4 ms·

With respect, Google Reader was used by a tiny fraction of the internet population, despite being operated by the biggest online brand for 8 years. The majority

With respect, Google Reader was used by a tiny fraction of the internet population, despite being operated by the biggest online brand for 8 years. The majority of people opted to get their news from sites and services which used some combination of human and computer powered filtering. I agree the Google Reader model is fantastic for information junkies (in a good way :) with relatively advanced technical skills; but empirically, it didn't work for the mass market Twitter is going after. So I don't see Edgerank as a band-aid fix, but a fundamental requirement for a large-scale system like this. Because (a) on the demand side, recommendations are what people actually want - let computers do what they're good at and reduce the human tedium of sifting through inane FourSquare checkins etc; (b) on the supply side, relevant native ads inside a recommendations stream are the business model. For these reasons, I disagree with OP when he says "Twitter should not abandon its unfiltered feed". I know it will cause shock, shock, when they do it, but they will do it because it's so fundamental to their objectives as a mainstream consumer service.

manicbovine 13y ago
I agree with you, but I think there is at least three related usability issues that made Google Reader less attractive to users. Namely, Google Reader required multiple tools to use efficiently. It also required users to have a basic understanding of RSS feeds. Finally, users had to cope with multiple information formats -- possibly hundreds. Here's just one example. Recently, a popular web comic changed its feed to contain thumbnails, links and ads -- requiring the user to click through in order to see the comic. Previously it just syndicated large, readable versions of the comic. That makes a lot of sense to me, but I have to wonder if a less experienced user might feel that something broke. This leads me to consider how it must feel for a "regular user" might feel or react while browsing between feeds and coping with information provided in so many different formats. Some feeds have pictures, some don't, some have complete articles, others just send out links and ads. Maybe I'm giving myself a false memory, but I can remember a feeling of cognitive friction while switching between feeds. I'm an information junkie and I'm technologically literate, so I dealt with it. I'm saying RSS would possibly fair better if the presentation format were standardized across sites. Maybe a simple style manual would unite enough content providers around a single format? The problem is actually worse with the popular content providers. They each seem to redefine the layout of an RSS feed, whereas the long tail of bloggers seem to coalesce around a very basic format.
- mmahemoff 13y ago
  Yes, you have a good point about consistency. That's one of several reasons the siloed equivalents of RSS (Twitter, Facebook, etc) "won". There's a consistent appearance, consistent metadata, and consistent interactions when it comes to liking and commenting; that kind of interaction was never standardised with RSS, despite some basic attempts. The best you could do was "star it on Google Reader" as a de facto standard. This is really orthogonal to the question of stream filtering, I maintain that's still an issue for RSS aggregators even though as you point out, there are other usability concerns too.
ProblemFactory 13y ago
> So I don't see Edgerank as a band-aid fix, but a fundamental requirement for a large-scale system like this. Because (a) on the demand side, recommendations are what people actually want - let computers do what they're good at and reduce the human tedium of sifting through inane FourSquare checkins etc I can't speak for everyone, but I believe the Google Reader model would work for most people. The interface is entirely based on the source, not the post content, the keywords, or an algorithmically derived importance. There is nothing technical about it: you notice who posts interesting content and who does not. There is no need for automatically ranking the importance of two posts from separate sources, people are very good at deciding who they find interesting. Edgerank or proposed twitter feeds of the future are based on the idea of algorithmically figuring out what is the most important news for you right now. But if you don't care about learning about the "news" two weeks later, then a source-based manual filtering would let the user do that far better than a ML algorithm can. And if it's out of date in two weeks, it wasn't worth reading at all.

manicbovine 13y ago

I agree with you, but I think there is at least three related usability issues that made Google Reader less attractive to users. Namely, Google Reader required multiple tools to use efficiently. It also required users to have a basic understanding of RSS feeds. Finally, users had to cope with multiple information formats -- possibly hundreds. Here's just one example. Recently, a popular web comic changed its feed to contain thumbnails, links and ads -- requiring the user to click through in order to see the comic. Previously it just syndicated large, readable versions of the comic. That makes a lot of sense to me, but I have to wonder if a less experienced user might feel that something broke. This leads me to consider how it must feel for a "regular user" might feel or react while browsing between feeds and coping with information provided in so many different formats. Some feeds have pictures, some don't, some have complete articles, others just send out links and ads. Maybe I'm giving myself a false memory, but I can remember a feeling of cognitive friction while switching between feeds. I'm an information junkie and I'm technologically literate, so I dealt with it. I'm saying RSS would possibly fair better if the presentation format were standardized across sites. Maybe a simple style manual would unite enough content providers around a single format? The problem is actually worse with the popular content providers. They each seem to redefine the layout of an RSS feed, whereas the long tail of bloggers seem to coalesce around a very basic format.

mmahemoff 13y ago
Yes, you have a good point about consistency. That's one of several reasons the siloed equivalents of RSS (Twitter, Facebook, etc) "won". There's a consistent appearance, consistent metadata, and consistent interactions when it comes to liking and commenting; that kind of interaction was never standardised with RSS, despite some basic attempts. The best you could do was "star it on Google Reader" as a de facto standard. This is really orthogonal to the question of stream filtering, I maintain that's still an issue for RSS aggregators even though as you point out, there are other usability concerns too.

ProblemFactory 13y ago

> So I don't see Edgerank as a band-aid fix, but a fundamental requirement for a large-scale system like this. Because (a) on the demand side, recommendations are what people actually want - let computers do what they're good at and reduce the human tedium of sifting through inane FourSquare checkins etc I can't speak for everyone, but I believe the Google Reader model would work for most people. The interface is entirely based on the source, not the post content, the keywords, or an algorithmically derived importance. There is nothing technical about it: you notice who posts interesting content and who does not. There is no need for automatically ranking the importance of two posts from separate sources, people are very good at deciding who they find interesting. Edgerank or proposed twitter feeds of the future are based on the idea of algorithmically figuring out what is the most important news for you right now. But if you don't care about learning about the "news" two weeks later, then a source-based manual filtering would let the user do that far better than a ML algorithm can. And if it's out of date in two weeks, it wasn't worth reading at all.