#not to do well by your non-existent algorithm

[anulithots]

I think this is my favorite drawing I've ever made.

It's Kamari!

I never know how to draw their full bodies. (hence doll for Anuli.)

So Kamari has wings, is digitigrade, and has bracelets and things that will clink against each other.

#the land of the fallen fairies #writeblr #Kamari the dryad #you know what? #Tumblr? #I want this to be seen by those who would want to see it #not to do well by your non-existent algorithm #that doesn't work #I don't trust the tags

[jams-sims]

Last night was a fucking fever dream. I laughed so hard until I cried and I woke up with a massive headache.

But I'm finally able to talk about what the problem is and people have already picked it out. Mainly when it comes to the red team.

They took a group of rper and forced them to play RUST, if you don't know what Rust is. Your YouTube algorithm isn't fucked like mine. It basically a game where no one wins. The whole point is that you build a base you farm materials for your base and you raid other people for their materials. That it's there is no end goal, but people sweat at that game.

An that's what Purgatory is, instead of each team going off to build multi bases, they have one home base. Instead of setting spawn points they only get one spawn point. An when they die their stuff stays on their body for anyone to take just like RUST.

The problem is they want player conflict, and for it to be something different and new unlike the leisurely life they led on quesadilla Island. They want that conflict that beautiful fitmc, where he turns back into the hardcore 2bt2 player and now that just everyone's problem.

I'm specifically going to look at the red team because they decide to put all the Lore heavy people together and they stuck survivalists with non players/builder. Charlie, Jaiden, Baghera are all pretty casual players. An Carre is never coming back after yesterday, if he does hell has frozen over.

Now I can take or leave whether the red team is cursed or not. I'm sitting directly in the middle of yeah it would be too obvious that red is cursed. Cause they legit have admin fucking with Philza. To it would be ironically funny if it was just the red team. Philza being the one to make the team curse cause the narrative has been beating his ass.

Anyway- the first thing that would have to be changed is the system of which people gain point. They want conflict, fighting is the best way to do that. Make it so killing people is gives you more points. But in turn you have to turn up the specialist factor so people don't just go on manhunts. You have to make it so if you are gonna go out into the world to kill. The world has to be beating your ass to make it fair.

This is the only way you can allow survivalist characters such as Philza and Cellbit to survive is by forcing them to make homes and equipment to survive natural disasters and other players. An you gain more by adventuring around as well. Movement should be awarded.

It's no longer race to the top of a volcano to turn in quest. It's now, race to the top of a volcano while it's erupting, to turn in quest to get equipment.

This allows players like Baghera, Jaiden and Charlie to focus on more homestead task. Building, food supply, and mining material. To make it so that if you can't turn in a quest you aren't fucked. I'm assuming the admins want NO working together with other teams. They want them to suffer thus we gotta rearrange with that suffering means.

For instants: If you build a base whether that be underground or not. You can be subject to raids, which means you actually gotta plan and hide your shit. If they find it oh well.

Your base can be destroyed but hard to craft items such as let's say enchanting table and villagers are a no go you can not take them or destroy them. Anything that would destroy a player morale shit like that is a no. We want it to be something you can work back towards and think about how do we make this thing better.

Natural disasters are more common than the minor problems, the your inventory moving on it own is less, and your controls swapped are less. We change that too, every 2 to 3 hours the weather changes. And it stays that way for the next 2 to 3 hours. In game or out would be up to the admins.

Task still exist, global still exist, but now it's a race to the top. The only time you can turn anything in is during a Natural disaster. Once a player picks something up. From their turn-in they become immune to damage certain amount of time. Then it's a free for all but that gives the turn in team time to run.

An once they enter their base they are considered safe. You have earned supplies for your team, immunity from raids and a shit ton of points. That's for the team who turns it in first.

Player can still turn in task and earn suppliers. That supplies is limited but even if your last you still get something.

How do you stop people from camping someone base or Gobal.

Easy, immunity for the team to do turn in first last the whole day until the next day. Where it loses its effect, that gives the team time to move their shit if they so wish or be ready to defend.

Global is even more easy global is locked until it is time and 2 if an admin catches you camping any where near it. You get one tapped don't be a camper.

How to stop people from camping others bases. (I think that just falls under don't be an asshole. An maybe making a shield around the spawn point with an area of effect could also help.)

I didn't want to complain and not give like legit fixs for the problems they have. Cause it's one thing to scream and say stuff is broke and it's another to actually come up with ideas to fix it.

#qsmp #philza #jaiden animations #slimecicle #foolish gamers #baghera jones #cellbit

[goldsasa]

Dear Sirs!

(or have some ladies also signed?)

A few days ago, you, Mr Musk, together with Mr Wozniak, Mr Mostaque and other signatories, published an open letter demanding a compulsory pause of at least six months for the development of the most powerful AI models worldwide.

This is the only way to ensure that the AI models contribute to the welfare of all humanity, you claim. As a small part of the whole of humanity, I would like to thank you very much for wanting to protect me. How kind! 🙏🏻

Allow me to make a few comments and ask a few questions in this context:

My first question that immediately came to mind:

Where was your open letter when research for the purpose of warfare started and weapon systems based on AI were developed, leading to unpredictable and uncontrollable conflicts?

AI-based threats have already been used in wars for some time, e.g. in the Ukraine war and Turkey. Speaking of the US, they are upgrading their MQ-9 combat drones with AI and have already used them to kill in Syria, Afghanistan and Iraq.

The victims of these attacks - don't they count as humanity threatened by AI?

I am confused! Please explain to me, when did the (general) welfare of humanity exist, which is now threatened and needs to be protected by you? I mean the good of humanity - outside your "super rich white old nerds Silicon Valley" filter bubble? And I have one more question:

Where was your open letter when Facebook's algorithms led to the spread of hate speech and misinformation about the genocide of Rohingya Muslims in Myanmar?

Didn't the right to human welfare also apply to this population group? Why do you continue to remain silent on the inaction and non-transparent algorithms of Meta and Mr Zuckerberg? Why do you continue to allow hatred and agitation in the social media, which (at least initially) belonged to you without exception?

My further doubt relates to your person and your biography itself, dear Mr Musk.

You, known as a wealthy man with Asperger's syndrome and a penchant for interplanetary affairs, have commendably repeatedly expressed concern about the potentially destructive effects of AI robots in the past. I thank you for trying to save me from such a future. It really is a horrible idea!

And yet, Mr Musk, you yourself were not considered one of the great AI developers of Silicon Valley for a long time.

Your commitment to the field of artificial intelligence was initially rather poor. Your Tesla Autopilot is a remarkable AI software, but it was developed for a rather niche market.

I assume that you, Mr Musk, wanted to change that when you bought 73.5 million of Twitter's shares for almost $2.9 billion in April?

After all, to be able to play along with the AI development of the giants, you lacked one thing above all: access to a broad-based AI that is not limited to specific applications, as well as a comprehensive data set.

The way to access such a dataset was to own a large social network that collects information about the consumption patterns, leisure activities and communication patterns of its users, including their social interactions and political preferences.

Such collections about the behaviour of the rest of humanity are popular in your circles, aren't they?

By buying Twitter stock, you can give your undoubtedly fine AI professionals access to a valuable treasure trove of data and establish yourself as one of Silicon Valley's leading AI players.

Congratulations on your stock purchase and I hope my data is in good hands with you.

Speaking of your professionals, I'm interested to know why your employees have to work so hard when you are so concerned about the well-being of people?

I'm also surprised that after the pandemic your staff were no longer allowed to work in their home offices. Is working at home also detrimental to the well-being of humanity?

In the meantime, you have taken the Twitter platform off the stock market.

It was never about money for you, right? No, you're not like that. I believe you!

But maybe it was about data? These are often referred to as the "oil of our time". The data of a social network is like the ticket to be one of the most important AI developers in the AI market of the future.

At this point, I would like to thank you for releasing parts of Twitter's code for algorithmic timeline control as open source. Thanks to this transparency, I now also know that the Twitter algorithm has a preference for your Elon Musk posts. What an enrichment of my knowledge horizon!

And now, barely a year later, this is happening: OpenAi, a hitherto comparatively small company in which you have only been active as a donor and advisor since your exit in 2018, not only has enormous sources of money, but also the AI gamechanger par excellence - Chat GPT. And virtually overnight becomes one of the most important players in the race for the digital future. It was rumoured that your exit at the time was with the intention that they would take over the business? Is that true at all?

After all I have said, I am sure you understand why I have these questions for you, don't you?

I would like to know what a successful future looks like in your opinion? I'm afraid I'm not one of those people who can afford a $100,000 ticket to join you in colonising Mars. I will probably stay on Earth.

So far I have heard little, actually nothing, about your investments in climate projects and the preservation of the Earth.

That is why I ask you, as an advocate of all humanity, to work for the preservation of the Earth - with all the means at your disposal, that would certainly help.

If you don't want to do that, I would very much appreciate it if you would simply stop worrying about us, the rest of humanity. Perhaps we can manage to protect the world from marauding robots and a powerful artificial intelligence without you, your ambitions and your friends?

I have always been interested in people. That's why I studied social sciences and why today I ask people what they long for. Maybe I'm naive, but I think it's a good idea to ask the people themselves what they want before advocating for them.

The rest of the world - that is, the 99,9 percent - who are not billionaires like you, also have visions!

With the respect you deserve,

Susanne Gold

(just one of the remaining 99% percent whose welfare you care about).

#elon musk #open letter #artificial intelligence #chatgpt #science #society #democracy #climate breakdown #space #planet earth #siliconvalley #genocide #war and peace #ai algorithms

[betterbemeta]

Something I am noticing lately on our current 'content and engagement-driven' internet is the mass platforming of stuff that 'sounds like it makes sense to a human being' as its only basis. It resembles 'a real thing' that people can hear of, but it's served completely blind to knowledge of the topic at hand.

It's not quite misinformation because it doesn't have to be an intentional trick or fringe conspiracy theory, or 'fake news' style progpaganda, or something with any kind of comprehensible 'goal.' And its often fully automated: it's about feedback loops in the selection of content, not 'content creation.'

So 'accurate' information on how and how often people select A, B, or C... but no information about what A, B, or C are or any factual information about them. Sort of like how 'AI art' will return output based on prompts, vast amounts of indexed 'reference' and trained data, but doesn't know 'what it's looking at.'

The result is similar to 'enshittification' but doesn't have to do with gutting usability to deliver returns to investors. It has to do with automated systems self-gutting a platform's usability because they do not actually 'know' anything, they just evaluate the navigation patterns of prior users.

This is a really simple, innocent, non-malicious starting point:

This is an audio-based youtube video that was made by a human being (I assume) who mixed Blue Noise and Violet Noise with the sounds of actual blueberries being shaken in a box for relaxation purposes. There's nothing inherently wrong with it. Please don't find and bother the blueberry noise person, nothing (that I know of) is their fault.

The thing is, 'white noise/pink noise/brown noise' are specific audio patterns that are sometimes studied for their benefits on concentration, sleep quality, etc. Which is why people are searching for them on the internet, they want practical benefits. There are other 'color' noises but any benefits are even less well-known, or their application for study is very specific (Violet Noise is studied for tinnitus relief, I believe.)

While the blueberry noise video is just for fun, for the benefit of anyone who enjoys it, it exists within a 'topic' that a platform like Youtube has matched to be 'about' some practical benefit. But you don't need to actually know anything about Pink Noise or Brown Noise to get to that topic, you just need to search 'concentration noise.'

You could plonk down a 10 hour video of your own farts and title it "rainbow noise for homework focus!" and the algorithm cannot 'know' that no part of what you posted is "a real thing." It would only see the activity metrics, who got to it from where, and how popular you are, and its potential to keep users engaging with the platform.

People navigate 'white noise/pink noise/brown noise' for ADHD relief, to soothe their babies to sleep, or to help with chronic insomnia or headaches. Even if studies on the effectiveness of 'well known' color noises are still only preliminary (this study only had 22 participants!), there is some basis for their benefit... but no evidence yet that supports the benefit in mixing different color noise patterns, like the blueberry noise video. It's just for fun. The issue is that Youtube doesn't know what 'just for fun' is.

If you listen to enough audio pattern content to encounter our blueberry noise friend you probably are going to run into 'Alpha/Beta/Gamma Frequencies', 'Solfeggio Tones', and 'Binaural Beats.' But, again, the algorithm has no real human knowledge: it doesn't 'know' the difference between any of these things or what they're studied for, or if any real benefit has been confirmed. But content reflecting machine curation's inability to understand will float to the top, based on user activity patterns:

Alpha waves have nothing to do with being an alpha chad. But through association without comprehension, platforms have 'learned' what 'alpha' means and as a result content will appear and be promoted that bridges multiple ways people use 'alpha' as a search term.

High frequencies produced by the human brain have been associated with memory recall and learning, but listening to high frequencies as 'brain waves' has nothing to do with this, and especially not a dystopian drive to 'increase productivity and output.' But people are searching for how to do that, because of the demand their lives place on them.

There is no sound frequency you can listen to that heals the body or reduces inflammation but due to this principle I'm describing that's messing with the internet, you can search this on Google and it will now not give you any credible information on the first page, it will assume that what you're talking about is legitimate and show you results that 'you're looking for':

It will not assume you would be interested in a 'no' or 'this is not real.' None of the Page 1 Google Search Results pictured above are based in any scientific fact. Maybe if you're a stressed out zebrafish. Legitimate medical practices that use frequencies are like... shock wave lithotripsy for kidney stones to break them apart. But once a web platform records people searching for audio content that confers dramatic 'real' results, it will retrieve other kinds of content as if it's credible:

I think it goes without saying that this has no basis in fact and it's not just somebody's fun entertainment project, and it's not something a typical person would seek out or casually believe to be true.

Anyway. The blueberry noise is fun and fine. But the current internet is on a form of evil autopilot that can't discern white noise intended to soothe babies from the innocent only-for-fun blueberry noise from The Law of Attraction. It's like the perfect totally-blind robot salesperson without ethics or morals of any kind. If you ask it for medicine it will eventually sell you cyanide pills simply because they are pills and so many people are interested in cyanide these days.

[disruptivevoib]

Could you tell us about AI Mimb...

--🍷🧡 {Soul front 🔱}

EDGAR!!

Or rather, [Mod]eratorbot. While there are certain things that I can't say because they're not revealed to folks besides those of us in the AU itself, his Whole is a streamer!

@shadywoods plays the aforementioned Whole, named Sneeb.

@agent-8449 @shxwrunner are Soul and Heart respectfully.

Edgar, Allen, and Poe or Mind, Soul, and Heart are all bots made by Sneeb and utilized on his streams as part of the entertainment!

They are also, in chatroomverse canon, the creators of the chatroom through the usage of an Algorithm they created to interdimensionally sort and search for timelines or universes where instances of themselves or whole existed.

And then they lost control of that algorithm.

Edgar himself was insistent on remaining non-sentient. He is not alive and thus shouldn't be much more than code. Poe, or Heart wound up deleting a chunk of his code and in the aftermath, Edgar.. well.

Computer's aren't made to effectively sort through literal human thought and logic. Which means he is in a near perpetual state of having a "headache" of sorts. Though it has now been midigated.

He wants, most times, to do what is best or right for his Heart and Soul. Though Allen left them, in some twisted attempt to try and make things better (long story. Tldr. Allen knows things the other two do not.)

Edgar now fairly staunchly despises Soul (and missed him). And while he and Heart get along, he's still prone to insensitivity, even when trying to be helpful. Suppose sometimes your emotions need to be spoken to harshly.

Overall, he wants to do whats best and wants to keep Whole and Heart both with him and from spiraling further than either of them already has. He's got a knack for being a bit nosy and fairly petty at times too.

#typical mind but he really does care a lot just kind of distantly. #hard to concise him down into smth small with no precontext #chonny jash #voidthoughts #asks #ai au #ai edgar #ai mind

[prettymuscles]

I'm taking requests!

Do you have a male model / adult star / actor you would like to see transformed in the style of my blog?

Feel free to "ask" me or message with one requested picture per week or the name of a desired model and I'll see what can be done!

A few tips:

Do not send: Pictures with head turned far to the side, looking over shoulder (may work if toward camera), looking drastically up, down or away from camera; these usually generate bad results due to the algorithm (hair in places it shouldn't be, face not really feminized, etc.)

Best pictures to use are high quality, well-lit front-facing modeling or selfie-type pictures, similar to the ones below

No AI generated pictures please, try to keep it to actually existing people

Beards can make things complicated, but not impossible: if the picture is high quality enough, and the face is front-facing, faceapp can usually remove the beard without issue, although it does have trouble differentiating beards from skin on darker skin tones (which is a shame, because many non-white men look beautiful when the filter is applied!)

Low-quality / non-HD quality pictures are extremely difficult to make work; I recommend the picture be sized at least 700 x 700, but ideally higher than 1000 x 1000 for best results

The absolute worst pictures, unfortunately, are bearded men looking far to the side, down, or up - faceapp cannot deal with a confluence of several issues relating to the visibility of the face

If your submitted picture doesn't work for whatever reason, I may try to find another picture of the model / actor in question and use that one instead, but I can't make promises if there aren't many pictures of the person in question available.

Please see these two examples for how to get some of the best results:

#biceps #bodybuilder #bodybuilding #female muscle #fit girls #hot male #jock #men #muscles #muscular #pansexual #pecs #pretty #queer #sexy content #photoshop #digital art #male model #male beauty

[tired-fandom-ndn]

I hope you don’t mind me asking, but I’m curious about what your take is on the claim that ao3 promotes/doesn’t properly regulate racist content. What are your thoughts on that?

Well, to start, the claim that ao3 promotes anything is false. There is no algorithm on ao3; everything is found and filtered by the users, not recommended or promoted. So I tend to see any claims about ao3 "promoting" fics as ignorant of the site's functions at best, deliberately misleading at worst.

In regards to regulation/moderation. . . who is the one deciding what fics are racist and should be removed?

Is it fics where a person of color is assaulted or killed in the story? What about ones that use racial slurs? Would a fic that plays into racist beliefs about a culture or ethnicity be removed and if not, why are those acceptable but other forms of racism aren't? Would fics about historical events be removed? What about general explorations of racism? Stories about racialized violence and hate crimes that are darker than other exploration fics? Where do depictions of racism in a fantasy or scifi setting fit into this?

Would the author be consulted for their opinion and the context over the accusations of racism? Would it make a difference if the author is a person of color? Would they have to prove that they're "allowed" to explore racism in their works?

And again, who is deciding this? Because people of color and other non-white people aren't a monolith and for every single person saying that a trope or story or whatever is racist, there's another person who thinks it's totally fine. People in the same cultures and communities can't even agree on what counts as appropriation when it comes to the use of our sacred cultural artifacts, why would I trust any tiny number of people to decide which fics deserve to exist and which don't? And that's not even getting into how there's so many kinds of racism and so many groups people can and are racist to, we would end up having to depend on people making decisions about what qualifies as racist for communities that they absolutely are not part of and have no authority over.

In terms of things like just adding an archive warnings tag for racism, the only actual purpose that would have is maybe making an author think about whether something in their fic qualifies as a depiction of racism. Otherwise, we have plenty of tags relating to racism that authors already use.

At the end of the day, racism is not going to be solved or curtailed by censoring fucking fanfiction. Racism on ao3 is a reflection of racism irl and the only way to combat it is by fighting racism in our everyday lives, including in fandoms. Fiction will continue to be racist as long as there's racism in society and fanfiction isn't the exception to that.

#Anonymous #ndn tag

[starberry-cupcake]

Not to sound like a petty millennial but one of the things that bothers me the most (as someone who's forced to attempt social media to promote work) is that somehow it's become normalized to share more of your personal information than should be necessary to even exist in the realm of algorithm, versus showing your work.

The way in which video content became the most favored way of posting online (youtube, instagram reels, tiktok) has made it mandatory for people to have to share their face and/or voice in order to be picked up by algorithms, because text-only videos don't do well and non-personal sounding ones are signaled as potential content farms (or ai now I guess).

This means that a) you have to be comfortable with sharing your face and/or voice openly, b) you have to be comfortable in camera or recording and c) you have to be an engaging 'personality' who looks or sounds as 'camera ready' as possible to keep people's attention.

I understand if I'm not successful at reaching people because my work is crap but not doing it because I am forced to share more of my personal self than I'm comfortable with or because I'm not a charismatic youtuber personality is a black-mirror-level of wild to me.

[insecateur]

Hey Mr. Sam! I'm not super acquainted with the fanfic community, so I'd like to ask a question in order to put some thoughts in order. I've seen this recent phenomenon where some readers will try feeding abandoned/hiatus'd fics into some sort of chaptgptesque thingum in order to finish them, and writers are, naturally, upset. My question then is: are they/you *largely* upset because a) said readers are basically saying the writing is boilerplate and predictable enough that it can be ML'd into a coherent continuation; b) there is a legitimate fear of how the data is used or the program is trained on it; c) there is fear that their unique style will be successfully "replicated" by AI; or d) some other reason? Thanks for the info!

heyyy sorry for not answering right away,

i want to give a caveat before i say anything else - i'm markedly more jaded than most fic writers i've seen on this topic, as a primarily visual artist who's used to my work being reposted, reused, sometimes even edited etc without permission. i haven't even bothered to edit my fics on AO3 to make them available to registered users only bc i'm just like, it is how it is. if machine learning algorithms want to put my ott kinky smut in their research data you know what, good for them i guess. maybe it'll make the results less biased toward vanilla straight people lol

but anyway, i think in general it's a mix of not wanting your data to be put into machine learning without your permission or knowledge, and just an overall frustration at how people who read fanfic, look at fanart etc online treat people who make that stuff like they're basically non-existant or just there to provide "content" to be "consumed." which i guess is part of why i'm already way past jaded about it; people have already been doing this to fanartists for years and years and years at this point LOL. i'm actually myself kind of frustrated seeing some writers talk about it as if it's a new phenomenon and proof that writers are treated less well than visual artists in fandom ? when ummm. i guess you just didn't notice until now bc it wasn't threatening you? which is fine but it gets kinda old after a while.

i personally just don't like people doing stuff with what i make without permission, as much as it is a losing battle at this point. so seeing people joyfully basically say like, "i don't care about the people making the things i like, i can just get this ''AI' to do it faster anyway!" is pretty disheartening, especially when it involves feeding works you might have spent months or years on into it. it's kind of an accumulation of negative feelings.

in the grand scheme of things i'm mostly over it tho. to me it's pointless to worry about it just like it's pointless to worry about the people disregarding my wishes regarding my art. all it does is cause me undue anxiety. which isn't to say measures shouldn't be taken against it, and i'm glad there have been moves recently to try and draft laws etc on the subject, but at my level there's not much i can do except say "please don't do that" so really it's whatever.

#la réponse d #idk if this is an interesting answer but there u go. if other people want to weight in feel free

[cambion-companion]

Not to join in on the anon-wars, but Ao3 has always operated very much on the “Dead Dove Do Not Eat” principle.

And there are things on that platform that are downright hellish, but it’s free, and if the tags are making you uncomfortable do not read; if the story upsets you, stop reading it; if you want to read stories more loyal to canon and you see a lack of such fanfics, write your own.

In any fandom and on any platform you’re bound to find people you disagree with both in interpretation of characters and the ethics of writing incest, non-con, violence, age gaps, and so on.

Half of the people writing Aemond as a horrible person probably don’t even care if the masochist they end up portraying him as fits canon. They’re just in to things like that.

And the writers and readers who enjoy seeing aemond turn black for Rhaenyra’s daughter/“y/n” might be team black, or they might be a sucker for a good old Romeo and Juliet trope, or they might like the idea of a person going “f*ck it all, they’re mine, and I’ll do anything for them.”

Bottom line is navigating in any fandom with the idea, that the content should align with your morals and your perception of characters, will only leave you endlessly disappointed.

I’m not personally fond of the lucemond-ship for obvious reasons, but the people that do ship it, obviously won’t care. Yes, I’m allowed to be disgusted and I could spend my time screaming in to the void, but I could also go about my day, not interact with it and let the algorithms do their jobs, so my feed remains free of it.

It’s free to read and to engage with, and I’m also free to ignore it’s existence.

🌼

Agreed! This is the point I was trying to convey with my last answer hehe 😅 thanks for putting it so elegantly! The same goes for real life as well. I've had friends cut me off because my "world view" doesn't align exactly with their own. That's no way to live life...sounds stressful tbh.

#aemond targaryen #house of the dragon #answered ask #aemond one eye #aemond stannies #aemond fanfic #aemond x reader #aemond imagine #game of thrones #hotd

[liz-allyn]

how can I get my fics to get attention like yours do? working your ass off and then getting 20 notes sucks. and I’m trying to become a better writer and honestly crave the validation like you get from your followers. hopefully you’ll see this! Obsessed with S&V

Hi my fellow 🖋️!

Thank you so much for your comment, and thank you for thinking so highly of me!

answers for fic writers after the cut

So honestly? I don't know how to get followers. This blog has been running since 2021, and I only started writing fic < 1 year ago. I've become so much more well known (ha) in the TASM fandom because I've hung out here a long time. I don't have a secret to engagement or understand how tumblr 'works' that well (ie the algorithm that social media sites use to decide what to put in front of you). There are still people I freakin' worship on this site who don't interact with my fics at all, haha. AND flip side, some of the best writers I've ever seen are on here and their fics have less than 100 notes. I have learned that persistently I'll look over into other peoples' yards and be like "fuck. how did they get 4K followers?" or like "wow, she just posted this yesterday and it has over 1k notes."

as for working your ass off and getting 20 notes—CAN CONFIRM: that sucks. some fics I wrote for fun. some fics I wrote for the creative challenge. but pretty much on all of them I've left a little piece of me, of my soul and the lessons i've learned in my meandering partial journey. blood on the keyboard. and those fics? the really great ones (or ones just as good as sugar and vice)? they still don't have many notes.

But I will acknowledge that my following dramatically increased when I started posting S&V. I think I had ~900 and that blew up to 1.5k. There are a few factors in this:

Always post fics with fic art (with gifs! if possible) - This is the number one thing that draws me to read fic. I like art that looks cool. I also have been playing around with Canva Pro for 3-4 years (it's my fav) and I'm animation-adjacent as a day job, so pushing myself to think about graphic design has been really good for me. I find it fun to come up with a whole aesthetic for your story, whether it's a mood board or a color scheme (I usually have 'branding' on the brain). If you don't want to play with fic art, at least find an appropriate gif. Use the actor's face. I also think the gif search on Tumblr sucks balls but it's really worth it to have the right gif.

Pay attention to the hashtags of other popular fics in your fandom - when I started paying attention to notes, this was the best advice. I was using all kinds of tags that made sense to me, but weren't effective. In the web version of Tumblr, I can search for a tag and see how many followers it has. It's significant if #peter parker x reader has way more followers than #tasm!peter x reader. Also trends come and go, tumblr has their hotboy of the month and it rotates. there have been months were TASM Peter was non-existent in the last year, and months where we're getting *fed*. Pro tip with tags: I have a separate Google Doc with a list of hashtags that I copy and paste to save time. After a while, you should try reblogging with different hashtags

Use a spelling or grammar checker - I'm not a grammar fac*st but if I open a fic and everything is misspelled, almost no punctuation is used, and I have to work really hard to understand your story, then it's not going to go far. (With respect to writers who post work in English, and it's their second language). I never pay much attention to the occasional misspelling or grammar mistake, but when it looks like you sent your fic via text message or tinder chat, I just can't. I personally use Grammarly, but I also hate it and think it's buggy and overpriced. If anyone else has a better option that works with Google Docs, let me know!

Never write your fics in Tumblr's post editor. That has nothing to do with engagement but that shit is buggy as hell.

Avoid putting too much text up front before the fic. I put author's notes at the end. I try to keep summaries short. I try to focus on what someone absolutely needs to know to understand this fic, and I *mostly* try to avoid apologizing to the reader ahead of time for what they're about to read, for it being too long, for it being weird, bleh bleh bleh

If you're writing a series, make people reblog to be tagged. I've seen lots of people do taglists and then they give up because it's a pain to manage (and it is), plus you have a bunch of people that go "add me to the taglist!" or "next part, when?" and they didn't even reblog what you've already written. I think that's been one of the really big factors behind S&V's success

Make navigation as easy as possible - this is web design 101 (which I flunked repeatedly), but a masterlist is good to have. If you're in a series, a link back to the previous chapter + next chapter is good to have, I've also recently started putting a link back to the masterlist. people that know more things about stuff than me told me that you want to make things as simple as possible for viewers to get from one page to another. (that's why youtube links are shit on tumblr, even videos are meh, but GIFs work so well and attract so much attention). this is a visual site, so big images or headings or links with color help to draw the eye's attention.

Write what you want to write. Because you really need to gain enjoyment from writing, and it should never feel like you're a content factory. Do this 9 out of 10 times. The 10th time, write what people want to read. My first mob fic was These Violet Delights which I thought was so smart, and I was so proud of, and I planned this whole series out for. But also I tag things accurately, and some people (even my moots) are uncomfortable reading about those topics. Then when I started S&V I was extremely self-critical because I didn't want to write something filled with tropes and cliches. And then I sort of got off my soapbox and let S&V be its own thing, full of tropes and cliches, because they're popular for a reason, and they're fun to read, and it's okay to have candy for dinner every once in a while. And I've been able to pivot my plans and create something I'm pretty proud of.

It does help to stay consistent in the fandom you're writing for. Either by writing or reblogging other writers. Just don't make yourself feel like you're a slave to Tumblr and you must stay active at all times f o r e n g a g e m e n t. Everyone should take breaks when they need to.

Don't listen to my advice because no one knows how to hack tumblr. There are some posts that have sooooo many notes and I'm like 'why?' Meanwhile, my heart's been ripped open in LED pixels on screen and sometimes it passes by.

I hope that this was useful! And if not, I'm sorry for the long post. I look forward to reading your fic (whoever you are), and feel free to tag me when you write your next thing!

💜

#on writing #writing advice #fanfiction #fanfic advice #peter parker fanfiction #tasm fanfiction #writing #writers on tumblr

[tacticalhimbo]

CS/CS/HB 3: Online Protections for Minors - Florida's New Surveillance Law

Now, I know many of you are looking at the title and raising defenses over my framing. But Jason, we care about minors! They should be safe online, don't you agree?

Absolutely I do.

However, this law will do nothing to actually make the online market "safer" for children and teens who use social media. It does not improve upon algorithms to make them less predatory. It does not instate safe, communicative spaces for online users, nor does it give them tools to avoid hate, harassment, and filter their experience.

It simply requires individuals to upload their personal, government identifications to verify their age, under the vague guise of "well, we told social media teams they have to delete the information within a certain time frame, and they legally can't sell the data." It erases anonymity for minors. Ties their online personas to their real identities, and makes it easier for bad actors to access this information.

But that aside, how else does it not protect youth? Well, it does not allow them access to queer spaces, as the State of Florida (and those representatives within it) have passed law after law sexualizing and dehumanizing the queer community. They have said that the mere act of "drag", watered down from its intrinsic art form to "dressing non-traditionally for one's gender", is considered sexual. Inappropriate for the public. They have said that simply existing within public spaces, as a trans* individual, is a crime. That daring to use the bathroom aligning with your presentation is the near equivalent of joining a criminal organization and kidnapping/sexually exploiting a child with the intent to traffic.

It uses the same, vague language as KOSA, the bill that, if signed and enacted, would require this type of personal data collection on the federal level.

Now, the ideal is to have a less predatory environment online, but we just cannot do that under capitalism. And any restrictions would be invasions of privacy by either 1) requiring an ID upload or 2) restricting account functions and monitoring activity through a probationary period (which is usually an overtly long period on purpose to encourage the former option).

Also, it's becoming funnier bc they're like "Nobody under 14"… most social medias have a 13+ policy, and you can't get a government ID in the state as a minor unless you have a parent present (I would know; I received a state ID at 16 to get a tattoo).

One thing I'm also looking at with this is the emphasis on parent permission and involvement. Are legal guardians going to have to upload their IDs to prove they're giving their child permission to use a site? Because it also says that parents can circumvent the child's account ownership and request the account be taken down if they feel they haven't given their permission for exposure to certain topics, but how do they know it's the parent? Is it an honors' system with an email address, or something more?

And what is harmful? Like I said, sexual content, sure, but also anything the state does not like. Example: Queer folk.

It's not explicit, but point 1 and point 2 fit the Florida view on queer folk.

"The average person applying contemporary community standards would find appeals to the prurient interest" (anything against the widely accepted social norm)

"Depicts or describes, in a patently offensive way, sexual conduct" (doing drag, again, is considered lewd and laviscious, queer art is considered lewd and laviscious, etc etc).

But it doesn't stop there. Any non-US social medias are exempt from this. You simply lose access to them if you cannot verify your age, because these laws do not apply to non-US entities:

And the terms for punishment, which I do not have screencapped at the moment, are vague. The bill says the responsibility falls upon the corporation, but what happens then? Are corporations obliged to report any users violating this law to the State? Where do internet service providers (ISPs) fall into this?

This law is not for kids; it's the testing ground for total media suppression. It will be enforced that way, just as various other bills in Florida have been, and it will become the new national norm. It will do so under a Republican presidency. It will do so under a Democratic presidency.

Florida is just ground zero, babes.

#txt #important #florida politics #long post #idk what else to tag this #and i could have gone more in depth but i just... i live this shit every day. i am not strong enough to be *the* beacon #if i find any better articles about it i will amend the post and post them too

[televisefur]

CW: Non-art tangent, Mentions of Twitter, mentions of hate speech

My opinion on Twitter and how we should use social media in 2023

I want to make this space a blog where I can express myself freely - so some posts will be art, and some will be like this. Thank you for staying with me!

Now, onto the topic of today's blog post.

Remember how we had tons of different social media websites, like MySpace, Facebook, Twitter, etc.? And now, most of the younger folk stay on Twitter, now known as X, and rarely ever go to Facebook, not to mention what happened to MySpace.

As part of this centralization, from 2012 and up until 2023 Twitter used to be my home online. Most of my followers are on Twitter and most of the people I know from the communities I'm in are from Twitter. I liked the convenience of having one platform where I could post art, talk to people, share my opinions.

As you know, Twitter, now X, has been irreparably damaged. As time went on, even before Elon's takeover, it became impossible to share your opinions without being supressed by the Twitter algorithm. I should've known I had Tumblr as an option where I could do so. Well, now as I've realized this, I'm making this post to talk about it. But things are so much worse now for Twitter when it comes to expressing your opinions or finding out what people think about things - which was Twitter's original intention.

When Elon came over, things became even worse. Elon started promoting alt-right hate speech, and made it much harder to get noticed or even talk to people in DMs without paying for Premium. But unlike Reddit, there was no giant tidal wave of people moving to Mastodon or any of the other alternatives (and people came back to Reddit anyway because some communities just stayed there or got reopened, and Lemmy has a host of issues that make it inconvenient for the majority of users).

Instead, the community splintered - Bluesky invites are flying off the shelves, Meta had their euphoric moment with Threads, and even Tumblr is trying to attract Twitter refugees by replicating its user interface, but most people stayed on the bird-turned-letter app- meaning we as artists or fans have to stay on our old Twitter accounts while at the same time spreading our attention thin between different platforms.

Tumblr used to be this extremely niche platform for artists, fans, various communities but it is becoming more popualr as a general social blogging website. I think that's a good thing, because I don't believe in gatekeeping, but I also don't like how uncomfortable existing Tumblr users are right now with the fact that a ton of Twitter users are coming into the platform.

Basically, I should've "switched" to Tumblr a long time ago and made that my priority. I should've just left Twitter a long time ago, when it started becoming the "only positive vibes" platform. The people behind Twitter manipulated it and the content you see on it for years, and that is the thing I dislike the most, second to the far-right push itself.

But now, the only thing I can do is get used to a splintered experience of social media once again. And I kind-of have a method for it, but I'm figuring it out. Like, for example, why should I use both Mastodon and Bluesky if both are virtually the same except one is open for everybody but more niche, and the other one is locked but is actually trending? Maybe it depends on which one is more popular at the current moment. Maybe it's based on how fun it is to use the platform. Maybe it is based on what I can do on it.

The thing is - I'm in the exploring phase. And that's the fun part.

#twitter #tumblr #social media #blog #opinion

[sol-consort]

Well now that it seems like you’ve gotten past Legion’s sacrifice I can say the part I left out of Legion’s pronoun ask

(If you haven’t seen the scene where they speak about what Legion says before sacrificing themself stop reading this ask)

Using singular pronouns like he/him are also ok because he refers to himself as I before sacrificing himself. Although the Geth most likely have like no gender so they/them is probably still the most accurate but I don’t think the Geth would care if you called them he/him or she/her

ALL GETH NON-GENDER CONFORMING LET'S GOOO!!! NEVER SEEN A RACE WITH SO MUCH RIZZ <33

And god it was a punch to the gut- how did I grow more and more in love with the geth?

Maybe when the game was released, this mission and A.I seemed like futuristic cyber fantasy but with the current times, it's growing more and more closer for comfort. Like I have how many years left to live, 60? And A.I is evolving, I might even get to witness the first prototype of an actual artificial intelligence that's not just a predication and logistics algorithm.

And that mission where you connect to the geth network and get to see their memories, see them being cared for by their creators. God that one geth asking what they have done wrong when they were being shut down really hit deep in my heart, like a child asking their mom why she's leaving, did they do something wrong? Can they fix it? Please don't leave me.

The fear of A.I replacing organic life has been one for debating since the day humans made the first ever computer. And it is reality, this is a real life issue the future will face. It's not a matter of if we create semi-sentient machines, it's a matter of when. People are trying and they will keep trying.

There is even this famous qoute:

"Man becomes, as it were, the sex organs of the machine world, as the bee of the plant world, enabling it to fecundate and to evolve ever new forms. The machine world reciprocates man's love by expediting his wishes and desires, namely, in providing him with wealth."

By Marshall Mcluhan.

Because we are the reproductive organs of the machines, the ones who will create them. The machines need organics to reach existence, they're simply not something that happens by nature.

And look at our machines so far, we've created them based on our own image. Sent them to space and names them curiosity, preservence, spirit and opportunity. These are real names for the NASA mars rovers.

We gave taught them how to sing happy birthday for themselves, all alone in space because we imagined how alone it must be, we wanted to comfort the metal that could reach worlds beyond our body's could ever handle.

But we're also scared, we're flawed and know that we are flawed. We are scared of them not needing us anymore, realising how weak and bad we can get, how we don't have a purpose in this world. We are scared they will not love us back, that they will leave us to rot in the wasteland we've turned the earth into instead of being the ones to lift us up into newer planets.

And the geth mission just made me sad man. It's like a parent shunning the child that loved them, and the child simply doesn't understand why their parents don't like them. They're just so innocent and their world view is so pure that they come to the conclusion any child would.

That they must've done something wrong, for how could the ones who gifted you life be the ones to forcefully take it away, look at you as if you were a monster when once you were beautiful in their eyes.

Ik it's a game but I really hope by the time a breakthrough in technology gets made that we can replicate even an ant's mind artificially, then we would be kinder parents to them than the quarians ever were to the geth. I hope humanity matures enough by that time.

10000/10 writing, hand in marriage for whoever wrote that Geth mission conclusion. It actually ranks even over Tuchanka for me because of how grey and muddy it is. It fully depends on your own views as a person and if you think organic life is superior. There is no clear right or wrong. For once it is the human playing the game who has to ask themselves what they believe in.

[pearls-and-vignettes]

A pearl. Bright blue in coloration. It details a conversation betweeen two parties.

1002.321 - PRIVATE Markings of Ash, Ground Coral into Lines

MA: Good thing they never revoked my access to the registrars'. MA: Here, the last entry on the ascension log: MA: "One-Spoked Wheel, Non-Associate, Subordinate of the Feathered Wing House of Void Studies; Self-ascended, Gilded Mask Superstructure Void Site 07; Notes: 'Fuck all of you.'" GCL: Poor little guy... Didn't even have a home AND they left him behind? MA: That is everyone. GCL: What MA: All of our residents have either left or ascended. I have sent a few inquiries to other groups' seniors and their logs further characterize this truth. We are effectively alone GCL: Wait so like GCL: We can party? GCL: Like do whatever we want? MA: Essentially, yes. There are no more engineers to monitor our actions, no more councils to decide what taboos we can and cannot have. We no longer have to police our own speech, thank goodness for the 1% extra processing capacity that frees up. GCL: One percent more than any of them could have processed individually. GCL: Or even collectively for that matter! MA: Look, the current moment isn't the time for commiseration. We resented them, and most of them resented us, I understand. But we each need to find a new directive. GCL: I thought that we could only focus on solving the Great Problem GCL: Isn't that what we're both doing? Right now? MA: I... MA: I may have developed a methodology over a long enough timeline that any analyst rummaging through my output would account for my 'erroneous' calculations with even a relatively small margin of error. MA: I can write onto Pearls recursive, open space, alongside an algorithm that calls for the copying of existing data onto the Pearl, as well as its simultaneous deletion. This process results in the deletion of both the source and the copy, as our cells do not have enough time to designate the copied information as such before the deletion order is sent. MA: At its mildest, this somewhat scrambles the affected data, rendering it inert if there is no action to read it, and at worst, the data is completely obliterated. Like a parchment set alight. MA: A certain engineer responsible for giving the first of our kind instructions intended to repurpose their instructions for the creation of smaller, less specialized computers responsible for more menial tasks. MA: This consequently means that not all of our cells contain the relevant taboo. GCL: So if we run that Pearl fast enough... MA: We can effectively obliterate the taboo. MA: It would've been a simple fix to implement, really. MA: What a /shame/ that their attentions were just too centered on themselves. GCL: I thought you said no commiseration! MA: I am submitting one of the Pearls I've generated to you as a research sample. I am also submitting a second Pearl with more detailed instructions. Up your water intake by at least 15% when running this, or you will get strained pretty badly. GCL: How can I send this to my own neighbors? Material transmission between two Iterators isn't really a thing everybody has... MA: I do not know. Perhaps you can dictate to them what to write on their own Pearl, but there is a considerable chance for error in that, and this process can be extremely damaging if not done correctly. GCL: Oh, I know! MA: Hm? GCL: I could get the lizards to carry them to their cans! MA: Very funny. MA: Enjoy. END OF TRANSMISSION

#cat2 #rainworld #rwpc`MA #rwpc`GCL

[nuadox]

The ChatGPT chatbot is blowing people away with its writing skills. An expert explains why it’s so impressive

- By Marcel Scharth , University of Sydney , The Conversation -

We’ve all had some kind of interaction with a chatbot. It’s usually a little pop-up in the corner of a website, offering customer support – often clunky to navigate – and almost always frustratingly non-specific.

But imagine a chatbot, enhanced by artificial intelligence (AI), that can not only expertly answer your questions, but also write stories, give life advice, even compose poems and code computer programs.

It seems ChatGPT, a chatbot released last week by OpenAI, is delivering on these outcomes. It has generated much excitement, and some have gone as far as to suggest it could signal a future in which AI has dominion over human content producers.

What has ChatGPT done to herald such claims? And how might it (and its future iterations) become indispensable in our daily lives?

What can ChatGPT do?

ChatGPT builds on OpenAI’s previous text generator, GPT-3. OpenAI builds its text-generating models by using machine-learning algorithms to process vast amounts of text data, including books, news articles, Wikipedia pages and millions of websites.

By ingesting such large volumes of data, the models learn the complex patterns and structure of language and acquire the ability to interpret the desired outcome of a user’s request.

ChatGPT can build a sophisticated and abstract representation of the knowledge in the training data, which it draws on to produce outputs. This is why it writes relevant content, and doesn’t just spout grammatically correct nonsense.

While GPT-3 was designed to continue a text prompt, ChatGPT is optimised to conversationally engage, answer questions and be helpful. Here’s an example:

A screenshot from the ChatGPT interface as it explains the Turing test.

ChatGPT immediately grabbed my attention by correctly answering exam questions I’ve asked my undergraduate and postgraduate students, including questions requiring coding skills. Other academics have had similar results.

In general, it can provide genuinely informative and helpful explanations on a broad range of topics.

ChatGPT can even answer questions about philosophy.

ChatGPT is also potentially useful as a writing assistant. It does a decent job drafting text and coming up with seemingly “original” ideas.

ChatGPT can give the impression of brainstorming ‘original’ ideas.

The power of feedback

Why does ChatGPT seem so much more capable than some of its past counterparts? A lot of this probably comes down to how it was trained.

During its development ChatGPT was shown conversations between human AI trainers to demonstrate desired behaviour. Although there’s a similar model trained in this way, called InstructGPT, ChatGPT is the first popular model to use this method.

And it seems to have given it a huge leg-up. Incorporating human feedback has helped steer ChatGPT in the direction of producing more helpful responses and rejecting inappropriate requests.

ChatGPT often rejects inappropriate requests by design.

Refusing to entertain inappropriate inputs is a particularly big step towards improving the safety of AI text generators, which can otherwise produce harmful content, including bias and stereotypes, as well as fake news, spam, propaganda and false reviews.

Past text-generating models have been criticised for regurgitating gender, racial and cultural biases contained in training data. In some cases, ChatGPT successfully avoids reinforcing such stereotypes.

In many cases ChatGPT avoids reinforcing harmful stereotypes. In this list of software engineers it presents both male- and female-sounding names (albeit all are very Western).

Nevertheless, users have already found ways to evade its existing safeguards and produce biased responses.

The fact that the system often accepts requests to write fake content is further proof that it needs refinement.

Despite its safeguards, ChatGPT can still be misused.

Overcoming limitations

ChatGPT is arguably one of the most promising AI text generators, but it’s not free from errors and limitations. For instance, programming advice platform Stack Overflow temporarily banned answers by the chatbot for a lack of accuracy.

One practical problem is that ChatGPT’s knowledge is static; it doesn’t access new information in real time.

However, its interface does allow users to give feedback on the model’s performance by indicating ideal answers, and reporting harmful, false or unhelpful responses.

OpenAI intends to address existing problems by incorporating this feedback into the system. The more feedback users provide, the more likely ChatGPT will be to decline requests leading to an undesirable output.

One possible improvement could come from adding a “confidence indicator” feature based on user feedback. This tool, which could be built on top of ChatGPT, would indicate the model’s confidence in the information it provides – leaving it to the user to decide whether they use it or not. Some question-answering systems already do this.

A new tool, but not a human replacement

Despite its limitations, ChatGPT works surprisingly well for a prototype.

From a research point of view, it marks an advancement in the development and deployment of human-aligned AI systems. On the practical side, it’s already effective enough to have some everyday applications.

It could, for instance, be used as an alternative to Google. While a Google search requires you to sift through a number of websites and dig deeper yet to find the desired information, ChatGPT directly answers your question – and often does this well.

ChatGPT (left) may in some cases prove to be a better way to find quick answers than Google search.

Also, with feedback from users and a more powerful GPT-4 model coming up, ChatGPT may significantly improve in the future. As ChatGPT and other similar chatbots become more popular, they’ll likely have applications in areas such as education and customer service.

However, while ChatGPT may end up performing some tasks traditionally done by people, there’s no sign it will replace professional writers any time soon.

While they may impress us with their abilities and even their apparent creativity, AI systems remain a reflection of their training data – and do not have the same capacity for originality and critical thinking as humans do.

Marcel Scharth, Lecturer in Business Analytics, University of Sydney

This article is republished from The Conversation under a Creative Commons license. Read the original article.

--

Read Also

Large language models could help interpret clinical notes

#ai #artificial intelligence #chatbots #chatbot #chatgpt #gpt3 #gpt #openai