5 ms·

Ironically, it was "rethinking digg" and "rebuilding it from scratch" that killed it in the first place.

smacktoward 14y ago
Yeah, that was my first thought too. The second thought was that ground-up rewrites are almost never a good idea to begin with. You end up throwing out whatever's good about the old property along with all the bad, and risk turning off the members of whatever user base you have left (who presumably stuck around because they saw something they liked in the old version). I suppose the population of Digg users is probably small enough at this point that the new proprietors aren't worried too much about whether they stay or go, though...
- pteromyscus 14y ago
  >The second thought was that ground-up rewrites are almost never a good idea to begin with. For a complex system, like an OS or a browser. This is a bloody social news site, as trivial an application as it gets. One man can write a half-decent one in a week. And for ramping it up, even the scaling roadblocks and tricks are old news by now to the web building community in general.
  - jedberg 14y ago
    > This is a bloody social news site, as trivial an application as it gets. I'm gonna have to disagree with you there. It's a fairly complex transactional system. > One man can write a half-decent one in a week. No, one man can write something that resembles a social news site in a week, but building a good one that scales takes a bit longer. :) > And for ramping it up, even the scaling roadblocks and tricks are old news by now to the web building community in general. That one I'll agree with. Their initial scaling problems should be fairly well understood.
    - mikkom 14y ago
      > No, one man can write something that resembles a social news site in a week, but building a good one that scales takes a bit longer. :) That totally depends on your available toolset and experience.
      - jedberg 14y ago
        > That totally depends on your available toolset and experience. That's either ignorant or boastful (or both). Any decent programmer will admit that they can't build a good and scalable anything in a week from scratch. If you count your "toolset" as an entire framework that does what you want already, then sure, maybe it's possible. Otherwise you're just being foolish.
laberge 14y ago
Not really though, everyone knew that V4 was a terrible idea and they simply went forward without caring much for the community. I mean what kind of marketing research could you do to prove that using promoted stories is a good way to build a community?
- SupermanScott 14y ago
  "Promoted Stories" myth needs to end. It wasn't the case, there was a severe bug whereby a Regular Expression only matched RSS content. The Regular Expression acted as a gateway into the Popular Algorithm. I worked at Digg and I fixed that bug. It wasn't noticed before launch because we echoed the v3 popular stories into the beta version of v4. Digg was never paid for stories hitting the frontpage. And for all the flack it gets for this myth, it should have been.
  - laberge 14y ago
    Well the fact rests, the community generally believed that there was a ton of promoted stories because, you guys never told anyone otherwise or did a poor job communicating. And im certain promoted stories did exist, I remember them clearly marked as that.
    - SupermanScott 14y ago
      Digg ads existed that were placed in the feed at slot 3 and 14 which were introduced in v3: http://about.digg.com/blog/ads-you-can-digg%E2%80%A6or-bury http://about.digg.com/blog/ads-you-can-digg%E2%80%A6or-bury They were ads that were pieces of content. Digg was in front of this trend with promoted tweets / stories.
  - nowarninglabel 14y ago
    If you don't mind me asking, how good a job do you think John N. did? Did the buck stop with him on this engineering project?
- smacktoward 14y ago
  You can find a research firm to tell you anything you want. They may have to torture the numbers more in some scenarios than in others, but there's no shortage of people willing to take your money in exchange for telling you what you already want to hear.
rokhayakebe 14y ago
Homeopathy!

Yeah, that was my first thought too. The second thought was that ground-up rewrites are almost never a good idea to begin with. You end up throwing out whatever's good about the old property along with all the bad, and risk turning off the members of whatever user base you have left (who presumably stuck around because they saw something they liked in the old version). I suppose the population of Digg users is probably small enough at this point that the new proprietors aren't worried too much about whether they stay or go, though...

pteromyscus 14y ago
>The second thought was that ground-up rewrites are almost never a good idea to begin with. For a complex system, like an OS or a browser. This is a bloody social news site, as trivial an application as it gets. One man can write a half-decent one in a week. And for ramping it up, even the scaling roadblocks and tricks are old news by now to the web building community in general.
- jedberg 14y ago
  > This is a bloody social news site, as trivial an application as it gets. I'm gonna have to disagree with you there. It's a fairly complex transactional system. > One man can write a half-decent one in a week. No, one man can write something that resembles a social news site in a week, but building a good one that scales takes a bit longer. :) > And for ramping it up, even the scaling roadblocks and tricks are old news by now to the web building community in general. That one I'll agree with. Their initial scaling problems should be fairly well understood.
  - mikkom 14y ago
    > No, one man can write something that resembles a social news site in a week, but building a good one that scales takes a bit longer. :) That totally depends on your available toolset and experience.
    - jedberg 14y ago
      > That totally depends on your available toolset and experience. That's either ignorant or boastful (or both). Any decent programmer will admit that they can't build a good and scalable anything in a week from scratch. If you count your "toolset" as an entire framework that does what you want already, then sure, maybe it's possible. Otherwise you're just being foolish.

laberge 14y ago

Not really though, everyone knew that V4 was a terrible idea and they simply went forward without caring much for the community. I mean what kind of marketing research could you do to prove that using promoted stories is a good way to build a community?

SupermanScott 14y ago
"Promoted Stories" myth needs to end. It wasn't the case, there was a severe bug whereby a Regular Expression only matched RSS content. The Regular Expression acted as a gateway into the Popular Algorithm. I worked at Digg and I fixed that bug. It wasn't noticed before launch because we echoed the v3 popular stories into the beta version of v4. Digg was never paid for stories hitting the frontpage. And for all the flack it gets for this myth, it should have been.
- laberge 14y ago
  Well the fact rests, the community generally believed that there was a ton of promoted stories because, you guys never told anyone otherwise or did a poor job communicating. And im certain promoted stories did exist, I remember them clearly marked as that.
  - SupermanScott 14y ago
    Digg ads existed that were placed in the feed at slot 3 and 14 which were introduced in v3: http://about.digg.com/blog/ads-you-can-digg%E2%80%A6or-bury http://about.digg.com/blog/ads-you-can-digg%E2%80%A6or-bury They were ads that were pieces of content. Digg was in front of this trend with promoted tweets / stories.
- nowarninglabel 14y ago
  If you don't mind me asking, how good a job do you think John N. did? Did the buck stop with him on this engineering project?
smacktoward 14y ago
You can find a research firm to tell you anything you want. They may have to torture the numbers more in some scenarios than in others, but there's no shortage of people willing to take your money in exchange for telling you what you already want to hear.

rokhayakebe 14y ago

Homeopathy!