This is opinionated rambling.

As my room of the internet is moving on from a serious plagiarism scandal, Tom Ska uploaded a humorous video setting up a “plagiarism scale”. I liked the video, but the thing that stirred me up the most is what it did not mention: Each state on the plagiarism scale has a cousin. I shall call this parallel family “relational content”.

“Relational Content” is content which is in open relation to another piece of content in the public corpus. The term is purposefully incredibly broad as to accommodate an entire scale which I would like to riff on. Let’s start at the extreme.


I think it’s accurate to describe (media-) piracy as just as much of a permanent institution as the (legal/moral/societal) laws it breaks. And despite clearly being copied, it is not plagiarism. Why not? Just ask any 15 year old who tried uploading a movie to YouTube and put “All Rights to this movie belong to [Giant Media Conglomerate]” if that stopped the video from being taken down (or rather never uploaded in the first place).

So if plagiarism hurts both owner of the original content and the audience consuming it, piracy has never made an effort to deceive. It is completely in favor of the end consumer (which also explains its popularity compared to plagiarism) as it perfectly replicates the original content. Weakening this statement a bit, it is of course also well-known that public-facing piracy, scams, and walls of advertisements go hand-in-hand. But surely almost always it is providing a cheaper price to the consumer.

“Freebooting” is the act of taking content which is already freely available in one place and putting it somewhere else. Why do that when the original content is also freely available? Because viewers are lazy, you are still providing a cheaper price by putting it on the feed of whatever hellsite they are occupying their space in. Plus content that keeps users on the site is favoured by said hellsite’s algorithm.


In current year, I would say I’ve lived through about 5 different “reaction content” scandals in different internet complexes I’ve been housed in. The story mostly goes the same:

  • Big creator develops a questionable audience with content that goes down well with the masses
  • Creator runs out of ideas/Energy/etc. but wants to keep feeding the golden goose of a fanbase
  • Creator turns to watching other content (often sent in by audience members) and “reacting” to it

The reaction is often done as one long-form take (often streamed live to maximise engagement and revenue) and later uploaded barely edited to wherever else. The scandal tends to start once one of the owners of the “reacted to” content feels like the reaction acts as the substitute for the original work (much like a free-booted copy might). What’s the difference to free-booting then?

It’s mainly a question of what additional value the reaction brings (see parallels to legal “fair use” clauses). And this is, what I might argue, not a question that can be answered for the entire genre. Tom Scott is of course much better at talking about this than me, also see the part of the video that this link is going to.

What can be said is that of course even the laziest reaction stream is not the same piece of content as the original video, but many might as well work as a substitute. An argument for piracy which just says “well the cheaper price is what audiences want” is trivially unsound under copyright, but the same argument for a “reaction” becomes subject to the details and personal opinion.

What makes the situation even more complicated is of course that the impact on the original is hard to quantify. Some people have tried to defend their work by saying that they are providing original artists with “exposure”. The reaction streamer’s true work then is not in “reacting” but in “curating”.

Limitations in technology (and rather limitations in the willingness for big tech companies to direct their users off site) created the interesting effect that original creators do not directly benefit from any person seeing the “reaction”. Ad revenue and algorithmic metrics have been redirected to the reaction streamer. This is less the work of a museum curator as it is one of a rival museum down the street.

The benefit to the original creator is then lost in the nebulous “I will tell my viewers to check your channel out” and algorithm coupling of original video and reaction. The matter is of course mostly nebulous because it matters from case to case. It is also highly correlated to how much the “reactor” agrees with the “reactees” viewpoint, creating a horrible incentives to pander to big creators.

In their most extreme form then, “reactions” purely live off the herding of big audiences. If there is a lot more additional value added, I like to distinguish the next stage of this spectrum.


Commentary in my definition is still firmly rooted in its relation to a single or small set of other pieces of content which it discusses. For illustration purposes (and often to pad runtime), parts of the discussed content may be shown, interspersed with one’s own comments.

Again, there have been many cycles of “commentary YouTube” I’ve lived through as the genre is quite evergreen and generic.

First of all, compared to reactions which one might describe as “low execution effort”, commentary typically has “low creative effort” by virtue of the fact that it can adhere to the structure of the discussed content, this is regardless of added value.

Commentary content, just like reaction content, is the perfect vehicle to participate in the algorithm goldrush of trends as something “going viral” is often seen as reason enough to talk about it. This also means that it is well-suited to surface-level creators, a few years ago regularly denoted as “lifestyle” creators.

Due to its relational nature, commentary can frequently manifest itself as a “confrontation” with another thing. This activates the well-documented bloodlust of the average internet audience and is suited to in-group/out-group mentalities (something very well-known by political pundits).

At its best, commentary is a professional drawing from years of experience using another piece of content to educate the audience. This may be in tandem with the piece of content or in spite of the discussed content. But few subject matters are well-understood after consulting a singular source.


In the proverbial interNET, we are now moving away from things hanging off a single strand but have many points of contact. The term “review” might make someone think of something like a “film review”. I would argue that the best reviews typically connect the discussed material with the entire media landscape. But if this bothers you, rather think of a “literature review” (meta-study).

A review is an effort to connect. It may discuss a central piece of content, an entire life or a cultural institution. A review therefore needs to have something akin to a research phase. This then is true curation of content. It’s also the first category which firmly cannot be done live anymore. Your pay per hour (if any) probably just decreased.

I would call the majority of Wikipedia articles “reviews”. Notably of course for Wikipedia is that its standard is linking the sources. This of course also stems from the fact that reviews can barely act as “substitutes” anymore. But importantly: They can of course still act as reasons for people to not consume the original piece of media. See e.g. bad reviews or plot summaries.

Oh no

Annoyed readers may have noticed that as the last few sections have been walking down a single line of a spectrum, they have covered an infinitesimally small amount of ground. Specifically, we’re currently quite stuck in genre.

What about memes? The lifeblood of internet culture thrives off of staying in relation to content. Memes often use screenshots or clips from content which if shown in its entirety would classify under “piracy”. Memes relate to people, society and each other. We have already been about 8 years into memes suffocating in the vomit of their own culture. Memes may be the most relational thing discussed.

Good luck classifying fan art or fan fiction. Fan fiction is so evolved that we’ve had momentous cultural objects which at one point or another were fan fiction. So who owns the characters if they were unique enough to stand on their own by just having their names changed? Are people even making points about the original work anymore?

If characters in a biopic have an important plot moment inside a popular real life fast food chain is this advertisement or an inevitable reality of said fast food chain tainting several generations of experiences?

Content is often related to other content, because it can’t help it. Our world is full of wonderful pieces of media and art which influence us.


Relational content provokes a lot of responses because it by definition creates a dynamic to other things. In the spider’s web, it can’t help but push and pull on the other strains if it wants to remain in the net. The one trillion dollar question remains: When is it bad?

Previous sections have certainly not been without judgement, but I have tried to never taint an entire category. How derivative and how original content is are in a complex relationship with lots other factors.

Consider that the original content may be harmful and a review may defuse it. Consider that a freebooted copy leads to a vile person getting more (or less!) attention. Consider a review fraught with inaccuracies trying to discredit established consensus.

The most important principle to establish in my opinion is the following: Only propaganda survives purely off of its consumption. The creators of everything else need to eat and a place to sleep. Your knowledge of a work may influence you, may influence its popularity, but any work which is only known and not supported will have a hard time having more things similar to it made.

If you make relational content and are in accordance with the original source, link it. To quote Hbomberguy, quoting Harlan Ellison, quoting Isaac Asimov: “You don’t steal from friends.”. If you think you like the content talked about, go see it. If you can, pay for it. It is a way to truly move the landscape into a direction you’ll like.

Original Content

We have arrived at the opposite end of the spectrum. Original content is what I’d describe anything which as its primary purpose brings something new to the world.

What is holding up a spider’s net? If all strains are only connected to other strains, the whole construction would fall down and ball up. There has to be something holding the entire thing up. Original content is this Atlas. It is certainly hard to quantify what is “original”, but its existence has just been proven.

One of the reasons I am so fascinated with the idea of relational content is its relatively strong proliferation in my generation. As I’ve mentioned, I mostly blame the internet, but I have not yet tried to explain why the internet has led to this.

Certainly the first factor is purely just that the internet has led to a greater mass of total media produced which has warranted more relational content (one may imagine original content as the foundation of a pyramid and relational content as the blocks above).

But still, I feel like the proportion of relational content produced and consumed has also increased. The ease of communication on the internet has brought the barrier to publish the floor. The barriers to learn creative expression have certainly also been lowered, but I am not convinced they have as far as the ability to fling ones proverbial shit into other people’s faces.

One theme in this article is one of effort and the truth that relational content is in total lower maintenance. Especially in the long run. This is very similar to why people plagiarize. If the original content is acknowledged/linked to is really secondary to just keep pumping out whatever one can.

This article was actually about Generative AI, sorry

Bringing up “artificial intelligence” may seem like a non-sequitur or a weird homage to the aforementioned hbomberguy video. It is neither. Personally I think the section of that video was its weakest part.

Generative Articial Intelligence (GAI) are language-, picture-and videomodels capable of “creating new content”. The proposition is that the content is new if it combines enough different sources in probabilistically varied ways. It does not refer back to the content used, there is no linking or source attribution.

I do not want to be caught in the weeds of the GAI debates going around the internet in this article, but what is clear is that these models have training sets. And they will continue to need training sets. I’ve heard the fear of these models cannibalising their own output quite early into the mainstream hype and I think it’s quite spot-on.

This is in my opinion a quite sensible way to re-frame the “ethics” of these models: Does using them and/or paying for them support the original pieces of content that make them so great? No. But they are creating a substitute and crowding its market.

Computers have replaced paper filing, many people have lost jobs. But the computers didn’t need to be filled with handwritten paper to function correctly. This is not your typical tech revolution.

In an internet where the barrier to post has already been lowered to hell, the barrier to create vomit is currently in free-fall and shows no sign of stopping.