Topic: Merit & new rank requirements - page 115. (Read 167726 times)

Referral links to external sites? No, don't do that. You'll get negative trust on your account for doing so, as this would be considered gaming the merit system for personal gain. That is in no way "helpful member activity".

Can I advertise merit as reward for using my referral link? Basically it's qualify as helpful member activity.

Is it a typo, my friend!
I guess what you implied is money.
It is definitely the most terrible endemic in the forum, in which we witnessed lots of spammers joined simply to earn money from airdrop,s bounties, campaigns.
They have done lots of efforts to change to forum as a dumping area.
Consequently, Theymos created merit system to change the nightmare, change the forum to better place.
I love him for the amazing idea.

It is too complicated!

You boys have a brain thinking in 1 direction : moneh, moneh and moar moneh!
Most smerit will be soon in the hands of the chosen ones, so this trading repetitive story will not be possible unless someone wanna trade a private message with a negative trust.

---

/cod mw2

Guess who is the admiral general

I think that the rank should be awarded as an army rank, if it deserved - an extraordinary. And with the current rules get the next almost how to accomplish the feat.

Maybe soon merit will become a huge cash business in here

I think the merit system is a good thing.
But it is impossible to get from Member to Full Member and to the next one.

Even if I write 1000 perfect comments, I will not receive 50 merits for it.
I see that even by a hero or legendary with hammer comments.

That must be come a change.

The "Required merit" distance between the individual ranks is too big. This is my opinion.

The new file is up Very fast calculation for a monthy period



Over 50% reduction in 2 months!
April tolal sMerit 23395.
There are 79 merit sources with a total merit generation of up to 17750 sMerit per 30 days, this mean if all sources sent 100% of merit, they have contributed for 75.8 % of total merit for April.

I'm not disagreeing that those posts are a contribution, however, I see posts that contribute in similar ways in lots of threads and they never get merit. My point was that the merit is distributed in such an apparently capricious (not that it is, but it often appears so) manner that people who could deserve to level up aren't getting merit, and other people get rewarded for merit for random posts.

I just don't think the current merit system is really effectively doing what it was set out to do.

I am sorry, i dont agree with you. Those first 2 posts are good because it really sucks when you are taken to a new page when you want to give merits, and later theymos made an adjustment on that mattets, isnt that a contribution?

I think what would probably work better is to have a lot more merit sources, and have the merit requirements higher to reflect that--in other words, dropping merit could be more casual if someone thinks it was helpful, funny, insightful, etc, instead currently where it seems like your post has to be legenDARY in order to be considered worthy of merit.

Of course, unless you scroll back to the first couple posts in this thread, in which case casual comments are showered with merits-- for example the following post on the first page got showered with merits from 28 people (some adding multiple merits):


Or this one by Lauda got merits from 13 people:


Even this post got a merit:


I would say the merits on those three posts were more like "Likes", not really merit as it has been portrayed. Now if merit points were more freely distributed and could be used more like "Likes" (those Like Likes are going to suck me up here any second), then I think they could be used more effectively to reflect good post quality and help with the whole point of merits was supposed to be: pushing bots and low-quality posts to the bottom of the heap.

Thats a great work. In my opinion there are 2 reasons, first is because merit sources are too few right now, or, second, because the quality of the posts are not good enough based on the sources point of view. So we can not only blame it from one side.

I will now explain more details to you. I know 3 languages: Ukrainian 10/10, Russian on 9/10 and English on 8/10.
I want to choose a section where you can communicate more actively, create new interesting topics and develop them.

Subs are completely fucked up, at least i see in the Italian one, there is almost no merit flowing or the sources are not giving enough, as you can see from this graph: https://bitcointalksearch.org/topic/m.34584678

Others are on the same boat as well

What's the point? Let's say for example that Brazilians have 2 moderators, so? Will this change anything?
Even knowing the number of smerit sources for each local sub seems logic, it will only generate pwoblems.

Hello!
Moderators, you can create a table that will record the number of merit sources that generate sMerit by country in the forum?
For example, the English branch is 10 people, German - 3 people, Brazil-4 people.

You didn't  even know what i was talking about, if you are not competent or ignorant in something, better to avoid posting about it.  



they are actually made to solve these problems, it's matter of training with the right data, like if you would need to teach to a child by examples to spot the cases.

Microsoft offer already some service: https://docs.microsoft.com/en-us/machine-learning-server/what-is-machine-learning-server
Google as well https://cloud.google.com/products/machine-learning/

Then again i'm not saying it would be easy but, after all this is the place that saw the birth of bitcoin, should be more open to ambitious projects  

Haha you are making me laughing out loud man, literally. You do talk like it is easy to make an AI in this time right now. Big company like facebook still need a lot of resources to make one, and even google doesnt have one for now, and you are here suggesting  AI for a forum. You are kidding right?

I wasn't aware of the "review threads" but I'll certainly look into it. Regarding the reason why it's important, there are a couple for me personally:

1. My main reason is that I want the merit system to become successful. Several of my regular threads are being overrun by scammers with fake accounts. My hope was that the merit system would resolve these issues. I'm expecting that in the longer term, we can set minimum merit requirements for threads, hide posts of people with negative merit, or something similar that would chase away spammers from BCT.
With the lack of merit being given out and the fact that newbie members can still post everywhere, the merit system does not meet my expectations. My main goal is to help improve the merit system into something that adds value.

2. I was looking forward to becoming a Sr. Member, since I was one day away before the new system got implemented (I had like 1 activity point to go). Suddenly I'm 150 merit away and I've been able to gather 2 new ones for the last months. It just "feels" unfair.

3. Merit points feel like "validation" of a post. Never getting any gets frustrating since it feels like my posts are worthless.

4. Definitely also the signature campaigns. The difference between member ranks for me equals around $1000,- per year. That is quite a lot given that I would have reached a higher status long ago with the old rules.

5. Lastly; I honestly believe that I do have the occasional post that would get "upvotes" on other platforms. Either helping people or giving free advice.
For example; I tried to help a project I'm not involved with this morning with some advice:
https://bitcointalksearch.org/topic/m.36188330

It's definitely not the best post ever, but I do think that these type of posts would warrant 1-2 merit on a platform where it's easy to give out.

Theymos is probably in the final stages of finishing exactly the kind of AI algorithm
that you are talking about  

Jokes apart, I think you are underestimating the difficulty of developing any kind of AI
solution. Even for tasks that are simple for humans it is incredibly hard to develop a
working AI solution.

Just imagine all the pitfalls of developing an automated way of awarding merits.

Here are just a few possible pitfalls:
-how does the algorithm deal with posts in Local boards? It surely can´t be for English posts only...
-how does the algorithm deal with posts that are of very high-quality, but are plagiarism from
other people´s posts or from other websites?
-how does the algorithm deal with posts that are beautifully crafted (think @nullius posting style), but
are completely garbage in terms of content?
...

I agree that we are indeed in 2018 and machines are able to do many marvelous things (e.g.
automatization in the production of technical devices or cars), but the respective industries had to invest
billions of $ into research to even get this far.