Forum
>
Site Discussion
>
Thread
Log in |
Register
Such a thing already exists. It's called 'not clicking on thumbnails of pix which don't interest you'.
Precisely. It's like people who complain about programmes on TV. Why would you watch them!
No but i do think we need a dislike button for videos that are just slide shows. or water melon crush videos that have dubstep as background music. and video with filters that make it look like vhs .despite being a modern video .or when they have seasure enducing effects.
If a video is just a slideshow, with static images or rapid fire scenes, then please do report them. For melon crush videos, if the muscle isn’t easily visible, then you can let us know too.
Such a thing already exists. It's called 'not clicking on thumbnails of pix which don't interest you'.
This is pretty much what I said when people were upset about Eunhee being on the site. If you don't like her, don't click it. Yet here we are.
This is why a block feature would be a good idea. Let people decide for themselves who should and shouldn't be in their feed instead of one person deciding for everyone.
It's just another filter that's already being used in different ways across the site. A filter that would only be run if a user is logged in and has blocks.
Its more that the building a database query to exclude certain entries can become very computationally expensive. It is not that it can't be done, just that implementing it is not straight forward. Also, consider developing and improving GWM is still very much a hobby project, so Chainer's day job and life will take priority. We may see this in the future, but I doubt there is any urgency in implementing this?
BTW there are probably less elegant ways to achieve this, like someone writing a piece of Javavscript in the front-end just to hide entries, but they you'll simply find yourself with empty blocks on the page. Something like Tampermonkey, but you'll need to code it yourself.
Its more that the building a database query to exclude certain entries can become very computationally expensive. It is not that it can't be done, just that implementing it is not straight forward. Also, consider developing and improving GWM is still very much a hobby project, so Chainer's day job and life will take priority. We may see this in the future, but I doubt there is any urgency in implementing this?
BTW there are probably less elegant ways to achieve this, like someone writing a piece of Javavscript in the front-end just to hide entries, but they you'll simply find yourself with empty blocks on the page. Something like Tampermonkey, but you'll need to code it yourself.
What's the site written in? I suspect it's built with Django, but I could be entirely wrong on that. If it is Django, I wonder if the following could work, roughly:
if request.user.is_authenticated:
blocked_list = request.user.blocked_list # There are various ways of accomplishing this part
DatabaseObjects.exclude(name__in=blocked_list) # then pass to context and you have your filter
Of course, it's hard to say exactly without knowing the backend. It's fun to theorize though. Conceptually, something along those lines could work (again assuming Django's ORM is at play here). A new field would have to be added to the User model to store a list of blocked names. Then the filter would operate off of that list. An additional conditional can be input to ensure a user has blocks first.
All this stuff about blocking things reminds me of the standard Timorous Tim Debacle.
I don’t think it’s a bad idea, it would be a sort of personal filter that would not affect other users view experiences. Not even disrespectful, just fine tuning each member personal preferences.
I also think this feature would be super nice
A crowdsourcing caveat:
The problem is that the names of the models are crowdsourced. Misnamed and unnamed images would slip past the block. This applies even more to any other detail, even if it is included in the site. This obviously does not cover videos with bad music or videos with bad quality filters. Or images with tattoos.
Imagine excluding "teens" and you would basically block most models under thirty. Women become facially "adults" around the age of 28, which sometimes seems to be the criterion for "teen" here.
About tattoos:
Filtering according to tattoos would face a few dilemmas. One is that the model might have tattoos hidden somewhere that are visible in no picture or just a few pictures. Another one is the token tramp stamp versus MMA level or prisoner level tattoo "artistry" or chicks who have the bad philosophies and ambiguous symbolisms of entire movements on their skin. To me, I still think it means that the individual in question has no personality or identity and imagines that the ink can give her one.
So we would have "model has lots of tattoos", "this image has lots of tattoos", "this image has a small tattoo", "the model is known to have a small tattoo somewhere" and "the image and the model are free of tattoos". Maybe half of the images on the site are free of tattoos, so don't worry, the prudes are doing just fine.
Technical limitations:
I don't think SQL has a feature to ask for "enough rows that fulfill these criteria". It only has "certain number of rows, then filter according to criteria". So the site, when showing one page of pictures, asks for that number of items from the database. It definitely does not ask for all, then filter and then pick a hundred.
Though if only one criterion was used at a time and there was a limited number of them, it would be possible to add indexes to the database for all of these. But combining any two criteria would run into the previous problem. And each index would take a few megabytes to store and an unknown amount of processing power to update every time there's one more or one less image that meets the criterion.
This because you would store "list of pointers to images that this criterion does not exclude" instead of those that it does, because that would give you no computational benefit at all. I wonder how a search according to multiple tags is implemented.
Can we block models we don’t want to see?