Online Personalization Creates Echo Chamber to Affirm Biases(nytimes.com)
nytimes.com
Online Personalization Creates Echo Chamber to Affirm Biases
https://www.nytimes.com/2011/05/29/technology/29stream.html
10 comments
An alternate perspective is available at http://glinden.blogspot.com/2011/05/eli-pariser-is-wrong.htm...
I'm not convinced of this rebuttal. The problem is that these systems tend only to improve serendipity within the comfort zone. Greg's One True Scotsman argument that a "well-designed" system would of course be better is not convincing either.
Naive systems optimize for things you already like, or variations on things you already like. The system doesn't need to present a broad array of ideas. Presenting you with something that causes a strong negative reaction is considered a bad thing.
Newspapers have the same problem to a lesser degree: I know folk who read the Weekly Standard and who wouldn't touch the NYT with a bargepole, and vice versa.
We just have many more, smaller echo chambers. People will always have to make the conscious effort to give strange views a fair hearing.
Naive systems optimize for things you already like, or variations on things you already like. The system doesn't need to present a broad array of ideas. Presenting you with something that causes a strong negative reaction is considered a bad thing.
Newspapers have the same problem to a lesser degree: I know folk who read the Weekly Standard and who wouldn't touch the NYT with a bargepole, and vice versa.
We just have many more, smaller echo chambers. People will always have to make the conscious effort to give strange views a fair hearing.
"[Pariser] went out of his way to “friend” people with conservative politics. When he didn’t click on their updates as often as those of his like-minded contacts, he says, the system dropped the outliers from his news feed."
There's the rub. The Internet, among other things, has created a flood of data into people's lives. Filtering is essential. Pariser's built-in mental filter took over. I understand the premise of what's being said here. But it's more complicated than this. If an article/story/data in general is uninteresting or repellent enough to someone that it's ignored, did it really help that it was on the page briefly as it's skimmed past?
I think the problem is that it _automatically_ dropped things from his feed. It might be good to offer a way to turn filtering off entirely and leave it up to the person. I can't imagine anyone would really want this for long, though.
There's the rub. The Internet, among other things, has created a flood of data into people's lives. Filtering is essential. Pariser's built-in mental filter took over. I understand the premise of what's being said here. But it's more complicated than this. If an article/story/data in general is uninteresting or repellent enough to someone that it's ignored, did it really help that it was on the page briefly as it's skimmed past?
I think the problem is that it _automatically_ dropped things from his feed. It might be good to offer a way to turn filtering off entirely and leave it up to the person. I can't imagine anyone would really want this for long, though.
Recommendation engines tend to suggest more variety than what your friends would. The whole argument rests upon silly anti-tech fears.
The vast majority of people get more political diversity in their Amazon recommendations than the NYT can produce.
The vast majority of people get more political diversity in their Amazon recommendations than the NYT can produce.
I think a big problem here is the scaling of online communities(there was a great article out on this a couple years ago). As homogeneous online communities grow their perceived quality goes down in part due to relatively heterogeneous members disrupting the signal to noise ratio. Sites like facebook(and I assume even reddit) get around this scaling problem by partitioning members off into homogeneous sets. This problem is related to the fact that there is a dialogue happening. Dialogues within heterogeneous populations degrades into culture war. So partition the user base via filters, create structure that mediates the dialogue/interaction, or create niche communities tailored to a homogeneous community.
The old article you mention might be "A Group Is Its Own Worst Enemy" by Clay Shirky?
http://shirky.com/writings/group_enemy.html
http://shirky.com/writings/group_enemy.html
That's the one, thanks.
There are not black & white opinions.
The issue is that the goal recommendation systems in commercial sites is [generally] profits. So, Netflix will not risk to recommend a film outside some confidence level.
In my experience, closer [taste] friends works better than recommendation systems, since they can see outside the typical statistics.
The issue is that the goal recommendation systems in commercial sites is [generally] profits. So, Netflix will not risk to recommend a film outside some confidence level.
In my experience, closer [taste] friends works better than recommendation systems, since they can see outside the typical statistics.
The issue is that the goal recommendation systems in commercial sites is [generally] profits. So, Netflix will not risk to recommend a film outside some confidence level
Sometimes it's actually irrational, though. On Pandora, for instance, I find my stations keep playing a relatively small set of songs over and over again (I believe a lot of other people have this problem). From a profit-motive point of view a better algorithm would be exploring the edges of my personally indicated taste space a bit more often but their algorithm is apparently a bit too dumb for this.
The bigger problem, I think, is not insularity caused by algorithms but insularity caused by personal choice. People with political ideology X don't miss out on reading stuff by people with political ideology Y because google doesn't suggest it for 'em, they miss out on reading stuff by people with political ideology Y because they don't want to read it. People live in information bubbles created by themselves, not by matching algorithms.
And the problem is not just that people refuse to step outside their own comfort zone. If I have ideology X and start reading political blogs by the opposing ideology, then I'll quickly get annoyed, because I'll find that these blogs largely consist not of "Here are some sensible arguments in favour of ideology Y" but rather of "hurr durrr people with ideology X are stoooopid and evil! Hey, did you know that people with ideology X say they believe X but in fact they really believe P, Q and R? What a bunch of disgusting freaks!" And so forth.
Sometimes it's actually irrational, though. On Pandora, for instance, I find my stations keep playing a relatively small set of songs over and over again (I believe a lot of other people have this problem). From a profit-motive point of view a better algorithm would be exploring the edges of my personally indicated taste space a bit more often but their algorithm is apparently a bit too dumb for this.
The bigger problem, I think, is not insularity caused by algorithms but insularity caused by personal choice. People with political ideology X don't miss out on reading stuff by people with political ideology Y because google doesn't suggest it for 'em, they miss out on reading stuff by people with political ideology Y because they don't want to read it. People live in information bubbles created by themselves, not by matching algorithms.
And the problem is not just that people refuse to step outside their own comfort zone. If I have ideology X and start reading political blogs by the opposing ideology, then I'll quickly get annoyed, because I'll find that these blogs largely consist not of "Here are some sensible arguments in favour of ideology Y" but rather of "hurr durrr people with ideology X are stoooopid and evil! Hey, did you know that people with ideology X say they believe X but in fact they really believe P, Q and R? What a bunch of disgusting freaks!" And so forth.
With friends, we learn to accept/reject their recommendations based on our experience with their recommendations in the area they are recommending. We filter the filters.
Yeah, I often find myself intentionally stopping this occasionally to allow growth into areas that I don't currently like, but could. This includes odd foods, hobbies, activities, etc.
Usually try to give it 4-5 "goes" at the novel activity to determine if they're good at picking out things that at first aren't interesting.
Usually try to give it 4-5 "goes" at the novel activity to determine if they're good at picking out things that at first aren't interesting.
You can watch the TED video here: http://www.ted.com/talks/lang/eng/eli_pariser_beware_online_...
This has been a fear since the earliest days of online communities. I don't think actual behavior has confirmed that it occurs more often online than elsewhere; if it did, Pariser would have more hard data (rather than just-so anecdotes and hand-wringing about potential-risks) by this time.
People who want to be insulated could always achieve it; however people also like novelty and (to a certain extent) challenges to their own views. (People even enjoy being outraged; I suspect right-wing-radio has as many listeners thinking 'ohmygod what an idiot' as 'finally someone who tells it like it is'.)
Online forums make it cheaper, easier, and even socially safer to incrementally explore alternative viewpoints without precommitting, via a public show of group-loyalty, to a standard school of thought.
I suspect the net effect is strongly in the direction of more awareness and respect for a wide variety of viewpoints, despite the sturm and drang of certain flamefests and partisan outlets. (And even despite the success of Pariser's own organization MoveOn, and its siblings across the political spectrum, with simpleminded outrage-driven us-vs-them fundraising campaigns.)
However, I'd like to see data or a well-designed experiment; none of the coverage of Pariser's book has yet suggested a qualitatively-rigorous case is made within.
People who want to be insulated could always achieve it; however people also like novelty and (to a certain extent) challenges to their own views. (People even enjoy being outraged; I suspect right-wing-radio has as many listeners thinking 'ohmygod what an idiot' as 'finally someone who tells it like it is'.)
Online forums make it cheaper, easier, and even socially safer to incrementally explore alternative viewpoints without precommitting, via a public show of group-loyalty, to a standard school of thought.
I suspect the net effect is strongly in the direction of more awareness and respect for a wide variety of viewpoints, despite the sturm and drang of certain flamefests and partisan outlets. (And even despite the success of Pariser's own organization MoveOn, and its siblings across the political spectrum, with simpleminded outrage-driven us-vs-them fundraising campaigns.)
However, I'd like to see data or a well-designed experiment; none of the coverage of Pariser's book has yet suggested a qualitatively-rigorous case is made within.
On a somewhat related note, I was looking at health insurance the other day. Wanting to make sure I remembered the terms correctly, I Googled, "how do health insurance deductibles work". By the time I typed the word 'insurance' my question was #1 in the suggestions pop-up. I found this strangely reassuring.
Later, I realized that it may well be at the top solely because I searched on the same exact term the last time that I was looking at health insurance, and Google remembered.
Later, I realized that it may well be at the top solely because I searched on the same exact term the last time that I was looking at health insurance, and Google remembered.
Surprise, surprise... The largest of traditional newspapers has a problem with one of the biggest drivers behind online media: that you can finally choose what you read.