Posted under Tags
Here's a list of the most common sources: https://docs.google.com/spreadsheets/d/15F5zLSDyMaFsS3cS20dAwbX_AFs1Q8UwsqINzpZKcvo/edit#gid=641174512. Hentai foundry is one. fc2.com is a top site, but I don't really know how it works, do things there have any kind of ID?
As far as I'm aware, FC2 is really just a blog site, with a page or more for pictures with direct hyperlinks, so it's not really indexed... and IMO not a good candidate.
However, given the plethora of results from the spreadsheet above, I'm leaning more towards an incremental approach, i.e. add an implication once the tag gets populated above a certain threshold...say 50... Additionally, the tag should have a wiki that tells what the bad_<SITE>_id even means, and how to determine it for that particular site, i.e. are there any caveats such as registering or following to determine a bad ID.
I just don't want us to get stalled thinking of every single site trying to make the implication request perfect instead of just processing things as we go.... Thoughts?
I'm okay with that. It seems better that way.
For now we just have all the major art sites that users are sourcing images from... well, except maybe weibo. I've only made that mention because I was thinking of what sweetpea said in the last discussion, but source:http://weibo.com/ doesn't pull up very much. Only like 14 images or so, so if you think we won't need it then you can remove it. I'm not sure if any of our weibo sourced images even have a bad id yet.
Interesting note though, I don't see bcy.net in that list either. Hasn't it been getting picked up more? Although I can't say much since I don't have an account there, apparently you need a Chinese SNS to register.
Got it... updated the implication post and request.
The bulk update request #1034 has been approved.
sweetpeɐ said:
Could you quote me on what you are referencing please?
It wasn't exactly about this particular subject, but I just remember a mention to Weibo in forum #124901, which is why I included it in my list. But we barely source from Weibo, and especially given it isn't primarily an artwork sharing site.
And it's usually not a great idea anyway since social networking sites usually recompress the hell out of the media they're given.
Mikaeri said:
It wasn't exactly about this particular subject, but I just remember a mention to Weibo in forum #124901, which is why I included it in my list. But we barely source from Weibo, and especially given it isn't primarily an artwork sharing site.
And it's usually not a great idea anyway since social networking sites usually recompress the hell out of the media they're given.
There are some gems on Weibo, they are just difficult to get to because of the language barrier and convoluted log in system. That being said, I have in fact found a good upload from their (post #2459751), which would not have been found from another source by this artist.
Just because it's social media doesn't mean it should not be considered as a source. For instance we source from Twitter frequently and to a lesser degree Facebook. It's always up to our best judgement; usually another source it better but not always.
So I would say go for weibo bad ids/samples. It would help a lot just to know when this source is improperly sourced or sampled because then we'd know how to correctly source from it. I believe my upload may have been sampled since it has a watermark but I can't be sure.
I use a userscript called "Weibo Big Picture (新浪微博查看大图)." I'm not sure if it actually finds the original but it at least appears to find a larger image.
Thank you for the link to my earlier comment.
sweetpeɐ said:
There are some gems on Weibo, they are just difficult to get to because of the language barrier and convoluted log in system. That being said, I have in fact found a good upload from their (post #2459751), which would not have been found from another source by this artist.
Just because it's social media doesn't mean it should not be considered as a source. For instance we source from Twitter frequently and to a lesser degree Facebook. It's always up to our best judgement; usually another source it better but not always.
So I would say go for weibo bad ids/samples. It would help a lot just to know when this source is improperly sourced or sampled because then we'd know how to correctly source from it. I believe my upload may have been sampled since it has a watermark but I can't be sure.
I use a userscript called "Weibo Big Picture (新浪微博查看大图)." I'm not sure if it actually finds the original but it at least appears to find a larger image.
Thank you for the link to my earlier comment.
I agree, it's just that I think weibo (and other SNS sites aside from Twitter) shouldn't be main sources. Of course there's images that can be sourced from there, but it's not usually where you'd find art unless you know an artist exclusively provides work there.
But I think if there isn't enough of a tag count (50, as brokeneagle said, is a good number) then an implication for such a rarely used tag would provide a little bit too much overhead. It would be simple to tag them all with a script anyway.
In regards to Chinese sites though, the rarity of images from those sources is only exacerbated by the fact, as you mentioned, that they have a high barrier of entry if you don't understand the language or live there (as bcy requires a Chinese SMS).
create implication bad_deviantart_id -> bad_id
The post count has already reached > 1K, so submitting this request.
The bulk update request #1046 has been approved.
create implication bad_artstation_id -> bad_id
The current count is ~93, which exceeds the minimum 50 threshold.
Updated
create implication yandere_sample -> image_sample
create implication bad_yandere_id -> bad_id
Both of these have exceeded the 50 count limit...
@Type-kun or @Hillside_Moose could you look at the yandere sample/id request?
The bulk update request #1074 has been approved.
The bulk update request #1103 has been approved.
