Could it be possible to do this (automatic updates):
(edited)
Delete ALL accounts with 0 rating or 1 quick rating (beer AND places), older than 1 year old and without any logging session for more than a year and with a url/website or phone number entered in the user description. (EXCEPT if AFFILIATED).
plus
1.force email verification for new accounts
2.make username, country and region required fields for new accounts
3.don’t allow urls (block http, www, .com, etc… ) or phone numbers (block more than 5 consecutive numbers in a row) in user description.
I know this might fell like it exposes the website real numbers, but it would be way better this way. @services
To make the website look less like of an amateurish farce?
A website plagued with fake and empty accounts doesn’t make it look better.
Sure it might feel good to tell they have 705083 members…
but they better display they have 9528137 ratings (almost 10 millions ratings) instead (most of them probably come from users with more than 1 rating anyway…I hope… so removing useless accounts shouldn’t hurt too much this number)
This also make a statement that the website owners are working to have a legit account database, especially since they are (were) a product review company first.
The “1&2 PLUS” suggestons you sugest are good step for quality control of the users and should be implemented. No.3 is not needed if 1&2 are mandatory. Im not so sure if I agree in the delete statement if they have performed at least one rating. For these cases the Admins should look into it and delete if its like an ovious scam like for instance giving 5 for a beer or a place that is so ovious from numbers of other trusted raters that this is just fake. We have seen more than once that place owners login and give 5 to a shitty place that should not even be listed on RB.
The figures you display tells me an average of less than 14 rating pr member.
Concidered that I all the raters I know have more than 500 ratings and I dont know any on RB with that low amount of ratings, and I can also see that by removing all these "one rating ratesr) the averate would be higher, however 1 rating only do not say its not an concidered rating. I feel sad if RB wants to remove users with few ratings just to get better statistic counts. Why did they leave after 1-2 ratings thats the question to ask, not just wipe them away to get better stats.
I would preffer to se the number of users that have done one or more rating (beer or place) last 12 monts. (or even better the numbers for each year).
That figure tell somthing but those figures they do not dear to come up with. Bet they are pretty low now. I will not be supprised if they are far below 10000 maybe even below 5000.
That figure tell something about the site. People are running away because of to much writhing to do an beer rating and the will from our excelent services to do fix bugs. (or was it maybe not will to).
If services do show some will fo fix whats broken and also make it easier to rate then maybe users come back too. The quality of content of beer and place rating in RB are ecelent and trusty for planning and get ideas where to go and what to taste, let it be the best.
Simple sample: they can fix this I have been complainging about this for almost a year now.
That should have been an easy simple fix for everyone that are capable to write an select count() statement and at the same time have learned to count.
the only thing we know right now is that we have 7583 users with 100 or more beer ratings out of 705000 users (around 1% of total users)
and out of those 9528137 total ratings, 6409576 ratings are from the top 1000 raters only. (around 67% of total ratings coming from 0,1% total raters)
let’s say those 6583 other users - with 2400 ratings to 100 ratings each - have an average rating count of 200 (very arbitrary), this make another 1316600 ratings of 9528137 (so around 14% of total ratings coming from 0,9% of total raters).
This makes around 1800000 ratings left for the remaining 697000 users (around 19% of total ratings coming from 99% of total users, so average 2,6 ratings/user for those remaining 697000 users)
I wouldn’t delete 0 rate accounts as some people just don’t rate & use the website as resource. Delete accounts that haven’t logged in during the past 5+ years maybe.
I can see the point in deleting Accounts that has ben not loged into for 6+ monts with no rating neither for beer or places. I would not delete if they have loged in last 12 months.
Keep in mind that even if people do not rate beer they in many cases find ratebeer as a great sourse of inspiration to find great beer to drink and places to visit so you should definitely not delete an account that is in use even if there is no ratings.
And delete an account that has not been in use for 5 year is definitely not an thing to do, that account might hold alot of great beer/place ratings that should be preserved. I know great people that has not been logge in for many years but they have more rating attached to their accounts and all of them are on the 100 beer list many with more than 1000 ratings, so they have been active and the history should stay as the great library it is in RB.
I don’t think it’s bad that we have lots of users with few or no ratings. It’s normal for these sort of things to follow power laws (more or less) - you would expect the top few users (in any country) would have double or more ratings than the next few, and the those next few to have multiples of the next twenty - and then for it to fall off very quickly.
So I’m not sure that having lots of inactive users is a bad look for RateBeer because it is typical of this sort of activity - the vast bulk of all activity is undertaken by a small subset and within that subset it is further concentrated.
What is a bad look is fake users - added to pad numbers or to shill for beer-related products or carpet cleaning companies. It makes the site look like it’s abandoned and surely affects how Google evaluates the site. I’d like to see them gone but stopping them joining and then seeing if older ones can be selectively pruned would IMHO be better than ditching our more reticent contributors.
That’s probably true. I get no problem either with empty accounts. This at least tells that the user who created an account had an enough interest to try the Site or the App.
The real problem comes from the fake accounts and the fake reviewers.
This is why they need to reinforce account verification. They should at least try to find a way to delete evident existing fake account too.
for example:
Delete ALL accounts with 0 rating or 1 quick rating (beer AND places), older than 1 year old and without any logging session for more than a year and with a url/website or phone number entered in the user description. (EXCEPT if AFFILIATED).
We admins barely keep up doing admin work (verify, editing, adding stuff, removing fake ratings…), if we can get spared the task of spotting spammers accounts apart from the ones with fake ratings (and right now, the company doesn’t have enough Devs to do that too), that would be that.
This would of course need a good account verification to begin with…which is actually totally deficient.
I really do not like this idea, it’s been brought up before, and we’ve never ever deleted beer ratings except those that were obviously spam.
Major points to consider:
Users with 0 ratings. What harm does the account do by being present? As others noted, sometimes people use the site for info without rating. Some are affiliated with breweries and cannot rate, but still contribute. Granted some have been gone for awhile, but deleting the account discourages them from coming back and again, what harm does it do to leave it?
Users with few ratings. First, how do we define few ratings? I don’t like the potential for a slippery slope here. Second, how do we know that these few ratings are bogus? Granted some are, but for those that are not, we could be losing valuable data. For example, maybe one of these users rated an obscure out of the way beer. I would hate to lose that information just because it was the only thing they ever did on the site. Now, for those ratings that are bogus, new questions arise: what is bogus?
2a: amateur ratings by people who don’t understand craft beer styles. These should NEVER be deleted, we do not rate to style, we use the hedonic scheme. If a person likes the beer, they like it and that is truth, and should stay. RateBeer is NOT a site for only for experts and trained judges. It NEVER has been, that was never the motivation behind it. It is about regular people drinking beer in regular circumstances.
2b: spam ratings. When it is possible to PROVE a rating as spam, them delete it. If a user’s entire account can be proven spam (regardless of how many ratings), then delete them and block their IP.
Following that, there is never a good reason to delete old accounts, inactive accounts, or few-rating accounts, they all may have value in one form or another and cause no harm by their presence.
I’m not sure what “reputation” we are trying to preserve here. RateBeer is over 20 years old, quite likely could make a case for the longest continually active beer rating website in the world (I don’t know on what date BA came online, but aren’t they gone? and I haven’t checked on the Oxford Bottled Beer database in forever, I doubt its still around). Whatever challenges that can be leveled at RateBeer’s data have been done so again and again for 2 decades and we’ve withstood it. It’s done. We won. (well, until we lost to social media, but we’re hanging on, and hey, if you want to talk about a database full of amateurish rates, look no further than untappd, crikey what a mess)
