[Experimental] Bitcointalk quality score

[fillippone]

What the hell? The posts are all empty and archived!
How can he have all these Merits?

Let's keep it short and say he went full drama, and burned his reputation and contributions.

Unfortunately, however, it is not an error in the formula! It is a borderline and isolated case, I would say.

I'd say this is something your formula should account for. I just don't know how, other than manually adding exceptions.

There are many users doing strange thing.
Symmetrick archived his posts and retained his merits.
Boblawblaw deleted almost his posts and became a newbie again. Then regained his merits.
I am curious to see his score as well, even if his past should be "invisible" now.

[LoyceV]

Boblawblaw deleted almost his posts and became a newbie again. Then regained his merits.

I didn't know that. BobLawblaw is currently Legendary with 1932 Posts (and Activity). I have 6889 archived posts, so it looks like his deleting posts to be a Newbie again must have been a while ago.

[giorgione]

Hello everyone, I am opening this thread to present a small script that I created experimentally, hoping that everyone will like it.

This board often discusses the quality of posts, but it is a somewhat peculiar topic since there is no precise metric to determine with certainty whether a post is of pure quality or not.
Therefore, I have created this tool that indicates how much quality users have or lack, and how many quality posts they have created during their time on the forum.

The tool uses data from the Tryninja website (thanks as always) bitlist.co and from the forum itself.

The data analysed are posts, Merit and profile age (data from the forum) merit 120 days and posts 120 days (data from bitlist.co).

The formula used is as follows:

I'm always amazed by your work. I'm really happy to have such talented people on the forum who contribute in this way. I truly say this from the bottom of my heart. I'm really happy that there's so much effort put into this section!
So, as others say, perhaps it's a good idea to put the numbers in order a bit.

[*Ace*]

Boblawblaw deleted almost his posts and became a newbie again. Then regained his merits.

I didn't know that. BobLawblaw is currently Legendary with 1932 Posts (and Activity). I have 6889 archived posts, so it looks like his deleting posts to be a Newbie again must have been a while ago.

I updated the logic
I left post120 === 0
Merit120 === 0
Inactivity ≥ 2 years

If by chance they receive even a single merit, the tool always labels them as former elite, and in order to return to their real score, there must be a condition that they must post during the reactivation period, i.e. during the last 120 days, the user must post something on the forum on at least 2 separate days.

[fillippone]

Boblawblaw deleted almost his posts and became a newbie again. Then regained his merits.

I didn't know that. BobLawblaw is currently Legendary with 1932 Posts (and Activity). I have 6889 archived posts, so it looks like his deleting posts to be a Newbie again must have been a while ago.

Something can be seen here:

This may likely be the last text post I make on this site.

Going to make an attempt to just post relevant market information without any comment. No more meming. Increased my ignore list, as an unfortunate side effect of everything that has gone on, after much self-reflection. Really have no desire to interact with people here any more, but recognize the thread has its place of importance in the Bitcoin sphere, with regards to staying on top of news.

Hope I'm able, at the very least, to share what I see happening on Twitter.

Good or Bad. Draw your own conclusions sorta thing.

Then, there are references to all his post deletions:

Thanks for that! The WO is already mentioned in some of the upcoming videos and there's a dedicated video upcoming about the Wall Observer thread. Anything special you guys want to see featured, something that we definitely shouldn't miss?

This is actually a nice one!
What are the most iconic WO moment?
I think I am quite new here, but I can think a few of those:

• The births of the hats
• @Mindtrust selling
• @BobLawblaw ragequit with epic post deletion
• @El duderino_ drunk dickpic
• SOON (TM) @Proudhon paying his bet to Fillippone

Merit shower to the one who’s able to link the above post!

What is your top-5?

[dkbit98]

What the hell? The posts are all empty and archived!
How can he have all these Merits?
Unfortunately, however, it is not an error in the formula! It is a borderline and isolated case, I would say.

It's a long story with that account, but I suspect old owner sold his account and he received a lot of merits before that for his activity.
I don't know who deleted all posts, old or new owner, but that is not important for quality score.
There are more examples where your quality score would not work correctly, like I said it's not easy to measure quality.

[*Ace*]

What the hell? The posts are all empty and archived!
How can he have all these Merits?
Unfortunately, however, it is not an error in the formula! It is a borderline and isolated case, I would say.

It's a long story with that account, but I suspect old owner sold his account and he received a lot of merits before that for his activity.
I don't know who deleted all posts, old or new owner, but that is not important for quality score.
There are more examples where your quality score would not work correctly, like I said it's not easy to measure quality.

No, it's not easy to measure quality, but unfortunately it's not the formula's fault if a high-quality user hasn't received enough Merit; the problem lies elsewhere, unfortunately! So I think I'll stop this script, or rather, I won't continue to modify the formula!
The score it gives is in line with every user's expectations. Obviously, there will be borderline cases, but I can't create a formula for every specific case; I'd go mad writing and rewriting the code.
Also, every time I fix one part of the formula, another part breaks, and it becomes an endless cycle 🤣



Just to understand better, how would you describe this data? Reliable? Or unreliable?

[Vod]

So it looks like you are developing two different scores at least.  One for active users and one for inactive.

My feelings - everyone changes and they don't always improve over time.   Past quality posts should not be discounted because of subsequent posts.    If you are going to compute the value of a profile based on their posts, then don't allow that to dilute over time.

Imagine if Einstein was remembered as an "idiot" because his final years were spent denying quantum physics.

[*Ace*]

So it looks like you are developing two different scores at least.  One for active users and one for inactive.

My feelings - everyone changes and they don't always improve over time.   Past quality posts should not be discounted because of subsequent posts.    If you are going to compute the value of a profile based on their posts, then don't allow that to dilute over time.

Imagine if Einstein was remembered as an "idiot" because his final years were spent denying quantum physics.

That's not quite right. There is only one score, which applies to both active and inactive users. The final rank is determined by reliability if a user is very young. This is to prevent a newbie with few posts and lots of merits from having a better score than a user like fillippone, who has a significant history on this forum.
Then there is an increasing penalty based on the length of inactivity, which will decrease the score based on the time elapsed since the last access to the forum and the last post written.
In the case of simmetryc, I put a special check since he has had no posts or merits in the last 4 months, so the final score is 0.

But for everyone else, the score of the formula is that of the reputation score.

I don't know if I'm explaining myself well! It's also a bit difficult to explain the formula 

[Vod]

Then there is an increasing penalty based on the length of inactivity, which will decrease the score based on the time elapsed since the last access to the forum and the last post written.

This. Why are you reducing post quality over time?  Have a seperate formula for reliability (if they get bored of bitcoin or stop living for example), but don't adjust the quality of past posts.   

[*Ace*]

Then there is an increasing penalty based on the length of inactivity, which will decrease the score based on the time elapsed since the last access to the forum and the last post written.

This. Why are you reducing post quality over time?  Have a seperate formula for reliability (if they get bored of bitcoin or stop living for example), but don't adjust the quality of past posts.   

Name:
Vod

Posts:   22700
Activity:   4312
Merit:   3497
Earned Merit:   2497
Final Rank:   19.22 — 🟢 Good user
Reliability:   100%
Reputation Score:   19.22

The final rank is reduced, only that. But as you can see, the reputation score is the same and will never decrease due to penalties, this is the real and only score.
Satoshi's final rank, for example, has decreased to 10% of the total reputation score, but if he were to become active again, he would regain his total score.
It is a temporary decrease, so to speak.

[Vod]

So it's more of a recognition score, like BPIP, not a quality of content score.  A post does not lose quality over time, but the recognition of the author does.    Obviously it's your new project idea, I just want to try a little consistency with terminology.