The LTB Network Forum

Avoiding shallow articles

Back to Content Garden

Comments

• crystalck04

  

  Offline
  Posts: 23

  Adam said: I have also recognized this problem @sara and have come to realize that ANYONE will be able to create a show or blog on the LTB network, and the best will be included on the front page by default, with specific articles or episodes that achieve a better than usual result on subsidiary pages being picked out and included on the front page.  LTBcoins for content creation on the incentive side are only paid out to content that appears on front pages.  There will be more than one Front Page, but each front page will basically have its own curation team and target a specific niche, or language.
  Still thinking about this.   I think the concept got a lot bigger once I realized how many people want to create content for us.

  I've been meaning to ask about this. Is this going to be one giant website, with everyone contributing? Or is someone's individual website somehow connected to LTB?
  

  Posted on March 22, 2014 at 4:05 AM
  Permalink
  0 likes
• Conrad

  

  Offline
  Posts: 11

  Adam said: Another idea could be to crowd source it. So for example.. article gets submitted. Kicks around in a "new" section or something for a bit, where it goes through a sort of peer review process. If a user can provide evidence that the article was plagarized from somewhere (or otherwise complete BS), then they can submit their evidence to the administration/moderators who can then look into and confirm the allegations and reward the user with X amount of LTBCoin for their efforts. The process could be streamlined so would basically be just a simple evidence submission form which notifies a mod. Mod checks it out, removes the article if needed and gives the user their reward with a single button click. 
  Yep, this seems like the way to do it.  Don't automate it, incentivize and outsource it to the audience.

  This may create the wrong incentives though: me and a buddy could work together to exploit it. He submits a plagiarized article and I collect the rewards for flagging it. I could always be the first or one of the first people to flag it ()because my buddy would tell me exactly when he will send it) and I would be certain that I'm correct. 
  One way to avoid these problems but achieve a similar effect would be for content producers to send a small amount of LTBCoin whenever they submit an article. If the article is not shown to be plagiarized they would get those LTBCoins back as well as any LTBCoins sent by people who wrongly flagged it as plagiarized.
  The possible problem with this solution is that people who submit content would first have to obtain some LTBCoin from somewhere (in the first phase this would create a chicken-egg problem, but in later phases when they could buy these coins on the market this still creates a barrier for people to submit content (they first have to perform the action of obtaining coins), and this barrier may result in fewer 'good enough or better' articles being submitted than there would be without that barrier. Then again, it may work to create a better ratio of desirable vs. undesirable articles submitted.

  Posted on March 27, 2014 at 8:52 PM
  Permalink
  0 likes
• Conrad

  

  Offline
  Posts: 11

  Thinking about this some more: it would be interesting if people could 'bet' on their own articles, not just 'bet' in the sense that by having to send LTBcoins when they submit an article they are 'betting' that they didn't plagiarize said article (that's the model I described in my previous comment) but 'bet' in the following sense:
  if I think the article that I'll be submitting is very good, then I could 'bet' that it is, or more concretely that it will be judged to be very good. This judging could be done by editors or it could be more automated via a proxy variable (page views, Twitter mentions). 
  I could then for example bet 10 LTBCoins that my article will be in the top 5 or something of that week, and if I'm very confident I would bet a lot of money. How could this investment then pay off? 
  Well, other people could take the other side of the bet, but the market may be too small for that to happen. 
  Or the number of LTBCoins that a top 5 article would receive could somehow be made dependent on the 'invested' number of LTBCoins. Basically, everybody who submits an article has to submit LTBCoins, and this pool of invested LTBCoins is added to the number of LTBCoins that in the original model would be rewarded to content producers, and the distribution of that combined pool would be made dependent on how good/important an article is judged to be (by editors and/or page views/Twitter mentions etc) and the number of LTBCoins invested by the content producer. 
  So somebody who 'bet' 1 LTBCoin when she submitted an article would receive fewer LTBCoins than somebody who 'bet' 5 LTBCoins when she submitted an article, even if they are judged to be equally good (if they are tied for 2nd place for example).
  Not sure whether/how this would work mathematically though.  Just thinking out loud.

  Posted on March 27, 2014 at 9:06 PM
  Permalink
  0 likes
• Conrad

  

  Offline
  Posts: 11

  
  
  So somebody who 'bet' 1 LTBCoin when she submitted an article would receive fewer LTBCoins than somebody who 'bet' 5 LTBCoins when she submitted an article, even if they are judged to be equally good (if they are tied for 2nd place for example).

  I guess in this scenario even an article that was for example judged to be the fifth best / most successful article could end up making its producer more money than the #1 article did for its producer, If the creator of the #5 article for example 'bet' 10 times more money than the creator of the #1 article did. 
  Not sure if this is necessarily a bad thing though.

  Posted on March 27, 2014 at 9:41 PM
  Permalink
  0 likes
• Conrad

  

  Offline
  Posts: 11

  Another thought: so far I've only talked about the producer of an article betting on / investing in that article (and one or more people taking the other side of the bet, if it's a bet)
  But perhaps the betting system could be generalized in that anybody who sees an article can bet on or otherwise invest in the success of that article. 
  This would function as a substitute for e.g. 'likes' or 'upvotes'. The thing about 'likes' and 'upvotes' is that there is little to no cost to them as you can 'like' or 'upvote' as many articles as you want (not sure if this is true btw). if you bet or invest on the other hand you'd have to do it with something that is scarce because your reward will be in the form of the same thing you bet or invest with and it will only be a reward if it is something that is scarce. 
  It would also function as a substitute for tipping. If you tip somebody there are no potential rewards for you. Betting on or investing in an article on the other hand means that if you correctly predict that an article will be successful you get more money back than you bet / invested.
  The advantage of this system is that it rewards those who are good at predicting (as early on as possible) what articles will be successful, and punishes those who are bad at it. 
  (Gee, if true, then decentralized content curation may be the killer app of prediction markets.)

  Posted on March 27, 2014 at 10:03 PM
  Permalink
  0 likes
• jason

  

  Offline
  Posts: 9

  Conrad said: Another thought: so far I've only talked about the producer of an article betting on / investing in that article (and one or more people taking the other side of the bet, if it's a bet)
  But perhaps the betting system could be generalized in that anybody who sees an article can bet on or otherwise invest in the success of that article. 
  This would function as a substitute for e.g. 'likes' or 'upvotes'. The thing about 'likes' and 'upvotes' is that there is little to no cost to them as you can 'like' or 'upvote' as many articles as you want (not sure if this is true btw). if you bet or invest on the other hand you'd have to do it with something that is scarce because your reward will be in the form of the same thing you bet or invest with and it will only be a reward if it is something that is scarce. 
  It would also function as a substitute for tipping. If you tip somebody there are no potential rewards for you. Betting on or investing in an article on the other hand means that if you correctly predict that an article will be successful you get more money back than you bet / invested.
  The advantage of this system is that it rewards those who are good at predicting (as early on as possible) what articles will be successful, and punishes those who are bad at it. 
  (Gee, if true, then decentralized content curation may be the killer app of prediction markets.)

  This is an interesting idea.. however, wouldn't there be a risk of this "comment prediction market" focusing the gambler's attention to how he thinks other people will rate the article instead of the value they actually see in it? Basically, wouldn't this just make people want to look for articles that are easy to guess people's reactions to and stay away from more controversial ones?
  I think the profitability of articles would probably rise & fall as time passes and crypto-related news stories propagate across the blogosphere. There may be no conclusive "value" to be found in any comments - they become transient things to bet on. I'm not sure if this is a good thing or a bad thing. It may be the next evolution of the DIGG / REDDIT type models.

  Posted on March 28, 2014 at 1:08 PM
  Permalink
  0 likes
• Conrad

  

  Offline
  Posts: 11

  jason said:
  

  Conrad said: Another thought: so far I've only talked about the producer of an article betting on / investing in that article (and one or more people taking the other side of the bet, if it's a bet)
  But perhaps the betting system could be generalized in that anybody who sees an article can bet on or otherwise invest in the success of that article. 
  This would function as a substitute for e.g. 'likes' or 'upvotes'. The thing about 'likes' and 'upvotes' is that there is little to no cost to them as you can 'like' or 'upvote' as many articles as you want (not sure if this is true btw). if you bet or invest on the other hand you'd have to do it with something that is scarce because your reward will be in the form of the same thing you bet or invest with and it will only be a reward if it is something that is scarce. 
  It would also function as a substitute for tipping. If you tip somebody there are no potential rewards for you. Betting on or investing in an article on the other hand means that if you correctly predict that an article will be successful you get more money back than you bet / invested.
  The advantage of this system is that it rewards those who are good at predicting (as early on as possible) what articles will be successful, and punishes those who are bad at it. 
  (Gee, if true, then decentralized content curation may be the killer app of prediction markets.)

  This is an interesting idea.. however, wouldn't there be a risk of this "comment prediction market" focusing the gambler's attention to how he thinks other people will rate the article instead of the value they actually see in it? Basically, wouldn't this just make people want to look for articles that are easy to guess people's reactions to and stay away from more controversial ones?
  I think the profitability of articles would probably rise & fall as time passes and crypto-related news stories propagate across the blogosphere. There may be no conclusive "value" to be found in any comments - they become transient things to bet on. I'm not sure if this is a good thing or a bad thing. It may be the next evolution of the DIGG / REDDIT type models.

  Well, the thing is, this would sort of take care of itself because different bets have different odds: the potential ROI of betting 1 LTBcoin on uncontroversial articles (low risk) would be considerably lower than the potential ROI of betting 1 LTBcoin on controversial articles / articles not many other people think will be popular (high risk). 
  Not sure what you mean by 'no conclusive "value" to be found in any comments..." Can you say more about what you have in mind?

  Posted on March 28, 2014 at 5:08 PM
  Permalink
  0 likes
• adam

  

  Offline
  Posts: 1980
  Location: California, USA

  Very interesting turn in this thread conrad!  I will catch up fully tomorrow and respond

  Posted on March 29, 2014 at 6:24 PM
  Permalink
  0 likes
• Conrad

  

  Offline
  Posts: 11

  Good to hear. I've thought about it some more and this may actually have potential as a way of curating content. Am writing a brief overview of how this could work and what advantages & problems it may have. Will send it to you today or tomorrow.

  Posted on March 29, 2014 at 6:47 PM
  Permalink
  0 likes
• Coinconomics

  

  Offline
  Posts: 16
  Location: BC

  
  Conrad said: Good to hear. I've thought about it some more and this may actually have potential as a way of curating content. Am writing a brief overview of how this could work and what advantages & problems it may have. Will send it to you today or tomorrow.

  I would be very interested to see this write up as well, please post it on this thread too. Also, as you create the overview, I want to stress to you the importance of having a very rigid mathematical structure in place. If it does not, the "game", while good in theory, could cause real damage to the fragile ecosystem during its early stages. I want to write more, as there are a million things flowing through my head as I think on this topic, but I want to refrain from muddling the waters so to speak.
  
  I eagerly await your post.
  

  Posted on April 7, 2014 at 2:34 AM
  Permalink
  0 likes

Post Reply