications? That McDonald's is closing down (no, it isn't really) is an "easy" story, but the problem of DRM-ed music is a much more "difficult" story, as many do not know what DRM is, for example. You would also have to determine a news outlet's popularity and influence. The New York Times has a huge influence; many, many people (thousands? millions?) read it every day. The personal blog of some unknown who talks about what happens in the world of tech and has about 100 readers does not have much influence. The number of outlets who publish a story on a topic is a measure of how popular or important the media thinks it to be, as well as for how long that story has a presence in the publication (and where... front page or last section?).
I know there are websites that track the news over one slice of time (i.e, right now), arranging the stories in a grid, where they are colored according to subject, shaded I think according to age, and sized depending on popularity. There are also sites that archive the news and do stuff with it (such as the Newsblaster project at Columbia, though it uses the news for natural language processing purposes, not for actual news analysis). What if we combined these two ideas? What if we took an archive of all the news stories from all the sources, and tracked their presence over time? Instead of a simple two-dimensional grid for one slice of time, we could have a three-dimensional shape, with the third axis being time. We could then produce a time line for any story, seeing how popular it was over time and whether it was covered at all. We could also plot when certain events happened. For example, we could plot Jobs' denouncement of DRM as one, and when each of the major record labels removed DRM as additional events. Then, we could see if these events have a correspondence with the structure of the news. Maybe the record labels made their moves after news reached a peak, or after a certain amount of time (though a simple time argument is difficult to make, for we don't know how long it would have taken them on their own to remove DRM). Though there would be an issue about what exactly an event is. Even some news articles could be events, if they broke certain stories or revealed key facts or whatnot. But that isn't an issue to worry about now.
While it would be an interesting project to undertake, I'm not sure what kind of use it would be. Sure, we could perhaps answer the eternal question -- are bloggers really journalists?? But we could also see what kind of impact the news has on events of all kinds. Maybe if we see a very high correlation, journalists would be more careful about what they say or do. But this isn't to say they should as it is. Or maybe it's just a chance to play with pretty graphs and pictures (and databases!). Whoosh!
The point is, many people did not (and, I guess, still don't) know what DRM was. Tech-oriented people who knew about it and dealt with it almost unanimously didn't like it (and, I think, at best, were against it but didn't really care because, for example, while they can't only play their purchased downloads from the iTunes Music Store in iTunes on a limited number of machines and on synced iPods, that's all they do so they're essentially unchained). Some of these people spoke out against DRM, and their advocates, like Jobs eventually, demanded that the record labels go DRM-free. Now, assuredly there was news coverage on DRM in the popular places (not just special-interest sources like many of the tech shows/podcasts I listen to), and this had to have had some effect on the people listening. At best, it would have educated them about DRM and probably made them not like it (or, perhaps, if they didn't do much with downloaded music, simply shrug their shoulders and ask, "who cares?"). Perhaps it made some of them speak out, put pressure on the record labels to get rid of DRM. Maybe because of this fire, fanned at least a little by news sources, the labels gave up DRM quicker.
Now, is this a common trend? Does the presence of a news story -- that is, essentially getting the word out to many more people at a time -- impact future events? It must, I think, because people talk about it in the political sphere, how news programs show only a certain side to a story and cause those who watch it to become one-sided in their opinions, perhaps causing things such as the President's embarrassingly low acceptance rate. Though those discussions are usually relegated to political topics. I'm more interested in all topics. The above story related to tech. What about science and economy and medicine?
Would it be possible to track what stories are in the news, be it in print newspapers, their online counterparts, web-only sources (CNET, etc.), personal blogs (as people certainly voice their opinions on news stories and simply publicize the news there)? Could we see if there is a definite action correlating to the presence of news on the topic? It would be hard to test this, though, as you can't have repeated trials (we can't put back DRM and tell everyone to forget what happened, and this time don't put it in the news at all, and then see how long it takes, if ever, for the labels to put out DRM-free music). We would need to be able to compare different stories and events to find ones that are similarly interesting or have the same expected outcome, the action believed to be coming will occur in the same time frame.
I'm not sure this is possible, as there are many, many factors. One would be the topic -- pop culture is more, well, popular than technology news (unless it's about a phone from Apple, *ahem*), another would be its "difficulty" -- what percent of the population even understand the story and its implications? That McDonald's is closing down (no, it isn't really) is an "easy" story, but the problem of DRM-ed music is a much more "difficult" story, as many do not know what DRM is, for example. You would also have to determine a news outlet's popularity and influence. The New York Times has a huge influence; many, many people (thousands? millions?) read it every day. The personal blog of some unknown who talks about what happens in the world of tech and has about 100 readers does not have much influence. The number of outlets who publish a story on a topic is a measure of how popular or important the media thinks it to be, as well as for how long that story has a presence in the publication (and where... front page or last section?).
I know there are websites that track the news over one slice of time (i.e, right now), arranging the stories in a grid, where they are colored according to subject, shaded I think according to age, and sized depending on popularity. There are also sites that archive the news and do stuff with it (such as the Newsblaster project at Columbia, though it uses the news for natural language processing purposes, not for actual news analysis). What if we combined these two ideas? What if we took an archive of all the news stories from all the sources, and tracked their presence over time? Instead of a simple two-dimensional grid for one slice of time, we could have a three-dimensional shape, with the third axis being time. We could then produce a time line for any story, seeing how popular it was over time and whether it was covered at all. We could also plot when certain events happened. For example, we could plot Jobs' denouncement of DRM as one, and when each of the major record labels removed DRM as additional events. Then, we could see if these events have a correspondence with the structure of the news. Maybe the record labels made their moves after news reached a peak, or after a certain amount of time (though a simple time argument is difficult to make, for we don't know how long it would have taken them on their own to remove DRM). Though there would be an issue about what exactly an event is. Even some news articles could be events, if they broke certain stories or revealed key facts or whatnot. But that isn't an issue to worry about now.
While it would be an interesting project to undertake, I'm not sure what kind of use it would be. Sure, we could perhaps answer the eternal question -- are bloggers really journalists?? But we could also see what kind of impact the news has on events of all kinds. Maybe if we see a very high correlation, journalists would be more careful about what they say or do. But this isn't to say they should as it is. Or maybe it's just a chance to play with pretty graphs and pictures (and databases!). Whoosh!