Graclic Bred

cute caracal! why is it in your house.

what a handsome alligator! why is it in a kiddie pool in a living room.

what a beautiful cobra! why is it being kept in a walmart $5 critter keeper.

such a beautiful fox! why is it in a two foot long metal crate.

I'm pretty annoyed at this post making the rounds, so I've decided to correct a bit of misinformation and misconception about how Tumblr as a platform perpetuates transphobia. Because there's a lot of things to criticize about Tumblr's rainbowashing, but I consistently see users criticizing the wrong things, often due to a misunderstanding of how social media platforms operate.

First off, lets talk about the idea of Tumblr staff "removing the trans colors from the flag".

The argument that staff made a brand new flag just to strip the trans colors off the real one is ignorant at best and disingenuous at worst. That's not a new flag, that's the Philadelphia Pride Flag. The Progress Pride Flag is free to use for non-commercial purposes and small businesses, but large companies—like Tumblr—would have to contact the creator to discuss licensing. Them using the Philadelphia Pride Flag—which has been published directly by a US government organization, and thus not copyrightable—has nothing to do with excluding trans people and everything to do with money. Obviously there is criticism to be made here: although Tumblr hasn't been doing too hot financially, they would have the money to support an independent queer creator. It's hypocritical of them to roll out a pride event feature like this while jumping through hoops to not give money to a queer artist.

Another argument I've seen made on that post is that Tumblr allegedly updated the animation to include the trans and intersex flags, as a response to "the backlash" (the new version is pictured below).

I personally doubt this. First off, "the backlash" wasn't even remotely big enough to prompt action from staff. A 90k notes post is not going to back staff into a corner. Especially when the thing they're being accused of—deliberately removing the trans color from the progress flag—isn't even true.

In fact, the trans flag was included in this feature from the very start. On browser, the flag you are shown changes based on the tags of the post: posts tagged as #lesbian will show a lesbian flag when liked, posts tagged as #intersex will show the intersex flag, etc. Posts with multiple LGBT tags will show a random flag among the labels tagged, and posts with more generic tags (like #queer or #lgbt) will show the Philadelphia Pride Flag. As far as I can tell this was always the case on browser: trans and intersex people were included from the get-go. You can see this feature being highlighted in this post and this post.

On the mobile app there is only one flag animation that will play regardless of the tag: the one pictured in the previous screenshot, with the Philadelphia Pride Flag being followed by the trans and intersex colors. I suspect that what's happening here is that the user originally alleging that staff removed the trans flag was on browser, while the users assuming it had been changed on a later date where using the app. Unfortunately I don't have a way of checking this, and there is a possibility staff had originally included only the Philadelphia Pride Flag on mobile and then decided to add the trans and intersex flags for the sake of being inclusive. However, if that's the case, it's just them making the app more consistent with the browser experience, definitely not them "trying to cover their tracks" like some users suggested.

And here we get to the topic of the posts I'm criticizing being tagged as mature, and how this ties in with trans people on this platform being disproportionately affected by bans (which is something that truly does happen, just not in the way people think—I'll get to that later). Some folks have been using what I debunked earlier as evidence of staff being transphobic and purposely targeting trans people. This argument highlights a fundamental misunderstanding of how Staff works: Staff is not this one cohesive entity, but it's rather made up of an engineering team, and a design/art team, and a public relations team (these are the folks that run the Staff account and reply to user feedback. On a side note: it's important to bring your feedback to them, but being an asshole is pointless. They're not the ones doing the things you are upset about in the first place), and a moderation team. This feature would have been rolled out by the art/design team with the support of the engineering team, so even if it was transphobic—which it is not—it would have had nothing to do with the moderators.

So here is the billion dollar question: Is Tumblr moderation transphobic? Yes, but not in the way most users think. There is this common belief that the people working as moderators for Tumblr are actively transphobic, and purposefully banning trans people—in particular trans women. This is not the case, at least not at scale. Like most large social medias, Tumblr uses a centralized hybrid moderation system. To put it simply: the process of flagging posts as mature (and it seems also banning users, unfortunately) is automated, with humans doing mostly reviews on the decisions that automated system takes. The algorithm doing the moderation is centralized, which means it was developed by people contracted by Tumblr, for Tumbr and Tumblr only, and that the public does not have access to its inner workings. We also know Tumblr has been aggressively downsizing its staff in the past few years, with headlines describing it as "running on a skeleton crew" as far back as 2023. From my understanding, there is less than 100 people currently working to push fixes and new features to Tumblr, and although the number of moderators is unclear, I doubt it's enough for the more than 130 million active monthly users. All this to say that posts are currently almost exclusively flagged by the automatic moderation system. It is possible to ask for manual review, but I don't think there is enough staff to actually go through all the requests.

So this is an issue of the automated moderation system, it's not a matter of staff being actively transphobic. Discrimination towards LGBT users was not hard-coded into the system, but rather a product of unconscious bias in development and pollution in the training data— likely caused by porn bots taking over LGBT tags—, and it is fixable. The automated moderator system targeting queer people was actually acknowledged in a January 2022 settlement, in which Tumblr committed to hiring an expert in LGBT discrimination and to take tangible steps to address the issue. Which is great, but things like this are not a one-and-done, and with how big the staff downsizing was since 2022 I doubt they're still keeping up with it.

Autistic people being difficult because they tend to approach problems differently and get flak for not doing it ‘correctly’ even when following instructions as closely as possible, and then asking minute instructions because they want to avoid that very scenario?

Neurotypicals asking “why’d you do it that way?” when they mean “this isn’t how I wanted it to be done” and then getting angry when the person begins explaining their thought process on why they did it that way.

Neurotypical people are shit at giving instructions...and it doesn't just suck for us in odd-brain-out community. It also sucks for anyone who doesn't share the culturally/experientially encoded assumptions that get left out of said instructions. That causes problems for people like, say, non-native English speakers or immigrants or people with disabilities affecting hearing and speech.

For anyone who hasn't seen them before, Hidden Search Operators are handy tricks you can use when you're either searching or filtering AO3.

summary: string is a generic way of explaining that you can search AO3 for a specific word that appears in a summary. You can do this from the search bar in the header, from the Any Field box at the top of the Advanced Search form, or from the Search Within Results box at the bottom of the filter menu.

Examples:

summary: Bruce

summary: "Bruce Banner"

summary: Bruce OR summary: Banner OR summary: Hulk

You need to put quotation marks around your search term if it is more than one word. The quotes make sure that the site searches for those two words together.

The other two operators listed work best in the Search Within Results box.

expected_number_of_chapters: 1 will return results where every fic has only 1 chapter currently posted.

You can use expected_number_of_chapters: -1 if you want results where every fic has more than 1 chapter currently posted.

I saw your tags @jackdaw-sprite and I figured I could offer you some help? You've got a few options to find fics that are actually about your fave who is often sidelined.

My own personal go-to is to use the Search Within Results filter. You can find the filters by clicking/tapping on any tag. If you're on a wide screen like a laptop, they'll be on the right hand side of the screen. If you're on a narrow screen like a phone, you'll need to press the Filters button at the top of the page to open up the menu.

In that menu, scroll down until you're near the bottom. That's where you can find the box labelled Search Within Results. In that box, you can type

summary: Renhardt OR summary: Rendog

When you filter for that result, you'll get 19 fics with that name mentioned in the summary, and if the name is in the summary you've got MUCH higher chances of them being central to the fic.

There's also a way to find fics with a tag in the X position. Fan coders have had this same issue, and so they've created userscripts that you can add to your web browser (like an extension). Here are a few that might be helpful for you to try:

AO3 Actually Relevant Characters - Sorts and highlights works based on how relevant the tagged character seems in the work's tags and summary.

AO3: Prioritise my Faves - Hide work if chosen tags are late in sequence, or if blacklisted tags are early

AO3 Secondary Character & Pairing Filter - Replaces works with placeholder if the tags you want aren't the first few tags, with option to show work.

Editing to add: this version of the post has a syntax error in it. To remove oneshots with a filter, use

never had existed if

we were all lizards.