I'm leaving my two cents anyway. Frankly, it's absolutely ridiculous that a serious site is still using the term "shoujo-ai" at all, and refusing to adapt to how the actual community uses words. It's a distinction that simply does not exist in the heads of any creator or scanlator, and the result of that is plain to see - you have manga being tagged with them apparently at random. A romance about a girl sexually harassing another girl and falling for her (Hino-san no Baka) or a comedy about a girl trying to buy another girl's body (Shouraiteki ni Shindekure)? Pure and innocent shoujo-ai, of course. Whole piles of manga with "yuri" literally in the title, often published in Yuri Hime? Those aren't yuri, silly, why would you think that? Cute manga about a guitarist girl and her underclassman having confusing feelings about each other (Whispering You a Love Song)? That's, uh... both at once? Somehow? Yagate Kimi ni Naru's anthology is shoujo-ai rather than yuri? Why? Girl Friends is both, but the sequel chapter is just shoujo-ai, and similarly for Kisses, Sighs and Cherry Blossom Pink stopping being yuri when it turns from an anthology into a full-fledged series about girls kissing and going out with each other. And that's before getting onto basically any yuri doujin, which end up as a complete crapshoot which term will be used - or sometimes, it's both! It's one thing to say "well, we don't necessarily use every term in the same way as the original Japanese", but when it reaches this point with a term this frequently used by both creators and fans alike, it simply becomes, as the previous examples and many more demonstrate, pointlessly confusing and ridiculous.
It's been suggested earlier in this thread that the solution to this is simply to report the "incorrect" tags. This suggestion misses the crux of the issue entirely. The problem is that nobody knows what the terms mean, or can agree on where the distinction is drawn, not even the people scanning and uploading them. If the actual creators and translators can't agree on the precise nature of the distinction, how on earth is a Mangadex user supposed to meaningfully use them to find the content they want? And that's before we get into series that have sexual content at the end, which would... suddenly change the category? Or maybe make it both? I don't even know. Like, most romance manga don't start out with sex, so is it "shoujo-ai" for the first few chapters, then it suddenly turns out to have always been yuri at the end? None of this makes sense, or is any actual help for an end-user who just wants to find and read manga - and all those "incorrect" tags from uploaders are doing is helping them find it. This doesn't strike me as evidence of a helpful, understandable, useful tag.
Now, in this thread, there's been a lot of fuss over the terrible issue this would cause for all those people who really want to see sweet pure girls love but want to avoid so much as seeing any icky, nasty sexual relationships between them. Now, because of the "nobody actually knows how to use these tags" issue outlined above, they already can't do so, but let's ignore that. Maybe suddenly a grand revelation happens, everyone now fully understands and agrees with each other on the distinction, Comic Shoujo-Ai Hime opens in Japan, etcetera. Personally, I'd say the bigger issue is that those people just don't exist. Most people who want to read yuri just, you know, want to read yuri, and they aren't interested in whether or not the fact a fade-to-black sex scene happens twenty chapters in or something. No yuri creator makes their creative decisions based on criteria like that, no yuri scanlator chooses what to translate based on criteria like that, and I do not believe that any yuri reader picks what to read based on criteria like that.
Teasday has said that "I've tried to stop assuming what kind of filters people want and need.". Okay, fair enough - maybe I'm wrong and, despite the genre itself making zero distinctions of that nature, there are thousands of Western readers who really want to. The issue is now what tradeoffs are being made in order to satisfy the need for that distinction - a need which, I maintain, is simply clutching at pearls and has never been demonstrated to actually exist. If there were no other issues created, after all, while it might still be mind-boggling that a site becoming increasingly central to the yuri fandom is still insisting on the ridiculous term "shoujo-ai", then fine, it doesn't actually matter. But it does. If I want to see what yuri works have been released, I can't simply select the genre, I have to go through the search interface and create an advanced search for two genres at once, which is just silly. One of the arguments in this thread has been that "outside of advanced search there would be no mechanism to filter out only "smutty" Yuri while leaving "tame" Yuri alone". Yet right now, there is no way outside of advanced search to just show me all the yuri, full stop, without risking missing out on series I and pretty much any yuri fan would clearly be interested in. Which one of those searches is the more fundamental, basic, and least "advanced"? Both options "assume what filters people want and need". The only question is which assumption makes more sense and is more useful.
Even more pressingly, it makes it impossible to search for yuri AND something else. If I want to read yuri fantasy or yuri comedy or whatever, the only thing I can do is create two separate searches, one for "yuri + fantasy" and one for "shoujo-ai + fantasy", and go through each set of results individually - sets of results which, again, because of there being zero consensus about what, if any, distinction there is between them, will have plenty of duplicates. That's an actual problem, not some theoretical "but what if someone want stuff that's gay but not too gay" distinction. The term "authoritative solipsism" has been used in this thread, and I have to agree, because that's exactly what I'm seeing here. If the people making this decision were the ones actually using the tags, I find it hard to believe that they would not already want these problems addressed rather than simply dismiss it with the fact that they "could see people" using the terms as they are now for something, probably, so eh, it's fine. The site needs to take feedback from the people actually using the terms into account. It seems to be that the tail is wagging the dog here, with the functionality of a tag or set of tags being determined not by the people who want to find that type of content, but by people who (theoretically, if they actually exist) want to avoid very specific types of content in a single very specific way.
