Digg v4 Troubles are Symptom of a Bigger Problem

Digg v4 continues to hit bumps along the road after its launch.

Technical glitches and software bugs in the system ran rampant at launch; the type of glitches that made you feel like the release was rushed or pushed out the door for some reason. Besides the hiccups I ran into, I saw a few emails from friends echoing much the same results. One of the more interesting issues in the Friend system that caused you to be friends with someone multiple times.

Digg v4 was also suffering from a surprisingly amount of down time. In the first few weeks of launch, it wasn’t that uncommon to visit Digg.com and see their Oregon Trail throw-back graphic saying that the site was down and should be back shortly:

Shortly after Digg v4 launched Kevin Rose announced that he was stepping down as CEO of Digg and being replaced by Amazon’s Matt Williams. Rose had only served as the CEO of Digg for 6 months and was quoted as saying it was a job he “would not wish on his worst enemy“.

My take on why you hear about Digg’s hard knocks so much in the news is because its community wants it to be the “powered by the people” site it was when it launched. Instead, Digg continually moves more and more towards and advertising/media/corporate site they don’t want, but makes Digg more money. The reality is that Digg hasn’t been “powered by the people” for a while now, but its domination by a relatively small group was much more subversive than it is now. Now, with Digg v4, the publisher friendly “Feed Auto-Publish” functionality floods the network with content from the leading publishers, drowning out what little attention individual bloggers could have gotten before.

In addition to the movement of Digg’s CEO, it was just announced that Digg has fired long time VP of Engineering, John Quinn.

If you choose to read through the comments section of that TechCrunch post, you have to filter through a lot people postulating that the firing was about Quinn’s choice to move Digg to a NoSQL-backed (Cassandra) data store and the technical hiccups Digg is experience as a result of that. I am saying this is incorrect. Even Kevin Rose was excited by the Cassandra backend (specifically called out in his interview here at 3:45).

UPDATE #7: Jonathan Ellis, project chair on Cassandra, points to a Quora conversation where 2 Digg engineers clarify the stability problems are not because of Apache Cassandra.

I don’t think anyone at Digg had the illusion that this almost rewrite would roll out smoothly, and given that the NoSQL rewrite even had Kevin’s support, the firing of Quinn is that much more suspect. I will say again that the technical issues the site has experienced since launch was not why Quinn was let go; I don’t know why he was, but the “software bugs” argument doesn’t hold up from what I see.

Firing the VP of Engineering, the architect behind the entire rewrite of your site, that has been eating/breathing/sleeping your technical concerns for the last 3 years and is the only one that likely knows how every part of the site works just because of a rocky launch and 2 weeks of questionable up time is stupid; Digg is not that stupid; no one is that stupid.

As one commenter quipped, if Twitter had fired engineers because of rocky launches and questionable code in the first few months of getting popular, there would be no one working at the company. (Ha!)

The reason Quinn was let go has to have something to do with the corporate culture at Digg and what is going on over there. TC had an interesting story on the reaction from Digg employees who found out that John Quinn was being fired via TechCrunch and not from their internal team members. That is insulting and alienating, but it also says a lot about Digg and how that team is running right now.

Given that Danny Trinh (the employee that tweeted about Quinn’s firing) felt comfortable enough to air that grievance, tells me that Digg doesn’t normally operate in this stealth mode where nothing is communicated with the other teams. If they did typically operate in an oppressive manner, I doubt Danny would have so nonchalantly quipped about internal company policy publicly for fear of losing his job.

The only time I’ve ever seen companies suddenly decide to operate in stealth mode at the highest levels is when they are in disaster-recovery mode; scrambling to put pieces back together and get a strategy together. The firing of Quinn (a major player behind the Digg technology stack) at the time he was fired (when most needed) suggests that Digg is in crazy-disaster-recovery mode and are desperately trying to find the right direction to point.

My guess is that Digg is franticly dealing with two crisis that they are trying to find answers to:

  1. Shrinking user base leaving because of the “publisher dominated” environment Digg v4 has created.
  2. Persistent technical problems

Users Leaving Digg

If you doubt the first point (shrinking user base) is happening, look at the type of site Digg was when it started: Mass story submissions, Mass voting, most popular stories bubble to the top. It was simple and brilliant at the time.

As the “social”-obsessed engineering continued, it became more about leveling up your Digg profile than the content itself and that has not changed. The MMORPG-esque nature of your Digg profile is intentionally engineered into the site to get you to return, this isn’t some elaborate anti-SPAM technique to establish “trusted” sourced (it does serve that purpose tangentially in some cases though). Kevin acknowledges this around 5:35 in his interview here.

I am not saying an RPG approach to profiles is bad, but it is clear there is a advertising incentive here to get users to return that is paramount over the concern of surfacing the best content.

With Digg v4, the idea is that you have your own personal Digg-view of the world, and the content you look at consists of all the content from the people you follow… directly. While they may open that up in the future and allow you to see content from friends-of-friends (a la Twitter re-tweets) the point is that now you have the burden of building a valuable circle so you can read valuable content and aren’t stuck reading garbage all day.

The cost associated with this, to users, that I don’t know Digg really thought through, is how much time it takes to build out a trustworthy network of people to follow and listen to. How am I suppose to create a content network as generally valuable as the millions of people that were submitting to Digg in the pre-bullshit days? On Twitter it has taken me about a year to find 80 people that I value following, I can’t imagine how long it will take on Digg or what the mechanism for me to discover additional people is going to look like.

Digg v4 allows me to import my contacts from Twitter, Facebook and Google, but the value-assumption here is wrong. When I imported my valuable Twitter followers, 6 of them had Digg accounts, and I think 2 of them were active and not necessarily posting the same valuable content I liked from their Twitter feed. My Facebook contacts are a cluster fuck of old high school classmates and relatives; I am hardly interested in seeing a Digg stream from these people nor do I think 1/100th of them use Digg. Lastly, my Google contacts… that would be a Digg stream to end all Digg streams. Content from half the planet, almost none of which I would want to read nor do I think most are Digg users.

Update #8: Jason Keirstead makes the point below, that without a giant vetted personal network of friends to pull your news from (something we doubt most people will take the time to create), Digg v4 will essentially become a giant RSS feed reader.

Because of this “What the hell do I do now?” moment, people do exactly what I did after signing into Digg v4… they follow the most recognizable names they can think of: Engadget, Ars Technica, Gizmodo, TechCrunch, etc. and now when I log into Digg, “My News” is nothing more than the front page of those sites… sites that I read already except now it’s all I see in my little world.

The result of this is that I am no long even exposed to individual written content; I have no idea how to reliably find that content anymore. Instead I just fall back to “Plan B” which is to follow names I recognize. Naturally, publishers love this because they have shaken the small-time writers out of the Digg tree, and Digg hopes to get more advertising revenue and publisher love for this.

It seems like the readers and users of the site that like the ideal of “social news” are the only ones that lose here and most people are voting by trailing off and finding other sites that itch that scratch better than Digg v4 does.

Update #9: ReadWriteWeb has confirmed traffic numbers from HitWise that Digg’s traffic has fallen 26% since the v4 launch:

Digg’s Technical Problems

Given the second point (persistent technical problems), I am still surprised that they let Quinn go. A few commentors in the TechCrunch thread said they had worked with Quinn and he was a person of the highest caliber; which only makes his firing from Digg even stranger at a time like this.

One thing that pops into my head as a reason for firing Quinn at such a bad time AND gives fuel to the fire that is #1 is a fundamental shift of focus for the site at the highest level: Focus on publishers, Focus on advertising income.

There may be movement afoot to make Digg go head to head with Google News, Huffington Post, CNN or any other mass-news-aggregation source for the big advertising dollar income. An environment generated by publishers and managed by the community for free while Digg collects advertising revenue.

Unfortunately, I think Digg underestimated how badly the community wants their own site that they run and aren’t willing to effectively editorialize the content mill that is Digg.com for free anymore.

If anyone doubts that Digg is shedding customers like fleas, since Digg v4’s launch, reddit’s traffic has been up as people look for more community driven sites.

Social News without Advertisement Interests?

From what I’m seeing, it doesn’t seem to work. Hacker News does a good job of this because it’s bankrolled by Y Combinator, reddit is a great source of content, but their financial woes persist and Slashdot has done an admirable job with a Social+Editor approach to news and a handful of editors to vet the highest bubbled stories; I also think the culture behind Slashdot allows it to continue operating mostly un-molested by business interests.

I am sure the readers have other recommendations here for social news sites that they enjoy, but in my limited view of the social news sites that have exploded, failed and persisted through the Hype Cycle, it seems to me the only ones that have become profitable and sustainable on their own, are the ones that have pandered to the almighty dollar, content be damned.

DISCLAIMER: I have nothing against Digg, Kevin Rose or his team. I, like many people, miss having a solution to the problem of “How do you surface good content online easily?” that Digg did relatively well before Digg v3 even. As a consumer of content online, I’m back to hunting through multiple sites, lists and Twitter feeds to find my own “most valuable content“.

Update #1: Fixed some grammar (thanks John) and added an entire section on why Digg v4 is causing a user exodus.

Update #2: Added some clarifying details why Quinn being fired for the “NoSQL debacle” is hog-wash. Even Kevin Rose was excited specifically about this aspect of the new design; Quinn had full support for this and as a result, as the site has stabilized, it is a lot faster.

Update #3: Clarified a bit about how your Digg profile becomes a lot like a World of Warcraft character that you have to work on “leveling” to get noticed and how this is engineered into the Digg system intentionally.

Update #4: Added a new section on social news and bending to the will of money.

Update #5: Added personal experience around importing 3rd party social contacts.

Update #6: A very interesting post from a self-professed “previous Top 20 Digg Power User” here. Claims the gaming of Digg has been going on for years and a tight-knit group of about the top 20 guys all know each other and use tactics that they only share with each other and no one else. They also get paid to have stories hit the front page. He claims he was making $8-10k a week doing this. He also mentions that the changes in Digg v4 to remove power from the “power users” was the only thing that could save the site. Unfortunately now, “power users” have just been replaced by publishers much to the chagrin of the community. The “gaming” of the site is now out in the open and clearly controlled by publishers. He also claims that Jay (when he was CEO) only wanted to sell Digg and the company is desperately trying to stay afloat and support the 100-employee, multiple-data-center bloat that it has created by targeting high-profit changes with the site.

, , , , , , , , , , , , , , ,

27 Responses to Digg v4 Troubles are Symptom of a Bigger Problem

  1. John Hunt September 8, 2010 at 4:40 pm #

    Not trying to be nasty or anything, but I recommend you fix your grammatical errors in this post.

    • Riyad Kalla September 8, 2010 at 5:22 pm #

      John, appreciate the feedback, but grammar and I have a tenuous relationship… we’ll see what I can do.

      UPDATE: Ok, cleaned things up a bit. Removed some of the lame attempts at humor and went with a more neutral tone. Also added an entire section on why Digg v4 is causing users to walk out the door. Thanks for the prompting.

      • Dave September 9, 2010 at 9:13 am #

        Your grammatical errors are still present.

        “…release was rushed our pushed out the door…”

        “On interesting one was a bug…”

        I stopped reading after the first paragraph. Sorry.

        • Riyad Kalla September 9, 2010 at 9:41 am #

          Dave, sorry you didn’t finish reading the article. I appreciate the feedback none the less (fixed).

    • Riyad Kalla September 8, 2010 at 7:18 pm #

      Alright John, I owe you some serious thanks. Your (constructive) criticism sat on my brain all day and since then I’ve edited the story probably 8 times, added 4 new sections and rewritten big portions of it.

      I think it’s a much better article now, and just wanted to give you my thanks for your part in that.

      • Steve Smith September 8, 2010 at 10:17 pm #

        While you’re fixing errors in your post, I think by “hickup” you mean “hiccup”

  2. Matt Arrington September 8, 2010 at 8:38 pm #

    I can’t help but think of this article from Coding Horror’s when I read about Digg 4.0.

    The single worst strategic mistake that any software company can make: Decided to rewrite the code from scratch.


    • Riyad Kalla September 9, 2010 at 5:57 am #


      Certainly a classic post for anyone in tech. Even more surprising, Digg v4 wasn’t just a rewrite from a backend implementation perspective, they fundamentally changed how the site worked and how people are intended to interact with it.

      It would be like Twitter rolling out a new version where there is no longer a message length cap of 140 chars and fundamentally reworking how following/retweeting worked at this late in the game.

  3. Jason Keirstead September 9, 2010 at 4:12 am #

    Digg’s big mistake with V4 is everything you posted, plus one more – it is still there.

    There is simply no way to make the current mechanism work. Users hate it, as is evidenced by the rapid exodus from the site.

    Web media companies have to wake up to how rapidly users can react. This is not Walmart where people have no options… there is Reddit, StumbleUpon, and a host of other alternatives for people to move to, with zero effot. And they have.

    The best thing Digg could have done after a few days of this non-stop debacle was say “I’m sorry” and just put the old site back. Sure they would have lost face but they would not have lost users. Now it is too late and even if they did restore the old site I doubt people would go back.

    • Riyad Kalla September 9, 2010 at 6:00 am #


      I *think* the new site can work theoretically if each user lived in a perfect world and suddenly followed everyone that they were most interested in (i.e. Twitter) — the problem is that no one lives in Perfectland, USA… I don’t want to start from 0 again and start figuring out, out of 10 million people, who I want to start following all over again.

      I’ve importing my Twitter contacts, but there is no guarantee that my Twitter follows behave and post the same level of content on Digg… in fact of my 100 people I follow on Twitter, only 6 had Digg accounts and imported… so now I’m at ground zero again figuring out who the hell to follow.

      I think this is intentional by design — it forces me back to the Digg site, to invest hours of my time building my profile, increasing my time on the site, and increasing Digg’s ad revenue generation.

      No thanks.

      • Jason Keirstead September 9, 2010 at 2:42 pm #

        No, it can’t work. Why? Because this just turns Digg into a giant RSS reader.

        I already have Google Reader. I don’t need another place to see the latest things that were already posted to large news aggregators.

        The whole point of Digg, Reddit, etc. is to bring things to people’s attention that they may not notice otherwise, things that other people have already told them “hey, this is important”. Digg is not a news aggregator, it never was. And it can’t compete with other already existing aggregators.

        The only way it can be successful is for submissions to be user-driven

        This is what Rose and company seem to have lost sight of with V4.

        • Riyad Kalla September 9, 2010 at 2:46 pm #


          Good point. Without a sufficiently large personal network (something I don’t know that many people will take the time to build) you are absolutely right… it really does just turn into a giant RSS feed reader of sorts… with comments… and ads.

  4. Anderson September 9, 2010 at 4:17 am #

    Slashdot survives because it’s become a niche site.

    When Slashdot added an “idle” section, the outcry was huge and was promptly disabled by quite a few users. I distinctly remember a user guessing that Slashdot had to add it and the social media buttons because of the hated “Guys From Marketing” to which CmdrTaco replied tersely that he wasn’t far from the truth.

    I’m a former Digg user and I liked it because of how it _felt_ user generated – it wasn’t, obviously, but it felt like it so I was reading things that my peers liked. Now? I’ll stick with with 4chan-lite called Reddit until a new Digg comes around.

    • Riyad Kalla September 9, 2010 at 6:02 am #


      I didn’t realize there was outcry over the Idle section on Slash — very interesting. As far as it being niche, that *has* to be true… the site has been around since before fire and water and I still read it just like everyone else. If they fundamentally juggled the recipe that works (like Digg v4 is doing) I imagine a lot of people would drop off and head over to reddit.

      Is that really it for our choices? reddit, Stumbleupon, Hacker News… and that’s it?

  5. javipas September 9, 2010 at 5:12 am #

    Brilliant article, Riyad. I agree totally with you, and I’m one of those Digg users that are letting the site in the oblivion. Right now there are alternatives such as HN, Reddit and Slahsdot that are (and have been for a long time) big sources of user-generated content. The one that you haven’t read yet on Techmeme, Twitter or Facebook, that have become the media mirrors for sometime now, something that IMHO Digg didn’t need to.

    Congrats 😉

    • Riyad Kalla September 9, 2010 at 6:03 am #


      Thank you for the kind words. I actually never really gave Techmeme much of a chance, but I’ll give that a peek. Thanks!

  6. Dylan September 9, 2010 at 1:43 pm #

    Well done Riyad. the section on why Digg users left was spot on. Great cultural analysis. I left precisely because of the “What the hell do I do now?” moment. I never collected friends on digg, i had like about 3 after a good three years or more of regularly visiting digg, this didnt mean i didnt comment regularly, or submit content often. it just meant i didnt want to use it as a friends connection site. i was looking for content. rose and the team at digg didnt consider those who didnt want to do it their way. I find that thoughtless of them and as a consequence slightly rude. I wont be going back. Ive got my rss (netvibes) and ive got reddit that ive come to like more and more. ho hum. its a lesson tho for a lot of other sites out there. How not do do things. Great article btw. Thanks

    • Riyad Kalla September 9, 2010 at 2:07 pm #


      The kind words are very much appreciated, I also have to totally agree with the “I never used Digg as a friend connect network and don’t intend to do so now” — very true.

      I think there is a large subset of us that may just stop using Digg, and see if in 3 or 6 months it’s still there, and doing well we might give it a second look or just wait until better “friend mining” tools are released.

      I doubt anyone wishes Digg any ill will, it’s just so odd to see such a big, funded company to change it’s foundation like that mid-race. I guess we’ll wait and see what it eventually becomes.

  7. Pureinfotech September 9, 2010 at 2:10 pm #

    Great article on Digg v4.

    But i am still wondering, if someday, we’ll know what actually really happened.

    Keep up the great work.


  8. Asif Youssuff September 11, 2010 at 11:09 am #

    The advertisement model used by social news sites is a bit troubling, but it can be saved, imo.

    I think of projects like quippd, where users are able to add links to content in a wiki editable fashion, keeping content purely about the story being referenced, and adding high quality sources of news. This keeps the publishers honest — instead of users simply consuming what a single source of news has to say, they are constantly challenged with new sources of information.

    If we can surface the entire *ecosystem* of news, we can get a better idea of what people are saying about a particular story, and not have a single voice dominating the conversation.

    quippd is an interesting evolution of the social news model: http://quippd.com

  9. Charlie September 24, 2010 at 2:31 pm #

    Digg has died of dysentary

    • Riyad Kalla September 24, 2010 at 3:45 pm #

      haha, maybe. Traffic has dropped off since the launch, which I suppose to some degree was expected, but it will be interesting to see if it builds back up or not.


  1. Interesting reading #573 – Mixologist 5000, Iconic prototypes, YouTube’s Vacationer, Cyberwar hype, Smart cities and much more… – The Blogs at HowStuffWorks - September 9, 2010

    […] Digg v4 Troubles are Symptom of a Bigger Problem – “Digg v4 was also suffering from a surprisingly amount of down time. In the first few weeks of launch, it wasn’t that uncommon to visit Digg.com and see their Oregon Trail throw-back graphic saying that the site was down and should be back shortly…” […]

  2. Empezar de cero no siempre es buena idea | Incognitosis - September 10, 2010

    […] Kalla lo comentaba muy bien hace pocos días en The Buzz Media, donde explicaba lo que ha ocurrido con este servicio: With Digg v4, the idea is that you have your […]

  3. Links « Beautiful Discovery - September 11, 2010

    […] Digg v4 Troubles are Symptom of a Bigger Problem Some surprisingly conservative, old-media thinking comes out of the folks at Digg. […]

  4. Digg Traffic Continues to Fall | The Buzz Media - December 21, 2010

    […] of the social networking site. Digg v4 fundamentally changed how the site functioned for its users leading to a long series of backlashes against the site.By the end of September the bugs in the Digg v4 system had been worked and the articles about the […]

Leave a Reply