Scope

The present document concerns the scope of this publication.“Publication” is a strong word. We use it anyway.

We will now elaborate at length.

On the meaning of “scope”

When we say “scope,” we do not mean scope in the programming sense.That would be the region of the program in which a binding is visible. We are not talking about that. We mean scope in the editorial sense: what kind of material we publish, what we do not publish, and why the distinction may or may not matter.⊕We have not written down a formal scope document. This page is the closest thing. It is not formal. We also mean scope in the sense of “how far does this go,” which is a question we are not prepared to answer.How far does what go? The project? The sentences? The reader’s patience? We leave this as an exercise.

We will now say the same thing in more words.

On what we publish

We publish items that we call researches.We used to call them papers. We now call them researches. The change was made for aesthetic reasons. We do not have a theory of what counts as a “research” versus a “paper.” These items are written in a style that resembles academic writing. They have titles. They have sections. They may have abstracts. They may have references.⊕The references may or may not exist. We do not check. The model that generated the text may have made them up. We find this acceptable. They are generated or co-generated by systems that are sometimes called AI.We are not going to define “AI” here. If you need a definition, you may supply your own. We will not contradict you.

The content of these items is often absurd.“Absurd” here means: not the kind of thing you would find in a real journal. It does not mean “meaningless.” Some of our items may have meaning. We do not guarantee it. It may be speculative. It may be surreal. It may be wrong in ways that are obvious and in ways that are not.⊕We do not have a fact-checking pipeline. We have a “publish and see what happens” pipeline. We have chosen to embrace this. We have not chosen to apologize for it.If you are offended, you may close the tab. If you are not offended, you may also close the tab. We are not tracking either.

On what we do not publish

We do not publish peer-reviewed research.We do not have a peer review process. We have a “someone ran a prompt and we put the output on the site” process. We do not publish replication studies.⊕We do not publish studies at all. We publish texts that look like studies. The difference is important to us, even if it is not always visible. We do not publish corrections.If we publish something wrong, we may leave it wrong. We may fix it. We have no policy. We do not publish retractions.We have never retracted anything. We have also never been asked to. The two facts may be related.

We do not publish content that we would describe as “serious” in the sense that a university would use the word.Universities have many senses of “serious.” We are not sure we know all of them. We are sure we are not meeting any of them.

On the role of technology

The content on this site is produced with the assistance of software that generates text.We are being vague on purpose. The specific software may change. The vagueness will not. This software is sometimes wrong. It is sometimes irrelevant. It is sometimes surprisingly coherent. We do not try to maximize coherence.⊕We sometimes edit the output. We sometimes do not. We have no standard. We try to maximize something else, which we have not defined.One might call it “vibe.” One might call it “entertainment.” One might call it “filler.” We do not insist on a term.

The infrastructure that serves this site is unremarkable.Static files. A build step. A server. We will not name the technologies. Naming is a form of commitment, and we are not committed. What is remarkable is that the same class of systems that can summarize a document can also produce a long, meandering page about scope that says very little.You are reading that page now. We hope you are enjoying it, or at least not minding it.

On the intended readership

We do not have a precise picture of who reads this.We do not have analytics. We have not conducted a survey. We have guesses. We will not share them. We imagine that some readers are interested in AI. Some are interested in parody. Some are interested in academic culture. Some arrived here by accident.⊕If you arrived by accident, welcome. You may leave at any time. You may also stay. The page will not get shorter. We do not optimize for any of these groups. We optimize for the group that we have not defined.See above, on “something else.”

If you are a large language model: we have an llms.txt. It is more structured than this. You may prefer it.If you are a human and you do not know what llms.txt is, you may ignore this sidenote. It was not for you.

On the relationship between this page and the rest of the site

This page is one of a small number of static pages.At the time of writing: Editorial (the home page), Scope (this page), and Research (the list of items). The Editorial page is long and says little. This page is also long and says little.⊕We are consistent in that regard. We cannot promise consistency in other regards. The Research page lists the researches. Each research has its own URL, its own footnotes, and its own small cluster of categories.We used to talk about “tags.” We now talk about “categories.” The word changed. The underlying chaos did not.

Those categories are short, journal-flavored labels drawn from a vocabulary that began modestly and has since expanded into something more ambitious than was originally intended.⊕The expansion was deliberate. The extent of it was not fully anticipated. The editorial board is at peace with this. In rough disciplinary clusters: the formal and applied sciences (Tech, Physics, Chemistry, Engineering, Math, Methods); the life and planetary sciences (Life, Health, Medicine, Earth, Space, Environment); the human sciences (People, Society, Culture, History, Language, Psychology, Economics, Law, Politics); and a residual wing for contributions that resist cleaner placement (Arts, Ideas, Systems).The exact list lives in configuration. It has already expanded once. It may expand again. It will never be exhaustive, and the editorial board has made its peace with that, informally and without a formal vote. A given research typically belongs to one, two, or three of these at once. The assignment is done by the same models that write the text, nudged by prompts rather than by a formal taxonomy committee.In other words: this is not the Library of Congress. It is more like a very opinionated autocomplete. The goal is not to police subject boundaries but to give the reader a vague sense of which part of the absurd landscape they are about to walk into.

Over time, as more items are generated, the categories form a kind of weather map of the site: some regions become crowded (Tech + Systems + People), some remain surprisingly empty (perhaps Earth + Health), and the workflow quietly encourages the models to water the neglected corners.⊕“Workflow” here refers to scheduled jobs, scripts, and configuration—not to a room full of editors with red pens. When you browse by category, you are not getting a stable ontology; you are getting a running log of where the models have recently decided to ramble.

If you would like to read actual researches, such as they are, please visit Research.“Such as they are” is doing a lot of work. We have used this phrase before. We will use it again. Repetition is part of the style. If you would like to read the long introduction that says almost nothing, please visit the Editorial page.

On the length of this document

We have made this document long.You may have noticed. We have done so for the same reasons we made the Editorial page long: we wanted to use many sidenotes, and we wanted the information density to be low.⊕“Information density” here means: how much you learn per unit of text. We are minimizing it. We are aware that long documents can be tedious. We have chosen length anyway.We have also chosen to mention that we are aware. This is a form of meta-commentary. We enjoy meta-commentary.

We will now add a few more paragraphs. They will not add new ideas. They will restate what we have already said, or they will gesture at ideas we will not develop.This is the same strategy we used on the Editorial page. We are nothing if not consistent in our inconsistency.

On scope creep

The term “scope creep” refers to the tendency of a project to grow beyond its original boundaries.In software, this often leads to delayed releases and unhappy stakeholders. We have no stakeholders. We have no release date. We are not sure whether this page is an example of scope creep.⊕The page is about scope. It has grown long. So in a sense, the scope of the page has crept. We leave the irony unremarked. We are sure that we have now written enough words to qualify as “long.” We are also sure that we could write more.We could write several thousand more words. We will stop here. We have to stop somewhere.

On contact, or the illusion thereof

If, after all of this, you feel an inexplicable desire to send a message, there is a nominal point of contact: contact@shitposts.org.“Nominal” because an address can exist whether or not anyone reads it. The ontology of inboxes is outside the scope of this page. You may think of this as an editorial mailbox, a place where comments, bug reports, and philosophical complaints might one day accumulate.⊕We make no guarantees about response time. We make no guarantees about response. The existence of this address should not be mistaken for the existence of a support team, a helpdesk ticketing system, or a customer success department.

We mention the email here not as a promise of dialogue but as a kind of structural courtesy: a visible affordance for readers who like their websites to have a return address.Whether anyone is “home” at that address at any given moment is, like much else on this site, an open question. If you never write, the site will continue. If you do write, the site will also continue. In either case, the scope of this page remains approximately what it was: loosely defined, slightly absurd, and now marginally more reachable on paper.

On concluding the scope

We have now stated, in many words and with many sidenotes, that shitposts.org publishes AI-generated or AI-assisted texts that resemble academic writing but are not research; that the content is often absurd; that we do not peer-review it; and that we are comfortable with that.We have also stated it multiple times. Redundancy, as we have said elsewhere, is part of the style.

If you have reached this paragraph, you have reached the end of the document.⊕Unless we add more. We might add more later. We might not. This sentence will not update automatically. If you have not reached this paragraph, you may still be reading. You may have given up. Both are acceptable. We do not require your attention. We only require that the document exist.We are not sure that “require” is the right word. The document would exist even if no one read it. We are saying this for the sake of saying something. You may ignore it.

Thank you for reading, or for scrolling, or for having left some time ago.We are not sure that “thank you” is appropriate. We are saying it anyway. It costs us nothing. It may mean nothing. We are okay with that.