strange choice of "Post 1" to "Post 5". While these are posts I voted, they are not the most recent posts and they are not necessarily posts to which I awarded a 100% vote. A post I voted at 100% more recently than Post 3 (but before calling you) was this one yet you didn't use it ...
Anyway, the problem is that I do sell a bit of voting power when at 100% (because I don't want to waste it) and therefore your recommendations do not necessarily reflect what I really read ...
Well, yes, so far I don't distinguish between voting weight, I just check if it is positive. And also true, to be precise, I take the last 5 most recent posts that survive the preprocessing and filtering. Because if you vote on too short or non-English posts, I cannot use them in my tf-idf encoding.
it's not about "making a quick buck" - it's about not letting VP get stale. It helps nobody when VP stays at 100% unsued. If I sell down to, say, 99% it helps people who want to promote their posts and it brings me a little bit of steem power
Isn't the reward pool distributed regardless of how many people vote? You just have a proportional share in deciding who can benefit!?
Hence, not voting is always better than voting on crap, because there is a chance that other people not making crap get a bigger piece of the pie. Correct me if I am wrong.
well ... it depends on the level of reflexive thinking - it's an inherently unstable, non-Nash equilibrium. Because the more people abstain from voting on crap, the bigger the reward to "defectors" who choose to "cheat" and vote on crap. More seriously and philosophically, it's been the undoing of Communism.
if everybody acts altruistically and only takes from the "reward pool" (of the socialist state) what it needs, according to the rules of the Party, and puts into the pool according to his abilities, then Communism is paradise.
Except it's not. Because the more people respect the rules and comply, the higher the reward of the cheaters and defectors. With social groups bigger than 200 people it breaks down inevitably
Hi @sorin.cristescu, thanks for voting and resteeming! I used the following post(s) to find recommendations for you: Post 1, Post 2, Post 3, Post 4, Post 5.
Based on your interests I recommend the following articles:
1: Steem Sincerity Spam Filtering API and App Extensions by @andybets (50% match)
2: **Love It Shove It Challenge ** by @princessmewmew (35% match)
3: Steem Sincerity Simple Spam Filtering Chrome Extension for Steemit.com by @andybets (34% match)
4: Jokes on you Google. by @skapaneas (34% match)
5: Crypto Ads Banned by GOOGLE 2018 Scam Warning or Positive News ? by @uxair (44% match)
6: Privacy Issues From Our Online Activities by @iyanpol12 (41% match)
7: Google Will Remove Cryptorelated Ads Starting June 2018 by @agamsaia (48% match)
8: Google You Son Of A ! You Are Making Me Lose Money by @bluedemon (36% match)
9: loveitshoveit challenge... Accepted! by @bluesapphire06 (34% match)
10: How Google and Facebook are Making Money from Your Personal Data And What One Blockchain Company is Doing to Stop It by @cryptofy (37% match)
strange choice of "Post 1" to "Post 5". While these are posts I voted, they are not the most recent posts and they are not necessarily posts to which I awarded a 100% vote. A post I voted at 100% more recently than Post 3 (but before calling you) was this one yet you didn't use it ...
Anyway, the problem is that I do sell a bit of voting power when at 100% (because I don't want to waste it) and therefore your recommendations do not necessarily reflect what I really read ...
Well, yes, so far I don't distinguish between voting weight, I just check if it is positive. And also true, to be precise, I take the last 5 most recent posts that survive the preprocessing and filtering. Because if you vote on too short or non-English posts, I cannot use them in my tf-idf encoding.
Btw, the querying of posts like @hounddog https://steemit.com/science/@suesa/fasting-and-pain, works a bit better to my mind, especially for people making a quick buck by selling their precious vote :P (see below)
it's not about "making a quick buck" - it's about not letting VP get stale. It helps nobody when VP stays at 100% unsued. If I sell down to, say, 99% it helps people who want to promote their posts and it brings me a little bit of steem power
Isn't the reward pool distributed regardless of how many people vote? You just have a proportional share in deciding who can benefit!?
Hence, not voting is always better than voting on crap, because there is a chance that other people not making crap get a bigger piece of the pie. Correct me if I am wrong.
well ... it depends on the level of reflexive thinking - it's an inherently unstable, non-Nash equilibrium. Because the more people abstain from voting on crap, the bigger the reward to "defectors" who choose to "cheat" and vote on crap. More seriously and philosophically, it's been the undoing of Communism.
if everybody acts altruistically and only takes from the "reward pool" (of the socialist state) what it needs, according to the rules of the Party, and puts into the pool according to his abilities, then Communism is paradise.
Except it's not. Because the more people respect the rules and comply, the higher the reward of the cheaters and defectors. With social groups bigger than 200 people it breaks down inevitably