Unintended Consequences
“Habr’s karma system and its impact on users” is a topic for at least a coursework
I could start this article by saying that I have been reading Habr for a long time, but this will not be an entirely accurate statement. The correct thesis would sound like this: “I have been reading articles from Habr for a long time” - but I was not interested in what was happening inside the community when I finally decided to register this spring. This is a typical mistake of a person who comes to Habr from a search engine to read useful articles about the intricacies of programming or interesting news from the world of technology. As long as you only see the portal from this positive side, you don't wonder what's going on under the hood. Of course, in the comments or articles, references to karma slipped from time to time - but after all, almost all major portals have karma (I naively thought), this is normal for self-regulating Internet communities.
I had to seriously think about it after I suddenly lost the ability to write more than one comment in five minutes.
At the same time, outwardly, everything was going great: my comments were always positive, my rating was growing - and suddenly it turned out that I had negative karma. All my long experience of Internet communication, all user habits, and even banal common sense shouted to me that this was some kind of mistake: the approval rate of a site user by other site users cannot rise and fall at the same time! But I decided not to cut off the shoulder, but to conduct a small study, both analytical (in the form of studying users' opinions about karma) and statistical (in the form of analyzing account performance).
The history of the war of users with karma turned out to be very rich. With varying success, it has been going on for more than a decade, with dozens of blocked victims and several deleted articles on its account. And, oddly enough, my problem (the discrepancy between ratings and karma) is practically not used in the argument - even in the days of the open API, these calculations were not used. Only one commentator came closest in a relatively recent entry:
“Actually, what is interesting to find: are there people downvoted for karma with big pluses to their comments?”
In the statistical part, you can see that yes, there are such people. But even without statistics, users, in principle, understood everything about karma for a long time.
Here's a post from ten years ago:
The big problem on Habré is that there are many users who put a minus in karma according to the principle: “Oh, you have an opinion different from mine, here’s a minus in karma for you.” Although for me such a reasoned comment with counterarguments and a well-stated opposite position does not even deserve a minus for the comment itself, not to mention the karma of the author. Unfortunately, on Habré there is practically no culture of reasoned dispute and respect for a strong opponent, many tend to just throw hats on.
In general, I have an opinion that the division of ratings into two counters "rating" and "karma" is unintuitive and therefore incorrect and inefficient.
Here's an entry from five years ago:
Only cases were analyzed when karma changed by at least 15 units, but this does not change the picture as a whole, tk. and in this case the ratio is 30% to 70%. As you can see, they drain karma mostly because of comments, but raise it for written articles.
Here's a three-year-old improvement proposal:
Sentence:
Allow article authors to vote for karma only in a certain period (for example, a week) after they publish an article. If a person has not published anything in the last week, he cannot be drained of karma for commenting. The rule does not need to be applied to reed-only accounts - they gain karma with useful comments.
A comment:
Too often, Habr users complain about the drain of karma for objectionable comments in other people's posts. For example, in this post the problem is described back in 2012. Woz and now there.
Here is another dialogue from three years ago on the same topic:
I can tell you why I stopped writing comments (I'll make this one an exception): because it's hard to earn karma, because for it you need to constantly generate some articles, but it's very easy to lose it. It is not true that if you write correctly, it is not wasted. It can be lowered for a variety of reasons: he did not agree with you in the dispute, he considered that some fact in the comment was incorrect, or he was simply in a bad mood.
Yes, this is an ancient disease of the habrasystem. It was assumed that those who have positive karma are adequate and will not just downvote anyone. Once upon a time, everything was even worse - the more karma, the greater the minus the user can put, which ended up with a couple of "starred" habrausers handing out left and right to anyone at -6, -8, after which the possibilities were cut to one. The creators of the economy of karma initially apparently did not take into account the perversity of anonymity
It seems to me that this system should have been re-balanced a long time ago by the fact that when someone votes, a certain amount of it is deducted from the user's karma. You don't need much - 0,2-0,5 is enough. This would greatly increase the responsibility of voters when choosing to vote for someone or not.
And finally, comments on a post from the beginning of this year:
Karma is not a very good tool for system self-regulation. Karma is often assessed by those who are dissatisfied with a person (or even with his position). As a result, it turns out that it is very difficult to collect karma, and it is very easy to drain it. This makes people think once again - is it worth expressing your opinion if it is not very popular? After all, if I speak out once, they will downvote and merge karma, and it will not be possible to express more. This leads to the fact that only one opinion remains on the resource, and all the others are forced out.
And here's a comment that explains why "writing articles" doesn't really save the karma system:
The article brings almost nothing in terms of karma, and for one unsuccessful comment, a person can be completely drained.
The problem here is the separation of rating and karma. This is how it works in people's minds:
1. Content rating is my attitude to an article or comment
2. Evaluation of karma is my attitude towards a person personally
Eventually
1. If you wrote the best article in the world, you will be given many pluses to the article (in the rating) and consider your mission accomplished.
2. If you wrote a comment that “does not fall into the stream”, then you will be minus the comment, and even a person, you apparently think so yourself if you think so, therefore it’s on you and in karma.
Many dissatisfied with the system of karma speak out in the sense that this is a purposeful policy of the administration - or . There is, of course, a lot of indirect evidence for this:
- The API was removed so that it was not possible to monitor the dynamics;
- We made a dynamic rating so that it was impossible to see the total ratings directly in the profile;
- They constantly refer to the “karmograph”, according to which there are more pluses than minuses (there is no talk of the ratio of karma and grades);
- There is a lot of, but unfounded, talk about how karma reflects the quality of posts and comments (which contradicts the statistics, as we see in the ratings).
I also remind you that nowhere and never was the justification for the existence of karma in the form in which it exists.
There is no way we can prove these conspiracy theories. But it seems to me that it’s not about them either - it’s the same problem as with people who minus karma: an impenetrable belief in their rightness, to such an extent that those who disagree with you are perceived as a “bad person”. So the leaders of Habr decided in the same way - we will evaluate users separately from their messages. And they have not been able to explain to them for more than ten years that this is an incorrect approach to ranking users. They are smart, they created a whole portal. Here you create your Habr - then we'll talk (by the way, it's funny that - "Get it first"
Personally, I assume that the very scheme of karma came to us with , where at one time most of the current owners of large Internet portals hung out. Habr began as the same Lepra - a closed club with invites and mutual assessments, dissatisfied - went away from the club. Those days are long gone, the club has not been closed for a long time, the ratings have long been given not to “another club member”, but to an ordinary user for ordinary comments and articles. But internal elitism does not let go of the administration. Everyone thinks - indeed, the guys have created a large profitable portal, have been writing articles on technical topics for many years - how can they not know something? So if everything is bad, then they, the villains, are well conceived. And in fact, administrators are just stuck in childhood. And the larger and more profitable the portal, the more difficult it is to admit one's long-term mistakes, out of falsely understood pride.
Entanglement
It's deep water, Watson, deep water. I just started diving.
Below, I will use the term "Karma" for karma, and the term "Rating" or "Total Rating" for the total sum of all the pluses and minuses that the user received, both for articles and for comments.
Having dealt with the history, we will try to look at the numbers. Recently there was a whole cycle of statistics analysis, but it concerned only the current year - I needed to understand the total user rating. Since we don’t have an API, and instead of real ratings in the profile, they show a dubious rating, I just had to study each comment and collect information about the author and rating from it. That is exactly what I did.
I opened each publication from the very beginning of time, got out of it the nickname of the author of the publication and the rating of the article, and then the nicknames of the commentators and the ratings of their comments.
Here is the main parser code.
import requests
from bs4 import BeautifulSoup
import csv
def get_doc_by_id(pid):
fname = r'files/' + 'habrbase' + '.csv'
with open(fname, "a", newline="") as file:
try:
writer = csv.writer(file)
r = requests.head('https://habr.com/ru/post/' +str(pid) + '/')
if r.status_code == 404: # проверка на существование
pass
else:
r = requests.get('https://habr.com/ru/post/' +str(pid) + '/')
soup = BeautifulSoup(r.text, 'html5lib')
if not soup.find("span", {"class": "post__title-text"}):
pass
else:
doc = []
cmt = []
doc.append(pid) #номер
doc.append(soup.find("span", {"class": "user-info__nickname"}).text) #ник
doc.append(soup.find("span", {"class": "voting-wjt__counter"}).text) #счетчик
writer.writerow(doc)
comments = soup.find_all("div", {"class": "comment"})
for x in comments:
if not x.find("div", {"class": "comment__message_banned"}):
cmt.append(x['id'][8:]) #номер
cmt.append(x.find("span", {"class": "user-info__nickname"}).text) #ник
cmt.append(x.find("span", {"class": "voting-wjt__counter"}).text) #счётчик
writer.writerow(cmt)
cmt = []
except requests.exceptions.ConnectionError:
pass
x = int(input())
y = int(input())
for i in range(x, y):
get_doc_by_id(i)
print(i)The result was such a label in the habrbase file:
I grouped users and got the result in the form of "User - Sum of his ratings" called habrauthors.csv. Then I started iterating over these users and adding data from their profile. Since sometimes the connection might break, or some strange error occurred while loading the page, we had to look at which user was processed last and continue from there.
Here is the rework code:
import requests
from bs4 import BeautifulSoup
import csv
import pandas as pd
def len_checker():
fname = r'files/' + 'habrdata' + '.csv'
with open(fname, "r") as file:
try:
authorsList = len(file.readlines())#получаем длину файла даты
except:
authorsList = 0
return authorsList
def profile_check(nname):
try:
r = requests.head('https://m.habr.com/ru/users/' +nname + '/')
if r.status_code == 404: # проверка на существование
pass
else:
ValUsers = []
r = requests.get('https://m.habr.com/ru/users/' +nname + '/')
soup = BeautifulSoup(r.text, 'html5lib') # instead of html.parser
if not soup.find("div", {"class": "tm-user-card"}):
valKarma = 0
valComments = 0
valArticles = 0
else:
valKarma = soup.find("span", {"class": "tm-votes-score"}).text #карма
valKarma = valKarma.replace(',','.').strip()
valKarma = float(valKarma)
tempDataBlock = soup.find("div", {"class": "tm-tabs-list__scroll-area"}).text.replace('n', '') #показатели активности
mainDataBlock = tempDataBlock.split(' ')
valArticles = mainDataBlock[mainDataBlock.index('Публикации')+1]
if valArticles.isdigit() == True:
valArticles = int(valArticles)
else:
valArticles = 0
valComments = mainDataBlock[mainDataBlock.index('Комментарии')+1]
if valComments.isdigit() == True:
valComments = int(valComments)
else:
valComments = 0
ValUsers.append(valKarma)
ValUsers.append(valComments)
ValUsers.append(valArticles)
except requests.exceptions.ConnectionError:
ValUsers = [0,0,0]
return ValUsers
def get_author_by_nick(x):
finalRow = []
df = pd.DataFrame
colnames=['nick', 'scores']
df = pd.read_csv(r'fileshabrauthors.csv', encoding="ANSI", names = colnames, header = None)
df1 = df.loc[x:]
fname = r'files/' + 'habrdata' + '.csv'
with open(fname, "a", newline="") as file:
writer = csv.writer(file)
for row in df1.itertuples(index=True, name='Pandas'):
valName = getattr(row, "nick")
valScore = getattr(row, "scores")
valAll = profile_check(valName)
finalRow.append(valName)
finalRow.append(valScore)
finalRow.append(valAll[0])
finalRow.append(valAll[1])
finalRow.append(valAll[2])
writer.writerow(finalRow)
print(valName)
finalRow = []
n = len_checker()
get_author_by_nick(n)There are a lot of checks there, because a lot of strange things happen on the habr pages, from deleted comments to some mysterious users. For example, how did the registration year 2001 appear in my sample? To collect user data, I parsed the mobile version of the site, and this version for some users not only reports that the user is deleted, but also gives this message: "Internal error (intermediate value).map is not a function". Everything remained in the comments, both deleted and unreadable, so I set them a registration date of 2001. Later, I discovered that some of these users are visible in the normal version of the site - if they were not deleted or blocked. But since there are only 250 of them, and half of them no longer exist, I decided not to touch them.
The final version of the habrdata table looks like this: ['nick', 'scores', 'karma', 'comments','articles','regdate']. It can be downloaded .
And here is how they are distributed by date of registration. I would say that in the long run there is some decline in registrations.
| Registration Year | 2006 | 2007 | 2008 | 2009 | 2010 | 2011 | 2012 | 2013 | 2014 | 2015 | 2016 | 2017 | 2018 | 2019 |
| Users | 2045 | 11668 | 12463 | 5028 | 5346 | 13686 | 11610 | 9614 | 9703 | 6594 | 8926 | 7825 | 5912 | 3673 |
In total, we have 114 users who have ever written comments or articles. Let's see what the karma and ratings graph looks like for users:
For these graphs, by the way, thanks to the chic visualizer .
We have absolutely insane outliers, you can see them on the chart as well. Let's say the user (UPD) for all his comments and publications, more than 268 thousand pluses have accumulated! And he is there in this stratosphere soars completely alone, the rest, more or less successful, hang out at an altitude of about 30 thousand. With karma the same story - the user karma is 1509, and everyday life starts around 500. I didn't cut the sample, I just zoomed in a bit so you can take a closer look at the distribution of ordinary users.
Here, at the request of workers, TOP-10 users were added by key indicators
A cursory analysis of the entire volume of users shows us that there are no clear correlations, either in pure form or with the reduction of outliers, so I will not dwell on this. It would be interesting to twist non-linear dependencies or see if we have any such clusters. Of course, I will not do all this - those who wish can download CSV and torment it even in R, even in SPSS. I'll go straight to what's bothering me - people who have positive rating scores but negative karma scores (and vice versa). We have 4235 users of these darlings. Here they are on the chart. 2866 users of them repeated my path, having pluses in ratings, but minuses in karma.
3-4 thousand out of 114 is a seemingly frivolous figure, within the margin of error. By the way, all users with negative karma are within the same margin of error. There are only 4652 of them. But let's look at the data not as aloof, as statisticians like to do, but as people.
All users: 114 343
Karma < 5: 89 447
Incl. zero karma: 67 890
Incl. negative karma: 4 652
Karma >= 5 and ability to vote: 24 896
Thus we see that the community is not really a "community" at all. This "“which can do nothing and therefore does nothing. A fifth of users have real opportunities to control the content of the portal, they are the community. So, when they shake the total population of Habr in a hundred thousand in front of you and say “One hundred thousand people are satisfied with everything, but you are not” - this is not entirely true.
And here is the same layout for ratings:
All users: 114 343
Score <5: 57 223
Incl. zero score: 26 207
Incl. negative rating: 9 737
Score >=5 and hypothetical ability to vote due to score: 57 120
And here we see that if the right to vote was determined by ratings, and not by karma, then more than half of the users could vote. And this is only in the opinion of those who can rate, i.e. owners of karma! In the case of free voting, of course, 90 percent could vote.
There is a fairly common but erroneous belief that "you just need to write an article" to get into this select community. This is not true - there are only 5 thousand authors of articles with karma >= 24 (another 900 users received karma more than 5 without articles for some special merits; apparently these are echoes of the previous rules and the karma they have preserved from those ancient times). Despite the fact that at least one article was written by more than 36 thousand users, a third of the authors of the articles did not receive the right to life.
Maybe the aforementioned third of the authors did not perform well, maybe their articles were bad and the community did not like them? No, all the same statistics tell us that 90% of those who wrote at least one article, and did not achieve more than 4 karma, also have a positive overall rating. But the assessment does not mean anything, because they have “low karma”. So you can have positive ratings, have articles, but at the same time not have high karma and the ability to “regulate the community”. It is neither yours nor ours. .
The ratio is also maintained within periods, for example, if you take only users from the registration date later than 2016 or 2018, when there were “project mergers”. 90% of users with at least one article have a total positive rating, but a third of them have karma less than 5 and cannot vote for articles. That is, “write articles to raise karma” works in about 60-70% of cases.
Here is another simple ratio that will tell you everything about what is happening:
78205 users from 114 343 have a total score greater than 0. This is how their articles and comments are evaluated, that is, actions useful for filling the portal.
24 896 users from 114 343 have the opportunity to vote. This is how their personality is assessed, that is, those who can already vote like their personality, or they don't like it.
At the same time, look at the graph of dependence of karma on the year of registration. Many say, they say, we have hazing - yes, this is what it is. In its purest form, as in the blockchain. These guys were the first to start, over the years they have mined karmas for themselves, and now it is from them that you constantly hear “yes, I don’t pay attention to karma at all and I don’t advise you.”
Не отдам своего сына в программирование, пока там не решится проблема с хабровщиной!At the same time, sixty thousand people can in principle write interesting or useful things, get positive marks, but at the same time they have to constantly look around so that they are not whipped for an unenthusiastic way of thinking.
Total:
- In principle, the commentator is not part of the community, even if he develops and supports it.
- The author of articles with a probability of 1/3 is also not part of the community, even if he develops and supports it.
- Even if the actions to develop and support the community were clearly approved by the pluses, the author can still be blocked by an extremely small part of users (literally 10-20 people out of thousands)
Who are these villains who put minuses on people developing the community?
When I was preparing this article for publication, a new topic appeared on a similar topic. In the comments, as expected, conversations about karma began, and another obvious conclusion:
You can nod as much as you like at the commentators who worsened the resource, but ... But they just can’t do anything on the hub:
They don't write bad articles.
- they are not voting for crooked reprints of hautushki with a lack of understanding of what kind of teams and why they are driven in there
- they are not voting in karma for hype news writers
- they do not evaluate the correctness of someone else's opinion
They cannot support the authors and express their respect in any way, except for comments.
And they can not protect themselves from others in any way.
Everything that happens on Habré is the work of those who have an article and karma.
Well, who is to blame - figured it out, let's see why all this is happening.
The part where he kills you
If every person can participate directly in management, what do we manage?
German Gref
As you understand from the comments quoted above, the essential issue of karma has not changed for many years. This problem is not technical, but psychological (maybe that's why they still can't solve it on the technical resource).
Let's look at its key components and analyze them in more detail.
- Karma does not depend on the actual quality of actions on the site
- Karma is psychologically asymmetrical
- Karma condones sociopathy
Item 1.
This is the same problem. with which I began my article: a plus person may end up with merged karma. If we discard various trifles, such as formulas for calculating ratings, we will see the key difference between Habr and all other sites: the separation of the user and user actions into two independent entities.
The most common and intuitive scheme looks like this: a user is an account, posts, comments are written from this account, “cute” pictures or photos are posted. The user is his actions. Other accounts like or dislike these posts and photos. The sum of likes and dislikes determines the quality of both messages and the account itself. They are inextricably linked.
Everything else is unimportant. In some cases, the minus ones are blocked, in some they are not. On some portals, in order to rate, you need to already have a high rating - on others, no. Sometimes the authors of the rating are displayed, sometimes they are hidden. But nowhere is it possible for a person to publish several downvoted posts, comments, photos - and at the same time maintain a high rating; as well as vice versa - if the user's messages are upvoted by readers, then the user cannot be banned by them, because they like what he does. And this happens due to the fact that the user's actions on the site and his account are one and the same. Your actions are plus - it means you are plus. Your actions are minus - so minus you.
On the hub, the situation is fundamentally different. On Habré, the essence of the user and his actions are artificially divided. All your actions can be approved and plused. However, your account will be downvoted. And vice versa. If on other resources they throw pluses and minuses to articles and comments, then on Habré they throw pluses and minuses separately to articles and comments and separately to the author.
This is what karmic curse is based on. And then it spreads and begins to bring damage to the entire community.
Item 2.
A separate evaluation system inevitably falls under the influence of two psychological distortions.
The first distortion is the psychological readiness of people to look for the negative and produce the negative. Aggression is the main reaction to everything unfamiliar, incomprehensible or unpleasant. As a result, a person's willingness to put a minus is always higher than the willingness to put a plus. You can see this in a lot of situations, in marketing it's a classic feedback problem. If a business does not want to write fake positive reviews, it is forced to implement a bunch of complex methods for obtaining them: give discounts and gifts, beg and remind people, give us a plus, write positive reviews. I saw a lot of links to an article already in 2013 that on Habré more often they plus karma than minus. Perhaps this is so now; but from the same article, we know that karma is plused by those who wrote the article, and commentators minus it.
This is a very serious distortion - a dissatisfied, aggressive person is always ready to spend time and energy to express dissatisfaction, to dump his aggression. Even with pluses and minuses for ratings, we have a constant “war of minuses”, when an embittered interlocutor puts minuses on each of your comments in the current topic, and even runs into your profile to find old comments and downvote them. But comments are at least easier to upvote - if a person agrees, he moves the mouse by a centimeter and upvotes. It is already more difficult with karma, most often they get to karma on the fuel of aggression in order to put one more minus.
Karma only works as a rating below the articles because there are big up and down arrows that the reader can easily click on. To change the commentator's karma, you need to do several additional actions, that is, the question is how quickly the response impulse to the text fades. The negative impulse fades more slowly for psychological and biological reasons - therefore, those who want to put a plus get to karma less often, preferring to put only plus marks on comments.
By the way, most of those who advocate karma do not even think about such complex things. For example, in all seriousness, without any emoticons there, they ask the fighters with anonymous minuses - why are you unhappy with only minuses? Why, when you are anonymously given a plus, are you satisfied, but do you want justification for the minuses? And that's why. Because a person's readiness to put a minus is higher than the readiness to put a plus, the readiness for aggression is higher than the readiness for approval. This readiness must be flagged and limited, if only simply so that the pros and cons become equal - they have long been forgotten on Habré that they would be deserved.
The second distortion is the appearance of a caste of judges. I remind you that usually a fair system is "all users judge all users", everyone just rates the articles and comments of others. But the administration of the habr was very worried about the authors of the articles, who could be good in technical matters, but terrible in social interaction with commentators. And the authors were given carte blanche, from now on they could only be judged by other authors.
Indeed, we can meet such systems at various literary competitions, for example: each wrote his own story, each read the stories of the others and gave each one an assessment. This is also a fair system.
Only on Habré the system was again distorted - other authors could only be judged by часть authors. Not everyone who wrote an article has the opportunity to vote for karma. And most importantly, a huge number of users (commentators) have appeared who themselves cannot judge anyone, but they can be judged and executed, and without the right to justification. As a result, out of a huge number of users, a small part of the “arbitrators” stood out - a fifth of the total number of users - and began to do whatever they wanted with the rest.
There is an implicit assumption that a large number of users will flatten the scores. This is wrong. Since judges can change karma in the same way as they themselves, sooner or later, objectionable ones fly out of this caste, and obsequious ones, on the contrary, fall into it.
In fact, all the positive examples that we see are a survivor's mistake. They just got lucky to cut through the toxicity of the community.
The situation is aggravated by the fact that people often work in the technical field who are rather poorly oriented in the social field - this is all computer hickey autism, leading to "inability to maintain and initiate social interaction and social connections". Here is toxicity for you, here is aggression for you, here is your desire to remove everything unpleasant and unusual out of sight.
All this together leads to the next point.
Item 3.
If the minuses did not affect anything, there would be no problem. They often write in the comments - but I don’t even look at karma, ha ha, but why is it needed, you are all just karma scammers, etc. Usually these are people with very high karma. In fact, karma would not really be needed if its low value did not block the possibility of communication on the portal.
And that's why it's blocked - because Habr's karmic system is based on the idea of what is objectively bad и good people. See above - not "bad or good user articles", namely bad и good user. They will give me an example of trolls, "bad people"; yes, it is fair - but practice shows that even a specialist does not always distinguish a troll (or bot) from an ordinary person with a strange opinion.
Other portals introduce an ignore mechanism to combat this. If you once decided that a certain person bad - you don’t try to add a minus to his karma, because he bad, but you simply ignore it, and you no longer see any of his articles or comments. But the habr administrators are far from human psychology, so they decided that bad и good these are not evaluative categories, but an objective truth, as a result bad they are simply thrown out of the site into the Gulag without the right to correspond and are shot as enemies of the people.
Here is a user comment , an employee of Habr
As you can see, the employee firmly and sincerely believes that karma really reflects the usefulness of the person’s publications and comments. Where do people come from who have a total score of +100, and karma -10? And why are there so many people with such a deviation? Maybe thousands of users publish flood and unfounded statements, get minuses in karma for this, but then some magician arrives and puts pluses for the same flood and unfounded statements with the help of ordinary ratings? Of course not.
Just the usefulness of comments and articles show ratings next to comments and articles. And karma reflects bad or good this is the person according to the voting caste. Above, we discussed why people will make more efforts to harm bad person to help good. Therefore, the execution poor human in such a system is statistically inevitable. Sooner or later they will kill all the usual "bad" ones, then they will start looking for the "least good" ones, and so on and so forth.
Please note that all these difficulties are based on the complete inability of the administration to understand and calculate the actions of people. Focusing on the technical side of things, they completely overlooked the social side. Approximately the same people designed the Universe-25, and then for many years they tried to tell everyone that there was paradise there. Some still believe in it - as they believe that "karma makes Habr better." The creepiest thing here, of course, is that the administrators and many participants do not even understand what is wrong here. Yes, they say, people really are good and bad. So let's all the good ones get together and kill all the bad ones! And they love to kill.
“How did the reptilians on Habré:
a large number of users were given small things and motivated: “Guys, which of the passers-by you didn’t like, shoot at that one. Do not be shy, nothing will happen to you for this, and no one will know who the shooter was. A lot of hits - great, you cripple him and he won't be able to talk much. Make the world a better place and don't deny yourself anything."
What happens on Habré is a sociopath's paradise. As andorro said on another occasion: "Social networks are created by asocial people."
Interestingly, there is no ignore on Habré. If you are dissatisfied with a person, you can only leave or put a minus in karma (that is, force him to leave in this way). Kill or die system. Roughly the same scheme was implemented by Durov in his Telegram - there is also no ignore, and the only way to avoid an unpleasant person is to leave the group chat or force him to leave the group chat. The consumer approach of a “successful person” walking over the heads of others is very clearly visible. Compared to, say, IRC created by people for people, Habr or Telegram are created by "highly active sociopaths" for the "target audience". If you are not in the target audience - come on, goodbye.
Episode three
— What can save us from revision?
- Sorry, not us, but you
"Operation Y"
What can be done?
First, we should finally accept the idea that Habr is no longer a closed community with invites, but an ordinary portal, and it should have a simple rating system common to such portals. For the presence of articles, since they are so important, you can give a double rating. But the system should be unified - for comments and articles they put pluses and minuses, if you get pluses more often, then you are good, if you get minuses more often, then you are bad. To combat overly vicious commentators, it has been repeatedly suggested that only those who also have their own articles can vote for articles, this is quite enough.
Secondly, the lock boundary is just ridiculous. What does 10 or 20 minuses mean for a portal where thousands of people can vote? We see that the average score is 118, well, without outliers it will be somewhere around 100, so -100 should be made a real border, after which comments begin every five minutes and other horrors, and then a step of a hundred, not 10 .
Thirdly, the rating that is used now shows more activity (ie, time dependence). It would be more useful to show the rating “pluses per message on average” - then people will not once again flood with meaningless comments, and the top users will look more correct: whoever has more useful messages will be in the top.
Fourth, instead of artificially fueling toxicity and mutual hatred in the community, incl. almost officially endorsing the "war of cons" - you just need to finally add ignore. And not just to fold comments under the spoiler, but to hide them, something like "UFO hid this entry at your request." And in order to cancel the ignore, you need to go into the settings and manually enter the nickname of the one whom you ignored; that is, turning on ignore should be easy, and turning it off should be difficult.
Fifth, I think it's time to revisit the issue of "price per appraisal." To put a minus, a person should spend part of his rating. The reasons for this were discussed above - it is more likely that an unhappy person will put a minus, than a satisfied one - a plus. It is necessary to equalize the chances of pros and cons.
And lastly, you can leave karma in its current form just as an element of decor and tradition, but remove its connection with blocking. Then, finally, all these jokers and jokers with their jokes “why are you worried about karma, I’m not worried, it doesn’t affect anything” will finally be able to say this seriously.
Only registered users can participate in the survey. , you are welcome.
Are you satisfied with the karma system in its current form?
-
Yes
-
No
1710 users voted. 417 users abstained.
Source: habr.com