Digging an elite news source

Dave Winer has a great post about the next step in Digg clones.

This is something I’ve been spending a lot of time thinking about too.

What makes a great selection of links? Does it take a million people? Or only a handful.

I’ve been comparing my link blog to TechMeme and Digg and others enough to know that my links are pretty darn interesting. But there are times when I fall down on the job. When Milan is smiling at me, for instance. Or when I have chores. Or, when I have to speak, like I’m doing this morning at San Francisco State University, or with a bunch of journalists and famous VC’s on Wednesday night at a PRSA shindig in Silicon Valley.

One app in Facebook, FeedHeads, taught me that a small number of people can really pick some killer things. But one problem: that app is often down for me and I find I don’t trust it and it’s not fast like TechMeme is, so I usually default back to TechMeme.

The top shared item view in FeedHeads, though, is really much faster than Digg or TechMeme to get cool stuff up on it. Something about it is really intriguing.

Steve Gillmor has talked with me a lot about this, too. He’s noticed that if he chooses the news reading behaviors of only a handful of people that he can get much better results than if he has a larger group.

Why is that?

I think it’s because we like our news focused. So if we find a news junkie who thinks like us we’ll find that person to have high utility. If you put him in a group of other people his utility will go down and the noise we’ll have to slog through to find a good set of articles that interest us will go up.

That sure explains Digg and its problems to me. I used to love Digg. Now it’s just a stream of noise that largely duplicates what’s coming through my feed reader.

Google Reader tells me that over the past 30 days I’ve read 39,712 items and shared 1,045 items. The thing is that’s probably only about half of the good items cause I can’t read around the clock and have other things to do than just read feeds.

So, lets say I hooked up with five other people who picked the same kinds of items. We could hit nearly 100% of the feeds that I read (and we could add some other feeds). Five people could beat TechMeme. Why? TechMeme is slow. I often put stuff on my link blog before TechMeme gets it. If we had a team reading feeds around the clock we’d regularly beat TechMeme or Google News or Digg or Reddit or TailRank.

And we’d have less noise. At least if you liked the things the five of us pick. If you don’t, then Digg will look better to you.

Anyway, this is a long way of saying that I’m looking forward to seeing what Dave Winer does in this space. There certainly is a lot of great stuff that doesn’t get proper attention on sites like TechMeme, Digg, etc. If this effort helps great stuff get more people to read it, I’m all for it.

Question: what do you think about Digg or TechMeme or Google News or other news aggregation sites? What are you hoping to see?

Comments

  1. I agree that that would be a good idea to an extent. The problem being that no matter how good you think you and your friends are (and I don’t mean you Robert, just talking in general here), there is always going to be interesting stuff going on outside of your clique that you’ll miss because you’re too focused on the feeds you and your friends read.

    I know that you use other methods like blog searches (I saw you shared my post earlier via a google blog search), but it still won’t be able to match the raw processing power of millions of users on Digg. Of course it will also filter out all of the crap that is on Digg, but sometimes you have to take all the bad to find that one gem.

    My point is that a combination of all of these things will work best. I too find my feed reader to be 99% more useful and quicker than Digg, Techmeme, and the like but if I didn’t check Digg every once in a while I would miss some decent stuff.

    That is why I want what I’ve laid out numerous times – for Google Reader to share the ‘read’, ‘starred’, and ‘shared’ data network wide (for those who opt-in). There is so much great information out there completely unharnessed right now and Google is just letting it sit there.

  2. I agree that that would be a good idea to an extent. The problem being that no matter how good you think you and your friends are (and I don’t mean you Robert, just talking in general here), there is always going to be interesting stuff going on outside of your clique that you’ll miss because you’re too focused on the feeds you and your friends read.

    I know that you use other methods like blog searches (I saw you shared my post earlier via a google blog search), but it still won’t be able to match the raw processing power of millions of users on Digg. Of course it will also filter out all of the crap that is on Digg, but sometimes you have to take all the bad to find that one gem.

    My point is that a combination of all of these things will work best. I too find my feed reader to be 99% more useful and quicker than Digg, Techmeme, and the like but if I didn’t check Digg every once in a while I would miss some decent stuff.

    That is why I want what I’ve laid out numerous times – for Google Reader to share the ‘read’, ‘starred’, and ‘shared’ data network wide (for those who opt-in). There is so much great information out there completely unharnessed right now and Google is just letting it sit there.

  3. The job of story selection – which is what we’re talking about here – is one of the toughest in professional journalism, and it’s a bit that most people who’ve never worked in media dismiss as trivial.

    So congratulations on discovering what main stream media has known for years: a small number of dedicated, knowledgeable editors can do a better job for a group of readers than a large number of people with less knowledge.

  4. The job of story selection – which is what we’re talking about here – is one of the toughest in professional journalism, and it’s a bit that most people who’ve never worked in media dismiss as trivial.

    So congratulations on discovering what main stream media has known for years: a small number of dedicated, knowledgeable editors can do a better job for a group of readers than a large number of people with less knowledge.

  5. MG Siegler: I guess this is where we differ. I’d rather avoid spam than get every single potential article. There’s way too much to read anyway, so having a filter is very helpful for most people.

  6. MG Siegler: I guess this is where we differ. I’d rather avoid spam than get every single potential article. There’s way too much to read anyway, so having a filter is very helpful for most people.

  7. True enough – there is too much data out there for one person to go through. That’s why I think a combination of everything we have now works best. I just don’t think I’d be comfortable throwing all of my eggs in one basket – no matter how nice the basket…

    It’s an interesting discussion nonetheless and if Dave Winer or Steve Gillmor build such a site, you can bet I’ll be on it often. I don’t think it’s a bad idea at all – I just don’t think it’s the end-all.

  8. True enough – there is too much data out there for one person to go through. That’s why I think a combination of everything we have now works best. I just don’t think I’d be comfortable throwing all of my eggs in one basket – no matter how nice the basket…

    It’s an interesting discussion nonetheless and if Dave Winer or Steve Gillmor build such a site, you can bet I’ll be on it often. I don’t think it’s a bad idea at all – I just don’t think it’s the end-all.

  9. I think that Slashdot is the linkblog with a few guys picking what goes on that you are looking for. Unfortunately, /. is a bit slow.

  10. I think that Slashdot is the linkblog with a few guys picking what goes on that you are looking for. Unfortunately, /. is a bit slow.

  11. I’ve even canceled my subscription of you’re link blog because of duplicates (better said: too much about one topic) and its quantity at all.
    Just to fit the amount of time, i spend reading feeds everyday, into my daily actions.

    What do you think of this scenario? More quality, less quantity?

    It’s btw the same reason why I only watch you’re videos if there’s a link to a short version directly in your feed, but now i get offtopic…

  12. I’ve even canceled my subscription of you’re link blog because of duplicates (better said: too much about one topic) and its quantity at all.
    Just to fit the amount of time, i spend reading feeds everyday, into my daily actions.

    What do you think of this scenario? More quality, less quantity?

    It’s btw the same reason why I only watch you’re videos if there’s a link to a short version directly in your feed, but now i get offtopic…

  13. I think you hit the nail on the head here Robert. I’ve largely migrated to Jaiku recently but still try to keep up with my Twitter friends. Unfortunately most days I only get time to skim through. The one Twitter feed I don’t skim is your linkblog. I unsubscribed from Digg last year because there was about 2 hits for every 8 misses in terms of matching my interests. However I consistently find myself clicking through on 8 out of every 10 items in your linkblog.

    On days when I’m extremely busy I only read the feeds from my friends, techmeme and your linkblog. I feel like I’m still connected while tracking those.

  14. I think you hit the nail on the head here Robert. I’ve largely migrated to Jaiku recently but still try to keep up with my Twitter friends. Unfortunately most days I only get time to skim through. The one Twitter feed I don’t skim is your linkblog. I unsubscribed from Digg last year because there was about 2 hits for every 8 misses in terms of matching my interests. However I consistently find myself clicking through on 8 out of every 10 items in your linkblog.

    On days when I’m extremely busy I only read the feeds from my friends, techmeme and your linkblog. I feel like I’m still connected while tracking those.

  15. “if we find a news junkie who thinks like us we’ll find that person to have high utility” … Isn’t that true in everything? We find good friends because we have a lot in common. We find a movie reviewer that likes what we like, etc. I tend to agree with MG Siegler that the idea has value but one needs a variety of sources or one becomes a dull boy.

  16. “if we find a news junkie who thinks like us we’ll find that person to have high utility” … Isn’t that true in everything? We find good friends because we have a lot in common. We find a movie reviewer that likes what we like, etc. I tend to agree with MG Siegler that the idea has value but one needs a variety of sources or one becomes a dull boy.

  17. My personal dislike of sites like DIGG is that it is very much US/english oriented. There are fabulous Dutch websites, that will never ever make it to DIGG, since they are in Dutch. It seems to me DIGG leaves out the long tail, while some of the very interesting stuff is just there: in the long tail.

  18. My personal dislike of sites like DIGG is that it is very much US/english oriented. There are fabulous Dutch websites, that will never ever make it to DIGG, since they are in Dutch. It seems to me DIGG leaves out the long tail, while some of the very interesting stuff is just there: in the long tail.

  19. I would like to see news sites just plain focused. If I am interested in the iPhone then give me iPhone news or if I am a Rails developer then give me Rails news.

    I think the idea is good and focused long-tail sites would be worthy of exploring.

  20. I would like to see news sites just plain focused. If I am interested in the iPhone then give me iPhone news or if I am a Rails developer then give me Rails news.

    I think the idea is good and focused long-tail sites would be worthy of exploring.

  21. I think a media discovery service which also “learns” user specific reading behavior would be the best. However i am not sure whether the machine learning techniques have reached the stage where this could happen.

  22. I think a media discovery service which also “learns” user specific reading behavior would be the best. However i am not sure whether the machine learning techniques have reached the stage where this could happen.

  23. Tech for Novices: I checked you out and all I see are ads and links to other posts you made. Hey, if you’re gonna play up the “we’re small and we’re a victim” you’ve gotta have a better looking blog than that! Sorry.

    Jonas: I have to wonder if you ever read my link blog. I generally don’t put up much repetition unless it’s a topic like Kindle or iPhone or Facebook and, even then, I don’t put up stupid posts, but only those that bring a new point to the table.

    Even if you collected these few instances, though, generally only 5% of the posts, or less, repeat on a theme.

  24. Tech for Novices: I checked you out and all I see are ads and links to other posts you made. Hey, if you’re gonna play up the “we’re small and we’re a victim” you’ve gotta have a better looking blog than that! Sorry.

    Jonas: I have to wonder if you ever read my link blog. I generally don’t put up much repetition unless it’s a topic like Kindle or iPhone or Facebook and, even then, I don’t put up stupid posts, but only those that bring a new point to the table.

    Even if you collected these few instances, though, generally only 5% of the posts, or less, repeat on a theme.

  25. “…the news reading behaviors of only a handful of people that he can get much better results than if he has a larger group…”

    Clay Shirky’s 2002 article Communities, Audiences, and Scale. has some thoughts on the subject. Basically, as a group grows it transitions from a social network to an audience.

  26. “…the news reading behaviors of only a handful of people that he can get much better results than if he has a larger group…”

    Clay Shirky’s 2002 article Communities, Audiences, and Scale. has some thoughts on the subject. Basically, as a group grows it transitions from a social network to an audience.

  27. Hey Robert

    I have been stealthily running http://www.seriouslymedia.com and http://www.seriouslywallstreet.com for the last 3 months. They are running on Pligg and I have automated feeds from people I trust into two single places where trends in online and traditional media (seriouslymedia) and developments in wall street (seriouslywallstreet) are covered.

    I did these for my own use as an experiment.

    As usual, Dave is onto something. The devil is, as always, in the details.

    Keith Teare
    ceo/founder/edgeio

  28. Hey Robert

    I have been stealthily running http://www.seriouslymedia.com and http://www.seriouslywallstreet.com for the last 3 months. They are running on Pligg and I have automated feeds from people I trust into two single places where trends in online and traditional media (seriouslymedia) and developments in wall street (seriouslywallstreet) are covered.

    I did these for my own use as an experiment.

    As usual, Dave is onto something. The devil is, as always, in the details.

    Keith Teare
    ceo/founder/edgeio

  29. I am sure this will be no different from the Slashdot of the early days. Doesn’t make much sense to me. Even if it works out great in the beginning, I would give 3-6 months before it becomes history.

  30. I am sure this will be no different from the Slashdot of the early days. Doesn’t make much sense to me. Even if it works out great in the beginning, I would give 3-6 months before it becomes history.

  31. I don’t pay any attention to Digg, nor to TechMeme, because now it duplicates, or is duplicated by, BlogRunner. I am better off reading my own feeds. But that may be because I’m not just a techie: I read in politics and health care as well.

  32. I don’t pay any attention to Digg, nor to TechMeme, because now it duplicates, or is duplicated by, BlogRunner. I am better off reading my own feeds. But that may be because I’m not just a techie: I read in politics and health care as well.

  33. This seems like just another scheme/platform for Robert, Dave, and Steve to promote Robert, Dave, and Steve. Yawn.

  34. This seems like just another scheme/platform for Robert, Dave, and Steve to promote Robert, Dave, and Steve. Yawn.

  35. My workaround: A few of us who are highly interested in and/or expert in political candidate debates use del.icio.us and simply tag everything debatescoop. It is a tag no one else uses. We then use additional tags for subcategories like GOP, formats, winners/losers.

    I imagine you could use other sites or systems that retrieve tagged items for whatever unique group you have: just use one unique tag along with whatever other tags apply to the subcategory of your subject.

    I use my link blog on Google to communicate on a different subject (Middle East Policy.

    Problem: I have two audiences for two niche subjects. My solution to this date is to use a social bookmark site for one group and the quicker (in terms of my clicks) google link blog for another.

  36. My workaround: A few of us who are highly interested in and/or expert in political candidate debates use del.icio.us and simply tag everything debatescoop. It is a tag no one else uses. We then use additional tags for subcategories like GOP, formats, winners/losers.

    I imagine you could use other sites or systems that retrieve tagged items for whatever unique group you have: just use one unique tag along with whatever other tags apply to the subcategory of your subject.

    I use my link blog on Google to communicate on a different subject (Middle East Policy.

    Problem: I have two audiences for two niche subjects. My solution to this date is to use a social bookmark site for one group and the quicker (in terms of my clicks) google link blog for another.