Saw 8+ repeated rants today ☹️ #feelsBadMan

definitely a "mark as duplicate" option would be great

Not that I don't find duplicates annoying, but there were 308 rants posted on June 1 which translates to one of our highest days yet. We're definitely working to try to improve it and have a lot of ideas, but if it really was around 8 out of 308 I don't think that's terrible, though we can definitely improve.

Also, we definitely like hearing ideas on this. I'm not sure a mark as duplicate button would do anything. Just because something might be a duplicate, doesn't mean we want to take it down. That goes more for text rants as opposed to images. Text can be a duplicate but should still be fine because multiple people can have the same thing happen to them.

Basically I think it's a complex problem that most likely requires a decently complex solution to make it work well.

You can hash the text and username together and store that value in a table as a primary key? That would take care of exact duplicates by one person. All you have to do at that point is trap the SQL error and pass a more friendly error to the user. Just a thought...

@dfox this is a wonderful 'growing pain' challenge - not a real problem.

A real problem would have been if devRant had received a poor or mediocre acceptance by the dev community. Well, as we know that ship has sailed and devRant is rapidly becoming 'the community' for developers globally. In fact, I get the feeling there are more International members than US based. I am learning a lot from my brothers and sisters overseas. Set me straight if I am wrong.

Managing the devRant stream is a programming challenge and UX opportunity. It's what you and we all do.

After reading your devRant News Update from last week and enjoying devRant the past 2 months I would put up money that within 3 months we will be able to filter and follow data unlike any other community.

Just keep embracing these challenges. We will all help as best we can. It's what we do.

@jagalbraith good idea, but I don't think that really addresses the main issue(s).

Sometimes people post twice accidentally (I don't think this is what @localhost was talking about) where a solution like yours would work, is they are usually identical posts.

The main issue though, from trying to gather what people think, are repeated images from different users. Repeated image, unless absolutely identical, are pretty hard to detect. There are libraries and algos that will generate a hash of an image for you, and then allow you to determine the similarity to another image. There are a number of problems with that, ex. it has to know that memes look similar but have diff text, and also it's only used for calculating between two images, and not 1 image and every other ever posted like we'd need.

Then there's duplicate text posts, ex the binary joke. It's not usually posted identical to other text, but close. Then the question becomes what we actually want to do with these dupes.

@dfox You are right in the fact that making the slightest change in the text or image makes it programmatically unique, thus, a giant game of whack-a-mole in trying to de-dup.

We have a plan to implement a simple repost marking mechanism tied in to -1. Should be available soon.