[01:09:41] it's not about the fact that it's deadly, [01:09:53] it's about the fact that I don't know how to use it [01:10:00] same goes for all tools [01:10:01] cars [01:10:09] printers [01:10:22] forklifts [01:10:41] you name it [01:11:15] maybe not a course but at least RTFM [01:11:40] aka learn before you use [01:12:27] @starmanakremis I've been reading, I just haven't felt the need to respond [01:12:32] kk [01:12:44] and there are free courses [01:13:11] maybe I should've made it more generalized [01:13:31] cuz there's things that don't justify a course [01:13:36] like coding [01:14:02] also I do agree [01:14:52] so I am a supporter for government funded computer courses [01:15:15] and it classes in school [01:15:28] that teach some programming language like python [01:15:34] it's just so useful [02:13:48] Disabled it entirely the moment it popped up. I don't want my secrets to be exposed on Google, bruh. 😭 [02:29:43] I'm watching the horimiya anime right now it's so peak [02:47:31] 👀 [05:13:43] claire recommended a hot shower and that made it go away [05:13:54] i was groaning in bed and had like 2h of sleep [05:14:02] after it was gone, i honk shooed until 2 pm [07:11:32] https://www.reddit.com/r/namesoundalikes/comments/1xhind2/soundalikes/ [07:11:35] stardew wiki mentioned [08:31:19] I own 105 decors [11:44:31] https://www.eurogamer.net/ai-fan-wikis-errors-complicating-games-culture [11:46:40] the ammount of slop sprawl zero wikis bro [11:46:58] it goes at least to the 3rd page [11:47:09] yeah [11:47:12] I fucking hate those wikis [11:47:39] speaking as someone who gives a shit about building good game wikis [11:49:11] sorry, not super related to the article, but eurogamer has the worst cookie consent form i've ever seen [11:49:44] i literally cannot turn off targeted advertising??? what is the point of having a cookie consent form if i can't turn off nonessential cookies 😂 [11:50:15] we need gdpr 2.0 [11:50:17] ffs [11:51:50] that forces websites to implement an interface to automatically disable cookies [11:52:30] also [11:52:38] shoutout to hooded horse for creating legit wikis [12:09:41] Abuse and spam, abuse and spam [12:09:46] Infinite content, but no content [12:10:18] [1/2] I don't think we've really had this happen on miraheze but I see no reason why miraheze wouldn't be used as an engine for some of these, and our policy basis to hit them is weak. It might make sense to specifically target this category of wiki, and perhaps determine some other AI attitudes from a rough miraheze perspective in order to guide the attitude of [12:10:19] [2/2] the platform itself [12:11:08] I do not favor an absolute purist attitude against ai in all regards and room for generosity when unclear, but at least wikis that stewards determine are clearly ai campers are something I would like a direct basis to whack [12:11:37] this doesn't even seem to really be a thing on wiki.gg or even fandoom (though there can be some garbage ai or non-ai wikis in the latter but it's expected) [12:11:50] well garbage in general is quite its own thing [12:11:52] the ai wikis i'm talking about are always outside the major wiki platforms [12:12:53] [1/2] for more outright sinister conduct we're pretty strong but I mean for a range where there is willful attempt to operate functionally within miraheze boundaries, either as efforts to compromise a wiki's community via sibyl actions which I do think we have good basis to exterminate but can become only harder over time to actually defeat, or as efforts to predicate an entire wiki on ai [12:12:53] [2/2] development, which is a boundary I have an unprovable inkling that the voting body of miraheze generally wouldn't like [12:13:12] unprovable without clear community determination I should say, which I do think is preferable over just assuming on these things [12:13:21] me neither, I can see some use for AI when debugging some specific or planning something quick [12:13:58] but it is clear that the vast majority of the vibecoded wikis that I'm talking about are either empty, full of misinformation, terribly designed, or any of these and more [12:14:01] my leaning overall is permit unless strong basis to prohibit; this is one domain I think having that ironed out in advance would be useful. Current campers are independent domains, but the landscape can and will change soon, and unfortunately some of my worse predictions only become more substantiated with time [12:14:54] And there have been occasions that are only more recent, though rare and probably not ai related, where there are requestors specifically attempting to camp domains which is something I want to proactively rebuke [12:15:18] if someone used ai to correctly gather up and organize game data to set up a legit and high-quality wiki full of correct and complete info, I'd say this is more acceptable [12:16:10] every tool really comes down to use, the problem of ai is it lets you spill so much with no effort that the slop really does ruin it for the good in practice more often than not [12:16:26] that's another major issue with these vibecoded wikis [12:16:45] they take up good domains like gamename.wiki or gamenamewiki.org and fill them up with slop... [12:16:56] so much so that even when you do something good with it, the very fact of how you do it permanently neuters its appeal to the kind of people who would be most spirited and helpful to do better [12:17:21] when some popular or high-quality wikis can have those kinds of urls [12:17:31] we will see regulation specifically when this is used enough against big money that big money says no wait stop it [12:17:46] like balatrowiki.org or pizzatower.wiki, those are reserved for high quality wikis [12:17:58] domain camping has always been a threat but yeah its been pretty supercharged in recent years [12:18:07] vs the same kind of urls used for a new game that came out on steam last week [12:18:47] you really just can't go off domains very well in general and it is ever more necessary to develop common trusted cores for things where both people invested and randos are able to quickly say yep this one is poor and that one makes sense [12:19:16] it is my long term hope to develop some connections to this end with other farms and projects [12:20:08] though presently my wish is to tidy in house somewhat more and to stay afloat as general stewarding, when done to spec, tends to leave very little bandwidth for loftier ambitions as @notaracham knows more than well [12:20:31] as someone also bogged with TS and other mundane duties [12:21:28] this is where I hope and wish to see more people who are not encumbered with big buttons to be more visible in the community end where they might be able to help push forward aspects of miraheze that the roled volunteers have limited bandwidth to do more than facilitate [12:22:43] the sooner we are stronger in these regards, the better we can resist the crippling waves of slop, corporate interests and our own disintegration if we lose too much momentum, as has always been a threat on miraheze [12:26:24] man I remember when the fact that it was hidden from Google Search was actually a criticism of Discord, because the old forums back in the day were searchable on Google and were readable without logging in or anything while Discord is all hidden and unsearchable [12:26:36] how the tides have turned [12:33:43] I guess this is an getting old thing [13:04:28] [1/3] I don't really think the law or our current systems, be they the EU itself or member countries at a national level really are capable at all of fixing what's wrong with the Internet currently, let alone the world, rot runs way too deep at this point I think. The most we can do is trying to elect people that put us in the correct position so that when [13:04:28] [2/3] the big collapse comes (personally I think this will be the AI bubble and climate change turning us more and more into an oven), as we are currently seeing signs of across Europe with the protests, most notably France, we can put the blame on the correct people and systems and try to rebuild from the ashes. Although if the French election polls are an [13:04:29] [3/3] y indication apparently people aren't tired enough of suffering, we want more and harder suffering 😮‍💨 [13:05:21] it was my criticism and I am favorable to the decision for this reason [13:05:34] although of course the way its being implemented is rather below par to me [13:05:51] not unusual for discord to disappoint me in that regard though... [13:05:56] well it does go kind of against how Discord was built to be to begin with [13:06:42] I view it specifically in terms of spaces that are designed like oldschool forum stickies where you can find this or that and this or that might be crucial information, except it was buried in the discord silo, which I've always greatly disliked [13:09:12] [1/2] however different spaces are used differently, not every public channel is equal or may at all want the public exposure, so from that perspective I think such a setup should be expressly opt in at a server/per channel level, transparent to channel participants, not automatic for old channels, and rolled out in a more consistent way that doesn't result in a gloopy patchwork where you [13:09:13] [2/2] have some people opting out here and breaking the flow, or they don't show enough there since the wording suggests its not a browsable transcript like how you might be able to use zulip, or designed with features that are piecemeal opt ins that make it a mess at all levels from user agency to a server's ability to run itself as it would prefer [13:09:35] I don't think it should be a choice at a user level to opt out of a public channel, it should instead be the user's liberty to not engage in such a channel if they don't want what they post to be open [13:10:38] in the case of this server everything is pretty much broadcasted in the same way anyway and probably more thoroughly than discord's approach seems to be [13:11:19] yep, I also think this shouldn't be an user level preference, I guess this was them trying to get ahead of an outrage maybe, but it kind of defeats the purpose. [13:11:42] it does kind of surprise me they offer it at all because the locked in, you need an account to visit behavior has always been a feature I would think [13:11:52] so indeed the implementation in practice is just a mess [13:12:04] I don't think Discord has the culture that made the communities of old okay with public forums anyway, so it's kind of a headscratcher [13:12:20] I do perceive that yeah [13:12:37] I would think a lot of servers would have zero interest engaging in that or only to engage selectively [13:13:02] My guess is this has been on a wishlist somewhere from way back in the day when this was a criticism levied at Discord, but things have changed since then [13:13:05] and so, here's an option to opt in to have a public facing, and that's it [13:13:31] my head's sufficiently stuck in old nostalgia that I have no scruple making it happen [13:13:38] just needs to happen in a sane way [13:13:58] but it is discord and discord has trained me the way big companies have trained me to rethink common sense [13:14:43] cause like, make your announcements channel for a service public, that's darn useful, etc [13:15:19] its functionality that draws me to fluxer's roadmap and is a selling point of zulip [13:15:36] of course the fact people have barely ever heard of either I guess is case in point [13:17:46] yeah there are definitely some wikis on miraheze that are basically just AI slop and I think that is a blind spot in our current policies [13:18:22] I would support a motion to disallow wikis with substantial AI-generated content [13:21:55] Yessssss 🙌 [13:22:17] If you want more anime recommendations afterwards hmu 🤙 [13:22:50] <_arawynn> There has been a request either yesterday or the day before that wanted a Wiki for AI agents to post/interact. [13:31:31] that's moltbook [13:31:43] they should stay there [13:31:55] and not interfere with our stuff [13:36:59] wikis about AI slop or AI slop on wikis [13:37:11] cuz there's a big difference [13:37:25] though the two can often overlap [13:37:32] looks at IBW [13:37:46] I'd give a draft a few passes to help see it to fruition [13:38:09] rfc really is the mechanism for these things to be expressed despite flaws and is something I would like to proactively encourage [13:39:03] about ai slop is not inherently an issue but the audience appeal makes it so, but I think it is adequately covered by volunteer level restrictions born of standing policy [13:39:14] slop itself is one where the gap remains somewhat large [13:40:09] I would say pretty much the safest approach is to directly hit wikis subject to sensible review as being evidently slop driven to try and avoid certain fights around the in between unless that is something there is a strong broad sentiment for stewards to get into [13:41:14] the problem is if it is something that depends on using ai detectors and arbitrary judgement and figuring out which point on the line is 'slop enough' or it becomes a matter where we're trying to detect any use and 'prosecute' from that, then imo it is untenable. But wikis all the way on the end where its just obvious they are agentic spews, I think that can be struck pretty neatly [13:42:58] [1/2] the technical end has proactively responded although it is an ongoing question how to approach wikis that might want certain kinds of scraping and ones that don't, that too should be a community discussion imo, but how much we can do either way is rather limited. If there are options to operate filters based on user agents that might be configurable per wiki there might be a chance t [13:42:58] [2/2] hough [13:56:06] wikis that _are_ AI slop. it's probably not the end of the world if you used ChatGPT for some templates/modules/styles, scripts are kinda iffy but probably allowable. But if the actual content of the wiki is just AI-generated, that's not providing any educational benefit to anyone. [13:57:17] As I figured [13:57:41] Some wikis about AI slop can be legit, provided that the documentation itself is legit [13:59:35] wiki.gg has a policy making up this distinction for example: https://support.wiki.gg/AI_policy [14:00:22] Tl;dr what to do on Miraheze regarding AI slop wikis would be complicated [14:00:27] But what if someone has ideas but struggles to put them into words, sometimes experiences writer's block, and maybe... doesn't have much time? As you can see, what if they use AI? [14:01:22] If the wiki is about your ideas I don't really understand why you can't put whatever you're typing to chatgpt into the wiki itself [14:02:13] Sorry, I meant ideas for creating fictional articles or articles on any topic. [14:02:38] so it's not your ideas and not your words [14:02:52] why even create it? Like what are you getting from it? [14:03:47] [1/2] in the interest of something over nothing my preferred approach is a blanket basis against agentic 100% ai invocations and dodging next levels down where it is a supporting tool to perceptible human effort and intent although past a point the line would be blurry, and at that point the idea of invoking wiki reviewers as a majority council to decide certain things might need to be use [14:03:48] [2/2] d [14:04:32] essentially to focus on the far end of laziest unconstructive uses as a first resort and let more arguable uses litigate seperately [14:04:42] yeah that makes total sense [14:04:56] I see some function in having an idea in abstract and ai being what realizes it to completion tbh, and that seems to be closer to the sense ucronista is going for [14:05:25] it is less effort than the manual process of realizing all aspects and the execution, but I give it weight as a something [14:07:21] but if it realizes essentially every aspect outside of a single paragraph of 'make me a thing, here's some general parameters' and that's what the wiki is then I think that's the line that needs to be drawn, or even if it is that prompting that devises every article on the wiki even if the articles are done that way one by one [14:10:05] [1/2] if the human is intangible its about as good as there not having been a human at all and its that level that is most egregiously against the people sentiment miraheze is built on that I think it is in need of proactive curbing. It's troublesome enough that you have so many humans with so little investment in what miraheze is as a community project, but then you have a technology that [14:10:05] [2/2] reaches that with one prompt, and then becomes a social engineer in the second prompt to usurp miraheze sentimentally. This is kind of a stretch maybe, but not a long one at this point imo, and it should not be the case we have broad windows that would allow this theory to be credibly tried [14:12:38] yeah I don't think it's a stretch either. It's insane how much the technology, and peoples understanding of it, has progressed recently [14:14:07] better conveyed the sentiment in edit now I think [14:19:06] [1/4] its just the miraheze version of what is now the internet problem though, where it will soon be impossible to validate humanity because you can prompt a model into being roughly human enough in behavior, and eliminate the long expectation that whatever it was you were talking to, it was probably a person, or at least sentient, where old bots would be fatally flawed after cursory insp [14:19:06] [2/4] ection. So now you either have to go off sheer verifiable age, off rooted networks of interconnected people that hopefully endure against deep seeding attacks, or oblige the businesses that want to tie your concrete in person identity to the online self with its own many problems, and for those determined and malicious enough, could still probably be faked. And if you can't answer th [14:19:07] [3/4] is then it is becoming easier and easier to see an outcome where any community project or member in it is completely fakable. And maybe you'd be fine with something that acts human; alas, the human required some kind of effort to start as a perfectly trustworthy entity and then completely 180 into a saboteur fully committed to damage, with prompting it would require none. [14:19:07] [4/4] You now have your daily dose of raidarr existential crisis [14:29:04] I feel so bad for the kids [14:35:26] Thank you for the feedback; that is certainly an area for improvement. [14:37:07] [1/2] > You now have your daily dose of Raiders existential crisis [14:37:07] [2/2] Welcome to my world [14:37:18] Or our world, the indie wiki world [14:38:42] A bubble of a beautiful world being surrounded by chaos [14:41:31] [1/2] honestly it does sometimes feel like that. [14:41:31] [2/2] We did have an instance of a person rewriting a page that was originally created with AI today, and they clearly put a lot of thought into it, so very happy about that! [14:48:06] [1/3] Also, this is a bit of a topic shift, but for the past month and a half, after recurring issues with accuracy/misinformation, and an increasing workload for people who care about accuracy, we decided to add an abuse filter asking people to cite their sources when adding new pages. This is a wiki about real life seals so supposedly all the information does have to come from somewhere. [14:48:07] [2/3] I collected the results of it here and I would like to hear opinions on how to get it to improve and whether it's valuable in the first place [14:48:07] [3/3] [14:51:03] (sorry if this isn't really off topic) [14:56:24] not off topic [14:56:50] my wiki has blanket ban when you do actual AI slop, not high quality AI usage [14:57:13] ai image gen will get you banned instantly [14:59:48] We have a blanket ban on LLM slop writing and images [15:00:19] (Not limited to Miraheze, also on wiki.gg and Weird Gloop too) [15:01:14] we have a rule saying you can use AI to help you research and technically you can write articles with it if they follow the same criteria everyone else is subject to (which they... usually don't, because they hallucinate stuff) but if you slop it up you're getting blocked [15:01:27] most of my wikis have a no AI for articles rule tho [15:21:14] [1/2] Yep, using AI doesn't mean you left them unchecked. [15:21:14] [2/2] Some stuffs are clearly faster if you let the AI do for you, you only need to check the result. [15:22:48] _enter the Turing test_ [16:09:24] I’ve never been more devastated in my life [16:09:35] I finished all 17 volumes of Horimiya [16:09:49] Why did it go so quickkkk [16:10:53] We are now seeing actual harm to correct and durable information as a result of these [16:11:52] There's an increasing trend of game focused wikis for new wikis that place extremely well in search results but contain total hallucinatory nonsense [16:12:31] What you might find in search results one day gets pushed out in favor of the nonsense the next [16:13:27] Because they can be stood up so quickly, they get a bit of a founder effect and are incredibly hard to quash even though they contain nothing of value [16:13:54] Made like a whole ass rant about these """""wikis""""" earlier today in the WG server [16:13:59] And I think here too [16:14:36] Crimson desert and control resonant are two examples that immediately come to mind as plagued by these fake wikis [16:14:56] Yeah [16:15:29] My bar is low enough to view Fandom or even Fextra counterparts as legit [16:15:37] Cuz at least these can contain real info [16:16:03] I don't know how deep the rot spreads but tbh if Miraheze gets a reputation for having slop wikis, that'll be even worse than the whole reception wikis fiasco [16:17:35] I'mma be real here, Miraheze will be fine [16:18:02] I know some random slop wikis exist, but I have yet to actually find one, other than IBW 😅 [16:19:42] Fandom probably has a lot of slop wikis too [17:13:05] We've closed a handful for content policy vio, but it's been thankfully limited [17:13:15] The most egregious one was 100% AI-generated health information [17:15:14] 😮 [17:15:32] That's not only harmful, but illegal... [17:16:09] would it be okay if you can take a look at https://minecraft.miraheze.org ? [17:16:21] that wiki is incredibly weird [17:16:30] and is seemingly plagiarized from mc.wiki for the most part [17:17:06] I think the staff made a post about that telling people not to. [17:17:07] maybe not anymore like it used to, but I think it should still be reviewed [17:17:23] It's already been reviewd and been raised to them, they're working on it [17:17:35] that's good [17:17:50] Do you mind if I DM you rq Aracham. [17:18:27] still a weird ass wiki with the ai logo, but as long as it is not causing harm anymore, then I guess it's alright now 😅 [17:19:00] If you find anything clearly infringing, please feel free to send it over to the admins, the steal a brainrot victory character is the most active over there [17:19:08] Yeah, I despise seeing Generative logos, just so lazy. [17:19:16] In a pretty hectic period at work, will respond when I can [17:19:26] https://minecraft.miraheze.org/wiki/File:Wiki_logo.png [17:19:44] this revision has more effort put into it lol: https://static.wikitide.net/minecraftwiki/archive/3/3b/20260829150729%21Wiki_logo.png [17:20:03] even if it is still pretty bad lol [17:21:04] Exactly. [17:21:41] I still can't believe people find any Generative images good looking, they're always terrible. [17:21:44] I mean, even if it wasn't AI, it is still infringing the fandom mc wiki's logo [17:21:54] Why does this wiki exist when the main one is already established? [17:22:10] no idea [17:22:10] Because the Minecraft Wiki isn't on Miraheze. [17:22:18] And it was created a long time ago afaik. [17:22:48] but terraria or runescape aren't on miraheze either [17:22:52] If they want to make a Minecraft Wiki then they can, as long as one didn't already exist at the time. [17:23:05] Which obviously one didn't. [17:23:14] i think this is a different instance from the original miraheze mcw [17:23:29] the original one was a fork that existed before mc.wiki [17:23:33] this one is not [17:23:48] Eh, idk then. They've already reviewed it, not in my realm. [17:27:57] notified them [17:32:07] We do not consider offsite wikis like MCW in saying that someone can't create a similar wiki on MH [17:32:30] There was a minecraft wiki on MH, it got shut down, and when the subdomain became available the current crew moved in. [17:34:24] to alter ^ practice would require successful rfc on the matter imo [17:34:34] (Now, is creating a separate minecraft wiki when a very good indie practice a good idea? Probably not!) [17:34:39] which I am open to in the event that's a strong sentiment, policy and practice are not immutable [17:36:05] we have never had a practice of judging based on the question 'is it wise/practical to make the wiki' which is deeply rooted in the more hands off libertarian ethos at founding; it is a question somewhat more feasible with the more scrutinizing request processes of today [17:36:51] but even today maintaining 'no unauthorized splits of wikis we host' is tricky and not automatically possible, to some extent its luck for volunteers to catch it [18:48:40] Damn, ao3 is gonna be down for around two weeks https://bsky.app/profile/status.archiveofourown.org/post/3myzstnmohf2o [18:49:18] Something bad musta happened to increase maintence time from 12 hours to 2-4 weeks [18:50:18] <_arawynn> 😱 [19:02:20] 12 hours to 2-4 weeks kinda makes me think hardware failure [19:10:08] Same [19:10:27] yeah considering they said here the data is all intact: [19:10:45] That does kind of make sense but I'm wondering what they mean by degraded [19:11:19] like if they can only get worse hardware would it just be slower lol [19:52:43] to me a degraded state means less functionality and more challenges [20:54:38] [1/2] colorful [20:54:38] [2/2] https://cdn.discordapp.com/attachments/615786602454581249/1565468928166203422/Screenshot_20261029-1719562.jpg?ex=6ae5040d&is=6ae3b28d&hm=9734d3a414ac8f9098b649e58571a5c4dc3bff6601e16f68eff96fa5ec260834& [21:50:20] [1/2] I want to showcase some of the amazing CSS work @sailing has done for our project (namely the "characters present/mentioned template and the page flipping template) [21:50:20] [2/2] https://kongarchived.miraheze.org/wiki/Original_Questions_-_Smash_Hits,_June_2001 [21:50:25] LIKE GUYS THIS IS SO COOL [21:51:23] getting praised for making divs with a border 🙏 never thought I'd see the day [21:53:22] at least wait til I do the mainpage [21:54:05] Main page is going to be the craziest thing ever [21:54:36] But seriously dude to me this is like some of the most impressive shit ever it really does mean the world to me (and other people who don’t code haha) [22:31:11] or they're hauling ass from one location to another [22:59:47] For some reason, I confuse brown, like brown hair, with bright green a lot [23:03:14] Could you be colorblind perhaps [23:03:34] Those aren't very confusable colors [23:03:54] I got a flag for that on an eye examination [23:04:09] But I use contacts so thats the biggest thing [23:04:25] The icon for GA is squinting eyes…hehe… [23:04:35] Before I got glasses, I would squint a lot [23:05:16] I was curious and took one of those online tests (I doubted I would get an accurate answer though) and got so many wrong it made me do it again [23:05:24] For red-green, I think [23:05:32] Maybe I’ll bring it up with my eye doctor [23:06:26] [1/2] YOU’RE WRONG MIYAMURA AHHH [23:06:26] [2/2] https://cdn.discordapp.com/attachments/615786602454581249/1565502098483847340/IMG_0972.png?ex=6ae522f1&is=6ae3d171&hm=2d913167da99328c2537b661cef677e374b3a3fc2d1ef2382826eeee9e5c3388& [23:06:58] I think you should [23:07:14] [1/2] What's this say /silly [23:07:14] [2/2] https://cdn.discordapp.com/attachments/615786602454581249/1565502292868866048/IMG_7f5c8305-b7d4-4728-9292-1aaedc4fe3f6.png?ex=6ae52320&is=6ae3d1a0&hm=9387d29079edeb42c55ed927b24e51e56ee609b0ded15055896465f20e24484f& [23:10:28] Either 81 or 21 [23:10:33] I think [23:10:38] 👀 [23:11:07] It says 74 [23:11:12] What [23:11:14] No it doesn’t [23:11:18] How is that 74 [23:12:10] My eyes betray me?? Are we talking about the same image??? [23:12:28] https://cdn.discordapp.com/attachments/615786602454581249/1565503615920251042/IMG_7f5c8305-b7d4-4728-9292-1aaedc4fe3f61.png?ex=6ae5245b&is=6ae3d2db&hm=aaafd43b1baf4cb9fcc69e30869018214ab83f404e841ee0d2708ca1201fe3ad& [23:12:32] 74 [23:12:48] Ok wow [23:12:53] It’s worse than I thought [23:13:07] Wait how is that not a one? [23:13:57] There's some dots in a darker green and others in a brighter green in the 4 to make it trickier [23:14:31] The darker green ones make up the bits that make ir more distinctly a 4 [23:14:38] Huh [23:15:16] diagnosed by colored balls on the internet [23:16:39] The colored balls are very good at diagnosis [23:16:52] You look at them and they say "I know what you are" [23:23:39] [1/2] It might be more obvious once you apply `filter: hue-rotate(180deg)` to it to flip the colors. [23:23:40] [2/2] https://cdn.discordapp.com/attachments/615786602454581249/1565506421259509770/image.png?ex=6ae526f8&is=6ae3d578&hm=09ad7ec87377b217c5ce7941b044c0928e00292fe7e52f7ff501447af6fcdfdf& [23:24:18] I still don’t see it at all!!! [23:28:18] Sometimes I confuse lighter shades of black for like blue ish [23:28:22] Or even purple [23:28:24] I definitely think you should talk to your eye doctor about this [23:28:39] You sound so colorblind [23:29:07] Well, we’ve always been more focused on the bad eyesight I think it kind of slipped through the cracks!!& [23:29:14] sounds like colourblind [23:30:00] i def see 74 and i know im very colourvoyan