An Outsiders View of Digg: There is a Flaw in the System
Since becoming fascinated with digg.com I've spend some time there.
Just to re-cap for the sake of this article I'll try and explain how Digg works. Digg is what is known as a "social bookmarking" site. The concept is that we all keep a "favorites" or "bookmark" list in our browsers so why not share these lists and index or rate the bookmarks in some way. Digg consists of entries of bookmarks in the form of news headlines and these entries are submitted by users of the site. The headline is a link to something on the internet and then the user can provide a small description as well. Once the entry has been submitted, it can be voted on (dugg) by users. Once an entry receives enough votes it will be promoted to the front page. The front page is the output of all this activity, viz., a current and evolving list of the hottest links on the internet. Each user's activity is tracked and a rating generated mostly based on the percentage of your entries that become promoted. I've so far had 3 entries promoted and my rating is 2301 (not bad!). The community is so tuned in that its very difficult to come across something interesting on the internet that has not already been submitted (more on this later as it is the crux of what I want to talk about).
There are basically three roles available in interacting with digg:
1. You can submit entries (Journalist Role)
2. You can vote and comment on entries (Editor role)
3. You can read the front page (Reader Role)
I have been engaging in Journalist and Editor roles because I wanted to try and get a better understanding of what I think is probably the coolest experiment happening right now on the internet. I also think that what is happening at Digg will soon become standard practice, indeed, AOL has now relaunched the Netscape Portal as a Digg clone. In related news, AOL has offered the top Digg users $1000.00 a month to quit Digg and use their site.
Ok, so now for my thoughts on this. I think the concept is amazing when compared to traditional news methods because the output is much more organic than any new organization could ever hope to achieve. In addition, the speed at which information is obtained is startling. Since I'm now tuned into to Digg, I'm seeing stories that will show up on cnn.com 3 to 4 days later. It is without a doubt the best way to get news of everything the quickest.
The implementation is awesome. There is Digg Spy which is a live scrolling page of all the entries that are coming in or being voted on. You can customize to show any and all of the following: new entries, recently voted on entries, recently commented, promoted, or upcoming. The Spy interface is where the inside work gets done. Once a story hits the front page it takes on a new life. It is really difficult to get a story up to around 50 votes but once its promoted then the votes rapidly increase.
My first story to get "dugg" was submitted before lunch. When I came back and checked it was on the front page and had 800 votes. I currently have a 10% success ratio. The big mystery is: what does it take to get a story promoted? Thinking about this raises questions about internet memes and marketing and human nature. Very interesting. From what I've seen its half what the story is, how you write your headline, how you write your description, when you submit it, your popularity amongst the community and half mystery.
There is a flaw in this system however. The flaw comes from human nature and I'm not sure that any kind of software fix to Digg could change this. The core idea is that the resulting output of news stories should be generated by a large community of users. This is essential to what makes Digg different from traditional news organizations in that they have a small editorial staff. Digg is supposed to be more "democratic" you might say.
Because of the rating system and the addicting nature of trying to get your own stories promoted, personal interests starting taking precedence. This is to be expected because the "staff" are not paid; they do it to do it. I have to say there is a certain high to getting a story promoted and this provides the motivation to keep doing it. Doing so generally means you were the first person to discover something, you wrote a good headline and the something you discovered was just confirmed to not be lame. There is a feeling of engaging in the cutting edge of the net ... or something.
You can imagine then that the problem is that people will go to any length to get promoted stories. Cheating is handled through how Digg is coded so that's not yet a big problem although sites have popped up that have tried. The idea is to build a community of users on a separate site that all vote on each others entries. Generally they haven't been too successful.
What has been successful is that a core group of diggers has formed, a clique if you will, that will all vote on each others entries. Other people have written about this that in fact most of the front page stores come from a small subset of Diggers and so the end result is a more traditional news editorial system than rather than the cutting edge social experiment it is thought to be. My thought is that it is somewhere in the middle. Yes, there is a core group who vote on each others entries, but they still need non-core people to vote too. They do get an advantage because they can get to 50 votes much easier. There are some users in the core group with success percentage in the 90's. That is just not possible for a regular user like myself.
The second part of the flaw is duplicate entries and I'll show how this relates to the core group problem. The instructions for submitting an entry are clearly displayed each time you do it:
* Quality Content: Is your story on topic? Make sure your story is appropriate for the topic you're submitting it to.
* Link Directly to the Source: Save people time by linking directly to the original news story.
* Search First: Avoid duplication by searching to ensure nobody has submitted the same story.
* Be Descriptive: You're the story's editor, so explain what it is and why it's cool.
The "Search First" rule is the big problem in that it is very rarely done and Digg Spy is full of duplicate stories. The Editors can label ("Bury") something as a dupe and if it gets enough of those it disappears. I'm seeing that there are too many duplicates and the editorial system is not working. If you try and submit a duplicate story the system warns you and tells you to vote for the existing story. Because you can ignore that and just go ahead and submit and because the community of editors can't stay on top of all the duplicates, this rule has to be followed on the honor system. Human nature dictates that the honor system generally doesn't work.
In my view, the basic honor system should be that someone makes a genuine effort to search and make sure they are not submitting a duplicate. If they find it is a duplicate, go ahead and give a vote to the existing story. This to me reaffirms the basic idea that Digg is about finding cool stuff first.
This is not what happens. People submit duplicates constantly. This causes the situation where you submit a story and it is getting votes slowly. A few hours later someone will submit the same story (rather than voting on the existing one and bringing it more attention) with a different headline and it catches on and becomes a front page story.
There is not much that can be done about this. The first time it happened to me, I posted a comment asking why, if I had submitted this story 3 days before, is everyone voting for this new one and they didn't for mine. The answer was that my headline sucked and that I shouldn't whine about it. Instead, they said, I could simply make a comment that this story was a duplicate, mark it as a duplicate and provide a link to my entry.
The next time this happened I followed the advice and put a link to my original entry of the story. I still got blamed for whining and people marked my comment as bad. I continued the discussion and found out that the person who submitted the duplicate was in the core group and that he has a large following of people who vote on anything he submits. You'd think someone so experienced would know how to search before entering. And of course he does and I'm sure he did. I'm sure he saw my story entered 3 days previous with 30 votes and figured he could resubmit it and get a front page with his army of voters. And he was right - it worked, the story went front page. Here are the two different headlines (I won't tell you who wrote which one - the point is that they are very similar and so the headline couldn't be why my story was growing slower):
"Role Playing Game Inspirational Posters"
"Gamer Inspirational Posters (Photos)"
Can you guess which one made it the front page?
It seems obvious now how the core group have such a high percentage of promoted stories. Probably many of their entries are duplicates but get promoted anyway.
Whenever you have an honor code you will find people exploiting it so in that way Digg is microcosm of the world. Since it is software, there may be a coding solution to this. My suggestion is that when the system discovers that the URL you are submitting has already been submitted that it does not allow the entry. This would solve some of the problem to a small degree although not all of it because the same story can usually be found at many different URLs.
My question to the Digg community is: do you want to operate with a code or will personal success win out over the goal of the experiment?
Speaking of honor, I am going to submit this story into Digg. By doing so I will be violating a social code which is that it is is not cool to submit stories you wrote on your own blog. I guess the story should be "found" be someone else. But in the spirit of how the honor code is broken at Digg I will submit it. Below is a link to my Digg story that you can click and vote for it. Let's call this my own little meta-experiment and see what happens.
|Digg This|
Just to re-cap for the sake of this article I'll try and explain how Digg works. Digg is what is known as a "social bookmarking" site. The concept is that we all keep a "favorites" or "bookmark" list in our browsers so why not share these lists and index or rate the bookmarks in some way. Digg consists of entries of bookmarks in the form of news headlines and these entries are submitted by users of the site. The headline is a link to something on the internet and then the user can provide a small description as well. Once the entry has been submitted, it can be voted on (dugg) by users. Once an entry receives enough votes it will be promoted to the front page. The front page is the output of all this activity, viz., a current and evolving list of the hottest links on the internet. Each user's activity is tracked and a rating generated mostly based on the percentage of your entries that become promoted. I've so far had 3 entries promoted and my rating is 2301 (not bad!). The community is so tuned in that its very difficult to come across something interesting on the internet that has not already been submitted (more on this later as it is the crux of what I want to talk about).
There are basically three roles available in interacting with digg:
1. You can submit entries (Journalist Role)
2. You can vote and comment on entries (Editor role)
3. You can read the front page (Reader Role)
I have been engaging in Journalist and Editor roles because I wanted to try and get a better understanding of what I think is probably the coolest experiment happening right now on the internet. I also think that what is happening at Digg will soon become standard practice, indeed, AOL has now relaunched the Netscape Portal as a Digg clone. In related news, AOL has offered the top Digg users $1000.00 a month to quit Digg and use their site.
Ok, so now for my thoughts on this. I think the concept is amazing when compared to traditional news methods because the output is much more organic than any new organization could ever hope to achieve. In addition, the speed at which information is obtained is startling. Since I'm now tuned into to Digg, I'm seeing stories that will show up on cnn.com 3 to 4 days later. It is without a doubt the best way to get news of everything the quickest.
The implementation is awesome. There is Digg Spy which is a live scrolling page of all the entries that are coming in or being voted on. You can customize to show any and all of the following: new entries, recently voted on entries, recently commented, promoted, or upcoming. The Spy interface is where the inside work gets done. Once a story hits the front page it takes on a new life. It is really difficult to get a story up to around 50 votes but once its promoted then the votes rapidly increase.
My first story to get "dugg" was submitted before lunch. When I came back and checked it was on the front page and had 800 votes. I currently have a 10% success ratio. The big mystery is: what does it take to get a story promoted? Thinking about this raises questions about internet memes and marketing and human nature. Very interesting. From what I've seen its half what the story is, how you write your headline, how you write your description, when you submit it, your popularity amongst the community and half mystery.
There is a flaw in this system however. The flaw comes from human nature and I'm not sure that any kind of software fix to Digg could change this. The core idea is that the resulting output of news stories should be generated by a large community of users. This is essential to what makes Digg different from traditional news organizations in that they have a small editorial staff. Digg is supposed to be more "democratic" you might say.
Because of the rating system and the addicting nature of trying to get your own stories promoted, personal interests starting taking precedence. This is to be expected because the "staff" are not paid; they do it to do it. I have to say there is a certain high to getting a story promoted and this provides the motivation to keep doing it. Doing so generally means you were the first person to discover something, you wrote a good headline and the something you discovered was just confirmed to not be lame. There is a feeling of engaging in the cutting edge of the net ... or something.
You can imagine then that the problem is that people will go to any length to get promoted stories. Cheating is handled through how Digg is coded so that's not yet a big problem although sites have popped up that have tried. The idea is to build a community of users on a separate site that all vote on each others entries. Generally they haven't been too successful.
What has been successful is that a core group of diggers has formed, a clique if you will, that will all vote on each others entries. Other people have written about this that in fact most of the front page stores come from a small subset of Diggers and so the end result is a more traditional news editorial system than rather than the cutting edge social experiment it is thought to be. My thought is that it is somewhere in the middle. Yes, there is a core group who vote on each others entries, but they still need non-core people to vote too. They do get an advantage because they can get to 50 votes much easier. There are some users in the core group with success percentage in the 90's. That is just not possible for a regular user like myself.
The second part of the flaw is duplicate entries and I'll show how this relates to the core group problem. The instructions for submitting an entry are clearly displayed each time you do it:
* Quality Content: Is your story on topic? Make sure your story is appropriate for the topic you're submitting it to.
* Link Directly to the Source: Save people time by linking directly to the original news story.
* Search First: Avoid duplication by searching to ensure nobody has submitted the same story.
* Be Descriptive: You're the story's editor, so explain what it is and why it's cool.
The "Search First" rule is the big problem in that it is very rarely done and Digg Spy is full of duplicate stories. The Editors can label ("Bury") something as a dupe and if it gets enough of those it disappears. I'm seeing that there are too many duplicates and the editorial system is not working. If you try and submit a duplicate story the system warns you and tells you to vote for the existing story. Because you can ignore that and just go ahead and submit and because the community of editors can't stay on top of all the duplicates, this rule has to be followed on the honor system. Human nature dictates that the honor system generally doesn't work.
In my view, the basic honor system should be that someone makes a genuine effort to search and make sure they are not submitting a duplicate. If they find it is a duplicate, go ahead and give a vote to the existing story. This to me reaffirms the basic idea that Digg is about finding cool stuff first.
This is not what happens. People submit duplicates constantly. This causes the situation where you submit a story and it is getting votes slowly. A few hours later someone will submit the same story (rather than voting on the existing one and bringing it more attention) with a different headline and it catches on and becomes a front page story.
There is not much that can be done about this. The first time it happened to me, I posted a comment asking why, if I had submitted this story 3 days before, is everyone voting for this new one and they didn't for mine. The answer was that my headline sucked and that I shouldn't whine about it. Instead, they said, I could simply make a comment that this story was a duplicate, mark it as a duplicate and provide a link to my entry.
The next time this happened I followed the advice and put a link to my original entry of the story. I still got blamed for whining and people marked my comment as bad. I continued the discussion and found out that the person who submitted the duplicate was in the core group and that he has a large following of people who vote on anything he submits. You'd think someone so experienced would know how to search before entering. And of course he does and I'm sure he did. I'm sure he saw my story entered 3 days previous with 30 votes and figured he could resubmit it and get a front page with his army of voters. And he was right - it worked, the story went front page. Here are the two different headlines (I won't tell you who wrote which one - the point is that they are very similar and so the headline couldn't be why my story was growing slower):
"Role Playing Game Inspirational Posters"
"Gamer Inspirational Posters (Photos)"
Can you guess which one made it the front page?
It seems obvious now how the core group have such a high percentage of promoted stories. Probably many of their entries are duplicates but get promoted anyway.
Whenever you have an honor code you will find people exploiting it so in that way Digg is microcosm of the world. Since it is software, there may be a coding solution to this. My suggestion is that when the system discovers that the URL you are submitting has already been submitted that it does not allow the entry. This would solve some of the problem to a small degree although not all of it because the same story can usually be found at many different URLs.
My question to the Digg community is: do you want to operate with a code or will personal success win out over the goal of the experiment?
Speaking of honor, I am going to submit this story into Digg. By doing so I will be violating a social code which is that it is is not cool to submit stories you wrote on your own blog. I guess the story should be "found" be someone else. But in the spirit of how the honor code is broken at Digg I will submit it. Below is a link to my Digg story that you can click and vote for it. Let's call this my own little meta-experiment and see what happens.
|Digg This|
Labels: Technology
0 Comments:
Post a Comment
Links to this post:
Create a Link
<< Home