The Fanfic Junkyard

Where I muse over fandoms, release plot bunnies, try to organize fanfic ideas, and just random stuffs.

[unknown-de-mordor-blog]

Monster, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Sartle of Giant hill, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Lucifer (human form), Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Tiamet leaders, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Dragon, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Balrog, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Rafael Howler, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Uriel, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Dr. Magaret Garcia, Daemon Hunters, Seiuchiroh, Todono

[unknown-de-mordor-blog]

Gabriel, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Dragoon army, Daemon Hunters, Seiuchiroh Todono

[unknown-de-mordor-blog]

Michael, Daemon Hunters by Seiuchiroh Todono.

[unknown-de-mordor-blog]

Because I’m an utter trash for Seiuchiroh Todono, y’all are going to be treated with a whole bunch of posts with just my scans of Daemon Hunters. If you’re not interested, just block the guy’s name in the tag for a while. Or ignore me. I won’t be around long, so that works, too.

[unknown-de-mordor-blog]

My Turin armour headcanon

I’m temporarily breaking haitus because I remember mentioning to you guys that my headcanon for Turin’s dragon helm is in Seiuchiroh Todono’s Daemon Hunters. Guess what, I’m at my parents’ house, and they still have my copy of Daemon Hunters! So y’all are treated to a gratuitous post of dragon armours that no one asks for -- because I love Seiuchiroh Todono’s artwork, and I just need to share this with y’all. This is going to be image heavy, but here we go.

First off, let’s start with my all time favourite, the Dragoon armour in full.

There are actually two-horned and one-horned helms as far as I can tell. I’ve always imagined Turin wearing the two-horned one, but the one-horned one is also badass, and I mean...

... he looks mean, doesn’t he? (And he is mean, if you must know.) For a closer look at the two-horned version of the armour (I’m not sure what exactly is the difference -- perhaps rank or specialization), let us look at the commander of the Dragoon army.

*drools over the detail like it’s no body’s business*

And here is the back.

The red cross is referring to the in-universe church. Not sure exactly what the religion is (artistically inspired by the Roman-Catholic church for sure, although there has never been a mention of a messiah or anything of the like, so I don’t think the influence goes any further than the wardrobe).

There is also another, special version of the dragon armour in this universe worn only by one of the four ‘gods’, who also happens to be the said Commander’s adopted son. The kid is kind of an in-universe Turin who doesn’t realize his own tragedy until it basically destroys his beloved adoptive family and he goes demoniacal. So let’s just say seeing him wearing his ‘father’ symbol even after he’s gone berserk breaks my very heart.

But can I say this? I hate that helm.

Why the dome shape?! Why that black horns?! Seiuchiroh, WTH!!

Okay, that helm aside, I love that armour to bits. I mean, look at that waist line, the elegance.

Probably not practical for sword fighting at all, but who are we kidding here. We’re standing in front of a god for god’s sake, and he’s going to cut you the eff down, corseted or not.

Okay, another thing I don’t really like about the Dragoon’s getup is the sword. I mean, I get it. It’s supposed to look manufactured and impressive -- these are soldiers funded by the richest and most powerful organization in-universe after all. But Gurthang is supposed to be unique and -- in my very biased opinion -- badass. And I like a big crossguard, so it turns out that I like this baby better.

The sword, not the huge guy holding it.

I know, I know, it looks too... undead... to be forged by elves, but I think the overall look works. You just need to change the design on the crossguard to something else.

You know, like a dragon.

I mean, this one is pretty elegant, right?

(Am I over-dragoning things? Probably. Do I regret it? Not at all.)

[unknown-de-mordor-blog]

Prolonged haitus

Okay, so I have hit my limit with tumblr and its policies on nsfw and how that has affected people who are doing good work on sex, sexuality, and the female bodies. And frankly, I can’t stand it anymore. I don’t normally post anything that violates the guideline, but the guideline is clearly bullshit, and a place that takes away the freedom of expression from positive and decent people while letting the hate groups run free is not a free space for me or, in my opinion, for anyone. So as much as I love the people here and as much as the Purge seems to have brought us closer together, I have to go. I thought I might go to another platform, but I’m not even thinking about that anymore. Right now I just need to go because I no longer want to give them any business. That being said, this blog will still be up because there is no way to lock the url down so bots can’t use it. I mean, at this point the bots are even taking existing blogs so clearly this platform is too broken to even deal with its actual issues. But I’ll leave this post as a reminder of why this place sucks, and a thank you to the users who have made this year a good one.

UdeM out.

[kingeomerofrohan]

Is the Tolkien fandom dead?

Reblog if you are a fan!

[unknown-de-mordor-blog]

I don’t contribute anymore, but yes?

[lauralot89]

Once, I encountered the funny story of an AI image descriptor with a sheep obsession. It had been trained on pictures of fields of sheep. Therefore, it tagged anything in a field as 'sheep', including an empty field, because they work on statistical probability. Therefore, it thinks "ah, a field! there's probably a sheep here." (It's a bit more complicated but basically that.) It also couldn't recognise sheep in places that weren't fields, such as petrol stations or barns. [cont]

Now, the alarming aspect of this story is that the very same technology is probably what tumblr is using to identify porn. Now, if it can’t tell that an empty field is not, in fact, full of sheep, what hope do we have that it can’t tell an empty room isn’t full of writing human forms engaged in passionate coitus?

this really does sound like an episode of black mirror

[architeuthism]

But wait, it’s even weirder than that!

[retroactivebakeries]

This is gonna produce some absolutely baffling pornography.

[crazy-pages]

…. oh my fucking god they actually are using open source software. They’re using a fucking one-layer unidirectional bicategory tag-trained neural network. This will never work. Literally, it will never work. There’s just not enough algorithmic complexity to do what they’re asking of it. I bet you I could prove on a mathematical level that this joke of a neural net fundamentally lacks the abstraction necessary to do its job.

This will never get better. Their algorithm will never stop fucking up, it will never actually flag porn reliably and it will always require a massive quantity of human hours to deal with the deluge of mistagged pictures. This isn’t just a case of an insufficiently trained algorithm, it’s just … this is the most basic neural network you can make. It probably hasa a lot of neurons and has loads of training data but like … you can’t just brute force this kind of stuff. One layer of neurons is just Not Enough.

[crazy-pages]

Also, just to make this clear, Tumblr lied. I mean, we already know this, but I mean they liiiieeeeed. All that stuff they promised about what would or would not be censored? That cannot be delivered on with a system this simple. Nude classical sculptures, political protests, male-presenting nipples (really Tumblr?), nude art outside the context of sex, all that? You cannot train a bicategory one-layer neural network to exclude those things. It cannot be done. Tumblr never intended for those things to actually be permitted, they were just lying. Because the system they have cannot actually do what they said it would and never will be able to.

Also, this kind of system is super vulnerable to counter-neural strategies. I bet you before the end of the month someone hooks up their own open source one layer bicategory neural network which puts an imperceptible (to humans) layer of patterned static over arbitrary images, and trains it by having it bot-post static-ed images to Tumblr and reinforcing based on whether the images are labeled nsfw or sfw. Seriously, within a month someone will have an input-output machine which can turn any image ‘sfw’ in Tumblr’s eyes.

This is genuinely pathetic. Like, I have real pity for whoever implemented this, because it’s clear Tumblr doesn’t actually have any engineers with any expertise with machine learning left at all and they foisted the job off on some poor bastard who has no idea what they’re doing and is going to get all kinds of flak for their (perfectly reasonable and predetermined) failure from management.

[janothar]

@crazy-pages, can you link me to the software? I’m curious about what stack they’re using, as a Data Scientist myself. When you say “open source software” I hope you’re no talking about Tensorflow or Pytorch or the like, because that’s fine…if their specific script and model are public, though…

[wetwareproblem]

It gets even worse. It will not require hundreds of man-hours to deal with the backlog.

Because the review team is another net, and I’d bet cash money it’s another copy of the same software.

That imperceptible static trick? I did basically that. Both copies got flagged by the initial censorbot.

One copy, the less justifiable one by tumblr’s stated rules, was approved. The other was not.

Someone pointed me to another blog - I forget the name - that posts the same picture every day, and was getting similar results. Inconsistency in both flags and reviews.

And that’s never going to get better either.

But the pornbots have learned to use non-sexual keywords and tag #sfw already, so there’s that.