Topic: An idea to democratize merit assignment to all levels (complementary) (Read 174 times)

I want to make it clear that I believe your premise is flawed, before we go any further. The "bottom rank layers" should not have an excess (or any) merit to distribute, because they would not distribute it properly. They do not deserve merit from the beginning, they do not have knowledge from the get-go and they are not knowledgeable in every imaginable way (or shouldn't be, if they are truly new). Therefore, I do not believe it needs to be addressed.


Do you believe that short posts cannot be worth merit, constructive, helpful, meaningful or informative?

Cause you're wrong, if so.


Why? Who would care or notice when someone picks their own post as worth merit?

There are already many threads of people actively looking for users to submit their posts (or others) that they believe to be under merited. They are then reviewed and receive merit or tips on how to improve themselves and their contributions.


So, if someone farms 100 alts they could easily mark their own posts and vote them up for guaranteed merit. This would encourage abuse in the worst possible way; I cannot think of a better way to subtly ask for a system to be abused. Limits every 14-days would be meaningless to someone that has almost endless accounts.


Democratization is only feasible if one person cannot falsely portray themselves as multiple people. Otherwise, it will be corruption. If you cannot verify the identity of those "voting" in the democratization process then it is meaningless and presumably skewed/abused.

Also, merit being generated automatically makes no sense. Again, it would just encourage people with multiple accounts to farm to a minimum level of "votes" and then sit pretty, while most of the "votes" were them to begin with. It would be like if you elected a President, but didn't know/care about the identity or how many times someone is voting; worthless. It ends up giving the ability and the most incentive to vote/abuse to the person that is running.

Yes and that's why I made a thread here : https://bitcointalksearch.org/topic/joeljantsens-merit-distribution-thread-newbies-welcome-2971165 (a way people can get merits even if they can't earn merits through constructive posts  )


That's bad because,
a) A post could be one liner,just one word or an image but highly constructive.
b) A post could be an essay or a paragraph yet contribute nothing to the topic.

This is just like the current signature problem,people get paid for shit posts just because they're 'lengthy' (and merit system was made to solve that error )  

Honestly,you will have everyone marking all their posts because no one here thinks their posts are trash until they're banned from SMAS or something.Merit system again keeps it subjective and let's anyone give merit if he finds a particular post helpful.

You know with the advent of account farming,it is very easy to cheat that system as one will obviously use alt accounts to mark posts as 'meritable'.

1. No. We should encourage users to post briefly and to the point (hint, hint).

2. In discussion threads, users should only post stuff they consider merit worthy in the first place.

3. and 4. No. The idea behind merit is that it is subjective and the more you have made useful contributions, the more merit giving power you have. We don’t need ‘democratic’ merit; if someone can’t earn smerit, they shouldn’t be able to give merit.

I’ve been looking over the rules of generation and distribution of the Merit System, trying to see if there is a way to improve it, whilst not complicating in excess the base of the system, and at the same time broadening the scope so as to allow all participants to Merit others without creating too much merit inflation.

The current system has a sound set of objectives, being in summary to enhance the quality of posts in general.
Regardless, the merit system is not going to be able to stop junk contributions on its own, but it should narrow it down a chunk (no one is going to write only meritable posts for many reasons, but we could move to lets say 1 every 100 to 10 every hundred as users move to making keener efforts to acquire merit for ranking purposes).

Not all posts are going to be meritable as we know. For example, joining bounties won’t be, neither will those small posts created to encourage someone. Frankly, neither will many post that are used as stepping stones to boost activity levels. But some are now going to need quality and visibility thanks to the merit system.

The current merry system is rather rank based, and perhaps it should, but it does leave the bottom rank layers without a word in terms of merit assignment. To address this, one possibility would be as follows:

1. Set a minimum length for a post to opt to being meritable (let’s say a minimum of 1.000 characters).
No post under the minimal length may be merited.

2. Let each user (even newbies) be able to mark (somehow) with a visible icon one of his own posts which he considers could be meritable.
Let’s place a limit on the number of self-chosen meritable posts (1 or 2 per 14d for example).

3. Users will be able to see other people’s self-marked meritable posts and cast their meritable/non-merritable vote for it. This vote cast would not be merit based (i.e. any user could cast it, maybe with limits per 14 days too so as not to over-use the vote per se).
Let’s limit the number of votes a self- marked meritable post can receive (let’s say 100). Once the 100 votes are reached, the post cannot receive anymore votes.

4. Based on the number of positive meritable posts received, the post qualifies to receive “system generated merit”.

For example:
 * Under 70 meritable positive votes -> no merit assigned.
 * Between [70 and 80) meritable positive votes -> 1 merit assigned automatically.
 * Between [80 and 90) meritable positive votes -> 2 merit assigned automatically.
* Above 90 meritable positive votes -> 3 merit assigned automatically.

All the above figures are tentative, and should be backed or modified in accordance to a bit of mining into the forum underlying data so as to set the appropriate parameters and no create merit inflation.

The above is just an idea to democratize merit assignment to all levels, give visibility to users trying to merit their (good) posts, and at the same time using a basis of consensus to generate merit. It’s a small complementary add-on to the current system. Or it could be simply an idea to discard.