My take on Merit

[cryptocrusher]

So it's been a day or so now since merit was introduced and I've shared some of my thoughts and opinions on it across a number of threads, now I feel I'll compile them all in one place.

1. Is the initial amount of merit people received fair?

Many people have complained at the unfairness of receiving the same merit points when they were 1 week away from a rank compared to someone who just ranked up. Frankly they're right and it isn't fair. However it's hard to have done anything that would have been fair, the most logical would have been to give everyone merit = activity (except for newbies and juniors) but even if we had settled on everyone receiving the merit between their ranks it would've been a much fairer solution overall. In this scenario all Members would have received 55 merit, all full 175 merit etc.

For anyone saying that they should've waited until the next activity period - the argument is completely false because there would be people each activity period who are one activity period away from a new rank.

I think it's safe to say this could've been introduced in a better way but now it's too late and it's just going to have to be something people suck up and deal with.

2. Will it combat spam effectively?

To answer this I think we need to consider the two main sources of spam - account farmers and signature campaign exploiters. Account farmers build up accounts and rank them up to sell them on. Signature campaign exploiters have many alts that participate in signature campaigns. So how do these changes effect each one?

Account Farmers

Account farmers are the ones that will be hit the hardest and this is where we'll probably see the biggest change. They can no longer just spam there 14 posts per activity period to ensure they achieve the highest rank, or even 1 post per activity period until they reach the potential activity they desire and then spam all their posts at once. They will now have to produce constructive posts in order to rank the accounts beyond Jr Member status. Mostly the farmed accounts are sold to people to use for signature campaigns (will probably require member+ we'll cover that next) or just because they want a high ranking account. This will make a huge difference in that many will simply not try and those that do try will have to up their post quality. In short, merit will probably combat this problem very well.

Signature campaign spammers

The second source of spam in the forum comes from signature campaigns, I myself have participated in campaigns and am a big supporter but it's no doubt people create multiple accounts and spam the forum as a result of such campaigns. Before we start, it's important to note that with the changes, junior members can no longer have links within their signature, this in turn will likely lead to signature campaigns being for members and above.

The merit system will struggle to combat this in a number of ways. Firstly all accounts that are already member plus will still be able to participate in campaigns and their post quality can remain exactly the same. This is going to make up a large number of the accounts. It can act as a deterrent to new accounts being created as it will require some quality posts to achieve a higher rank and higher rewards but still people can just farm member accounts and use these, once they have their 10 merit then their post quality does not matter.

Even worse, if signature campaigns decide to still allow Jr Members (this is entirely possible) and accept that they won't have links in their signatures then the spam could actually become worse. In the environment whereby accounts cannot rank up beyond Jr Member without quality posts there is no incentive to write anything of use or substance. Previously signature campaign participants had an incentive to post quality posts because they did not want their accounts to be banned for spamming and lose all of their progress. It would take months to reach a new rank and having to start again is a big set back. If Jr Members are still allowed in campaigns there is very little risk in regard to their accounts being banned for spam, they can simply create another Jr Member in approximately 1 month. In short, I think merit will have mixed effects here and won't be nearly as effective as it will be against account farmers.

3. Is the way that sMerit works well thought and will it be fair and effective?

The math

By now you probably know that the only long term source of sMerit is from selected accounts. Theymos has said that in total there will be 8175 sMerit to be distributed each month. Before we actually look at how this sMerit may or may not be fairly distributed let's just break that down with some numbers.

Each account previously if fulfilling its potential growth would grow by 30 activity per 30 days. Given that, dividing 8175 by 30 we get 272.5. That means that 272.5 accounts with the new measure could grow at the rate they were previously able to grow. Now as of writing we have approximately 1.7m accounts in this forum, of course many will be inactive but even removing those it's a lot more than 272.5

An alternative way to look at this point is to consider the posts per day. If we take the average number of posts per month from January 2015 to December 2017 (therefore discounting for the fact the forum has got a lot busier in the last months than at the beginning of this time period which would only serve to worsen things) we get an average monthly number of posts of 451,274. Given that of those 451,274 posts there will be a total of 8175 merit given out we can calculate that per post we are looking at approximately 0.0018 merit being given per post, or alternatively 1 in 55 posts earning 1 merit. Compare that to previously where a user could assume that per 1 post they'd gain 1 activity up to the limit then it's a much slower rate of progression. These figures were calculated from the official forum stats if anyone wants to check my calculations.

The logic

The math is all very good at showing that there's probably not enough sMerit to go around but it also fails to highlight a number of other things.

Will sMerits be fairly distributed across ranks and sections of the forums? I've seen many concerns that lower ranked members will never receive merit and that it will in fact always go to the higher ranked members of the forum, undoubtedly there is some correlation between post quality and ranking but it does not mean to say that a newbie can not have excellent post quality, will they be rewarded for that, that remains to be seen. Also, given that there are currently about 30 merit sources, will they check enough of the forum that all areas are covered and peoples contributions are rewarded, it's highly likely that they are more likely to be in sections that are regarded as more 'elite' such as 'development and technical discussion' or 'serious discussion'. Newer ranked members are less likely to post in these sections and even some higher ranked members may chose no to, I myself do not and find I am much more able to offer advice in other sections of the forum, it remains to be seen how the spread of sMerits will be across the forum. One seemingly obvious solution to this would be to have more sMerit sources, it does not have to mean more circulating sMerits but more sources would mean that more of the forum is likely to be scoured for good quality posts. Furthermore I think it is worth considering sMerits not being spendable/giftable to legendary accounts, they are effectively wasted as there is nowhere for them to progress from Legendary.

What of sMerit sources that become inactive, we only have a finite amount already, we can be sure that some of the 30 or so sources will become inactive and that even further reduces the scarcity of sMerit.

One thing I feel would be a nice solution in the long term would be to have some sort of way to nominate yourself or others for consideration for a rank increase or a merit boost. Undoubtedly some people's contributions will go unrewarded and they should be able to speak out about it. Of course some consideration of controls would be needed so as to not have many Jr Members spamming saying they deserve more merit.

From both the intuition and the math it would seem right now that we are set for a period of great scarcity and ranks will be much harder to progress throughout, of course it's early days and this could change but those are my initial impressions.

4. Why do ranks even matter, it's not like it makes a difference?

This is one question I found myself answering earlier and the real short of it is this. Rightly or wrongly people can make money by posting in this forum, as they increase ranks, their earning potential increases. The community is incentivised to contribute and help one another by such things as signature campaigns. If we end up in a scenario whereby the elite remain elite and those at the bottom remain at the bottom then it will only promote discontent within the forum. It is vital for the forums success that there is always a progression and a constant stream of new members. If these new members feel that the same opportunities are not afforded to them then they won't stick around.

5. Summing up

This covers most of my initial thoughts on the new merit system, I see a lot of good in it and a lot of potential but also it has a feel of something that was perhaps a little rush and not well thought out. It remains to be seen how well it will function and we should all give it time but subsequently there should be a willingness for people to listen to feedback and consider change. Spam will certainly be reduced in the forum but a correct balance needs to be found between limiting spam and limiting opportunity for new members. If things don't work out and sMerits are too scarce I hope that this is acknowledged and changes are made whether that be in increasing the number of sMerits in supply or reducing the merits required to rank up.

I'm really interested for some feedback and your inputs on this, especially your feelings regarding the total supply of sMerits and if you feel like it's enough to sustain the ranking up of accounts. If anyone has any access to more detailed numbers about things such as the amount of activity gained each month by the whole forum or anything similar please let me know as I'd be really interested to look in to it.

Now feel free to agree or disagree with some of my points but let's try keep it to constructive criticism and discussion instead of just flaming people.

[findingthemoon]

You are completely right that there is nowhere near enough smerit to go around, I suspect perhaps theymos is going to increase it after the initial smerit distribution is allocated, but he will need to increase it by a very large amount to even make a dent. Distribution of merit aside, I don't think the idea is that an average user should post 500*50= 25 000 posts to become a hero member. I also don't see theymos having the time to add and supervise hundreds of smerit sources.

Another point you didn't mention is that posts after the first few replies are much less likely to receive any merit as they won't be seen much, maybe it is good in the sense that some of those huge useless threads will be less lucrative to post on but it also gives less incentive to continue posting on good discussions.

As for distribution of merit, we can already see from the stats theymos posted that most of the merit is going to a few well known people in the community rather than being distributed evenly to good posts. Also it seems this meta section has been the one section where most of the merit so far has been distributed to.

Theymos also asked people to merit posts of high quality rather than posts they agree with but just as with likes on reddit this will never happen.

[cryptocrusher]

You are completely right that there is nowhere near enough smerit to go around, I suspect perhaps theymos is going to increase it after the initial smerit distribution is allocated, but he will need to increase it by a very large amount to even make a dent. Distribution of merit aside, I don't think the idea is that an average user should post 500*50= 25 000 posts to become a hero member. I also don't see theymos having the time to add and supervise hundreds of smerit sources.

Another point you didn't mention is that posts after the first few replies are much less likely to receive any merit as they won't be seen much, maybe it is good in the sense that some of those huge useless threads will be less lucrative to post on but it also gives less incentive to continue posting on good discussions.

As for distribution of merit, we can already see from the stats theymos posted that most of the merit is going to a few well known people in the community rather than being distributed evenly to good posts. Also it seems this meta section has been the one section where most of the merit so far has been distributed to.

Theymos also asked people to merit posts of high quality rather than posts they agree with but just as with likes on reddit this will never happen.

Theymos has said he'll be monitoring it closely and adding more merit sources as he sees fit, hopefully that will be the case and we won't end up in a scenario where ranks are much much harder to come by. I don't think anyone will have a major issue with a slower rate of progression, even if it was halved or so, but if it's 50 times or more then it's a bit ridiculous.

I think that's something that will need addressing and I'm not entirely sure as to how that can be done, hopefully in time if there are more sMerits to be spread around then people can reward good quality discussion. As you mention in your next point it's important that they're used to reward people who are challenging one another and learning instead of just the people that you are in agreement with.

[Ctn]

That is good analysis of the sMerit system. I loved the point where you talking about the "anti-spamming"stuff. Yes this system is going to overcome these problems surely and there won't be any shitty accounts anymore henceforth. I think we will start seeing the effect sooner than we think because by this time most of the account farmers might be running here and there to beg the merits. They know very well, if they don't achieve some in the future then they don't re-gain next month either, this cycle will completely halt for them at some point leading to broken chain of their farming.

Finally, this forum is going to take fresh air!

Pointing myself at your second last point :

It is vital for the forums success that there is always a progression and a constant stream of new members. If these new members feel that the same opportunities are not afforded to them then they won't stick around.

You speak the truth. If the new comers don't understand this system then they might just go away. Everyone who join the forum, join it with the desire to earn money (yes everyone, thats truth)and if they don't see anything quickly then they might just fly away.

But on the other hand I do think that when a newcomer will come, he might just think that the forum was like this since the first time and he might just stat posting in qualitative ways right from the first day.

Thats like even better for the forum! Hope this system really workout the way it is designed and fulfil the purpose.

[Jet Cash]

I think the merit system would be much more effective if it was only possible for a member to give one point to a post. I had a quick look at the posts that were getting 50 merits, and they were either a couple of years old, or posts about ponzi scams or air drops. It's hardly going to improve the forum by promoting ponzi pushers.

[vegita1233]

Theymos has said he'll be monitoring it closely and adding more merit sources as he sees fit, hopefully that will be the case and we won't end up in a scenario where ranks are much much harder to come by. I don't think anyone will have a major issue with a slower rate of progression, even if it was halved or so, but if it's 50 times or more then it's a bit ridiculous.

Going by your numbers:
Another viewpoint of math is that this is essentially asking the current 50 merit sources to classify 451,274 posts per month or 15024 (451274 / 30)
 posts per day per merit source.  This gives each merit source about 5.75 seconds per post in a 24 hour work period to read / judge / and assign a merit score.  

On top of this the merit score doesn't really discourage the old timers from posting shitposts either for campaign purposes.  The lower ranks still have to read those..

[Javi_Anibarro]

I think the merit system would be much more effective if it was only possible for a member to give one point to a post. I had a quick look at the posts that were getting 50 merits, and they were either a couple of years old, or posts about ponzi scams or air drops. It's hardly going to improve the forum by promoting ponzi pushers.

i could not agree more,
Theymos need to consider it.
also it would be better to set the limit and time.
except the merit sources,
it can prevent people from abusing their Merit.

[krishnaverma]

On top of this the merit score doesn't really discourage the old timers from posting shitposts either for campaign purposes.  The lower ranks still have to read those..

In a recently launched campaign, all rank members are being given higher payout if they have some earned merits. With time, we can see the earned merits being given more importance in these campaigns for old timers as well.

[actmyname]

I have some points to make, in response.

1. Is the initial amount of merit people received fair?

Many people have complained at the unfairness of receiving the same merit points when they were 1 week away from a rank compared to someone who just ranked up. Frankly they're right and it isn't fair. However it's hard to have done anything that would have been fair, the most logical would have been to give everyone merit = activity

I agree somewhat with distributing the base merit in correlation to the activity. This would be the fairest way to take the Hero Members (in the Legendary zone) into account since that is based on random chance. Don't want to ruin the surprise.

3. Is the way that sMerit works well thought and will it be fair and effective?

The math

By now you probably know that the only long term source of sMerit is from selected accounts. Theymos has said that in total there will be 8175 sMerit to be distributed each month. Before we actually look at how this sMerit may or may not be fairly distributed let's just break that down with some numbers.

Each account previously if fulfilling its potential growth would grow by 30 activity per 30 days. Given that, dividing 8175 by 30 we get 272.5. That means that 272.5 accounts with the new measure could grow at the rate they were previously able to grow. Now as of writing we have approximately 1.7m accounts in this forum, of course many will be inactive but even removing those it's a lot more than 272.5

Take into account the trickle-down effect. If they are members that are posting high-quality content then hopefully they will redistribute their sMerits as well. This means that we have an infinite summation: ∑_i(8175)/2ⁱ = 16,350 which is double the amount, so we're given about 1 in every 28 posts. Not a lot but it's something to keep in mind. Also, with the forum's current status I think that number is pretty high (if we're talking constructive posts in ratio to unconstructive posts)

With the proposed 100 (?) sMerit sources by the end of the year I assume that it'll reach 20,000 sMerit each month which results in 40,000 merit being distributed to those 450k posts, leading to approximately 1 in every 11 posts. Since we are in a forum, wouldn't it be good to see the spam:quality ratio turn from 49:1 to 10:1? Seems like a vast improvement. And if it doesn't improve, then that's fine: people who are making quality posts will get more merit.I personally had an idea for signature campaigns, where the top poster of the week gets 10 merit (or some amount). This would incentivize better post quality within the campaign.

[HashFace]

I think if I had sum up my feelings on the merit system in a few words ...

It was designed to punish bad posters and spammers, but will likely imped "normal" posters as well.

But that's not to say it can't be fixed if problems become apparent.  The Admins will hopefully look at this system again in a month or two and make adjustments as needed. 

But your point about only a couple hundred posters being able to stay "on track" for ranking up hits home.  It makes the case that not only will the bad posters be "punished", but pretty much everyone will be "punished". 

[onurgozupek]

In most local topics, you can see merit buying/selling posts. So anyone hoping Merit will increase the post quality will be disappointed...

[Thirdspace]

In most local topics, you can see merit buying/selling posts. So anyone hoping Merit will increase the post quality will be disappointed...

once found out, expect them to get tagged by DT1/2 and received neg trust
I believe this merit thing would slowly change the quality of the posts we see on this forum
only time would tell, lets hope the best for our beloved forum

[Sexy_Sees]

OK, so only big timers in here. So, I feel a little intimidated here, but here it goes. I can imagine you all already earned your stripes and I am considered to be a 'new kid on the block'. But I have a Signature Campaign accepted just recently and just like I do at work, I deliver. This whole new system with merit integrated has a lot of benefits, I must agree! On the other hand it also feels slightly dishonest.. I probably will be a Sr Member in around 2-3 years
Besides that, it's a Monopoly right now. Heroes and Legendaries own the hotels and us little ones will never win the game anymore.. The rich are getting richer and the poor are getting poorer. You feel the nasty taste it gives the people from the bottom and the supporting shouts from the top of the foodchain. It's a model we probably have to get used to now. We live, we learn, we adapt 

GB

[TheQuin]

I can imagine you all already earned your stripes

This is the case, people earned that by making a contribution to the forum over a long period of time.

I probably will be a Sr Member in around 2-3 years

It will take a minimum of 8.5 months as the activity system is still in place and the merit system is an additional requirement. Have a look at the OP to this thread and you will see that it has already attracted 14 Merit. Is it impossible to make 1 good post like that every couple of weeks?

The rich are getting richer and the poor are getting poorer.

This makes no sense. Nobody has had any status taken away from them so it cannot possibly make them poorer. The only change is that it now takes some effort to get somewhere, it is no longer an automatic right. You might feel aggrieved that those who came before you had an easier ride, but that is often the reward for being an early adopter, not just in Bitcoin but in life generally. The early bird catches the worm.

[dncdog]

I can imagine you all already earned your stripes

This is the case, people earned that by making a contribution to the forum over a long period of time.

I probably will be a Sr Member in around 2-3 years

It will take a minimum of 8.5 months as the activity system is still in place and the merit system is an additional requirement. Have a look at the OP to this thread and you will see that it has already attracted 14 Merit. Is it impossible to make 1 good post like that every couple of weeks?

The rich are getting richer and the poor are getting poorer.

This makes no sense. Nobody has had any status taken away from them so it cannot possibly make them poorer. The only change is that it now takes some effort to get somewhere, it is no longer an automatic right. You might feel aggrieved that those who came before you had an easier ride, but that is often the reward for being an early adopter, not just in Bitcoin but in life generally. The early bird catches the worm.

The big problem is the probability to get merits for good, high-qualitied posts is extremely low. Right? I guess, it is lower than 50%. To clarify, with 100 high-qualitied posts, posters can get merits for their 50 posts or even much more lower than the 50%. That's the problem. It is due to the fact that other users might not have time to read all topics, all threads, and they might not familiar with new merit system, or even not ready to send their sMerits (there are still lots of confuse, misunderstanding about Merits and sMerits, right?).

[TheQuin]

The big problem is the probability to get merits for good, high-qualitied posts is extremely low. Right?

Not really. I've been surprised by the points I've picked up so far. I just had a quick look at the last page of your post history and other than talking about the merit system (which we all are right now, it's a hot topic) I noticed most of your posts are in altcoin ANN threads. I don't really hang out over there so I don't if there are people giving out merit or not. In the Bitcoin technical sections, it seems not to be an issue at the moment.

[Husires]

What will happen next Wednesday?
Some accounts will have the allowed activity period but have not Rank up because of the new system.
They will be required to publish daily for some Merit.
I agree with you will be long posts or useful but generally not useful because they are mandatory shares.
hope theymos think again about this point.

[LoyceV]

I agree to most of your post, but there's one thing I disagree with:

Furthermore I think it is worth considering sMerits not being spendable/giftable to legendary accounts, they are effectively wasted as there is nowhere for them to progress from Legendary.

Merits can also be used to select posts worth reading. By excluding certain members, you make Merits less valuable to the forum. I assume (in the long term) theymos will give all sources combined enough sMerits to give them to all members who post quality. If 50% goes to Legendary members, the total can just be doubled.

Update: I forgot to add Legendary members can also "pay it forward" by giving away the sMerit they receive with their Merits. If you exclude Legendary members from receiving Merit, you also exclude the most experienced users from giving them away.

[1Referee]

They will be required to publish daily for some Merit.

No one is required to post for Merit. If you just want to post randomly throughout the week, then that's fine too. Sure, the chances of you not getting more Merits will shrink a bit, but the main point is that people should be posting naturally, and not for the sake of Merits.

I agree with you will be long posts or useful but generally not useful because they are mandatory shares.

People are forcing and exaggerating in their posts to get Merits, that I have noticed too, but that's still better than spamming with 'yes I think the price will go up', or 'Bitcoin will make us rich, but we have to trust in Bitcoin'. Everything that may potentially enhance the discussion field in a thread is a welcome addition, because that's what's missing now.

[A1exander]

So it's been a day or so now since merit was introduced and I've shared some of my thoughts and opinions on it across a number of threads, now I feel I'll compile them all in one place.

Excellent analysis.

The math

By now you probably know that the only long term source of sMerit is from selected accounts. Theymos has said that in total there will be 8175 sMerit to be distributed each month. Before we actually look at how this sMerit may or may not be fairly distributed let's just break that down with some numbers.

Each account previously if fulfilling its potential growth would grow by 30 activity per 30 days. Given that, dividing 8175 by 30 we get 272.5. That means that 272.5 accounts with the new measure could grow at the rate they were previously able to grow. Now as of writing we have approximately 1.7m accounts in this forum, of course many will be inactive but even removing those it's a lot more than 272.5

An alternative way to look at this point is to consider the posts per day. If we take the average number of posts per month from January 2015 to December 2017 (therefore discounting for the fact the forum has got a lot busier in the last months than at the beginning of this time period which would only serve to worsen things) we get an average monthly number of posts of 451,274. Given that of those 451,274 posts there will be a total of 8175 merit given out we can calculate that per post we are looking at approximately 0.0018 merit being given per post, or alternatively 1 in 55 posts earning 1 merit. Compare that to previously where a user could assume that per 1 post they'd gain 1 activity up to the limit then it's a much slower rate of progression. These figures were calculated from the official forum stats if anyone wants to check my calculations.

My ballpark estimate was 1 in 50 to 100 posts. There will be a trickle-down effect, but there also will be a large number of points received by Legendary members and these may cancel each other. With these numbers it may be possible for a Newbie to progress to Member, but not further. Higher ranks are essentially frozen, since they would have to make many thousands of posts to get enough merit points.