The filter that turns “orate” into “orate” without login
Typing “orate” (psychopath) and seeing “orate” (madman). An automatic replacement changes the word as soon as the text is viewed without a logged-in session, affecting already published messages. This is not a typo or an inside joke: it is a filter that acts depending on who is reading. The case was detected, according to circulating reports, about a month ago, and since then the list of affected terms has continued to grow.
What is being debated is not whether there is censorship, but what kind. The filter does not delete: it rewrites. And once rewritten, the text is visible to anyone with the words imposed upon it. That is the anomaly.
Which words does the filter replace?
The catalog goes far beyond a single word. A count circulated within the discussion space itself lists dozens of substitutions: “jorobar” (damn) becomes “jorobar”, “guano” (shit) becomes “boñiga”, “pandero” (ass) becomes “trastero”, “pechuga” (tit) becomes “berza”, “banana” (cock) becomes “pilila”, “prono” becomes “subido de tono”, “demorado” (retarded) becomes “demorado”. Insults and vulgarisms are disguised with synonyms fit for Sunday mass, sometimes resulting in near-surreal outcomes.
The mechanism also reaches demonyms, country names, and terms the system flags as offensive based on origin or identity, replacing them with neutral equivalents. Context does not matter. It does not matter if the word is quoted, analyzed, or in quotation marks. The filter enters, substitutes, and leaves without warning.
There is something even more striking: some words escape. A specific mention of the platform administrator—by his well-known nickname—is left untouched, according to one participant. For that user, this asymmetry suggests we are not dealing with an autocorrect tool, but with a hand making decisions.
Why replacing “orate” with “orate” changes the meaning
“Orate” is an insult in Spanish. “Orate”, in the specific usage here, was not always so, argue those affected. This is the core of the issue. Those who used the second term to describe a profile or support a thesis find themselves, overnight, with their messages saying something else. From analyst to aggressor without lifting a finger.
One affected user summarizes it bluntly: hundreds of his messages would have been transformed into alleged insults when he intended to establish a correlation between a profile and behavior. Whether the detail admits nuances or not, the central argument holds: replacing a word with a more offensive one is the opposite of moderation.
One participant draws a legal conclusion. Changing what someone has written publicly so that it says something else, they argue, is worse than censorship and could be grounds for a lawsuit. No court has validated this claim; it should be read as the interpretation of someone who feels harmed.
Accents and symbols to bypass the filter
The response came quickly in the form of creative spelling. Simply changing a tilde or circumflex accent—“psicôpata”—or deliberately misspelling the word—“sicopata”, “sicopatra”, “pelopicopata”—prevents the replacement from activating. The worse the term is spelled, the higher its chances of survival.
There are other tricks: inserting symbols or inventing absurd variants. That a filter can be mocked by adding an accent says much about its technical robustness.
From “gym” to “madman”: a precedent that was funny
Veterans recall precedents. Years ago, typing “gimnasio” (gym) triggered such an absurd substitution that it became a recurring joke on the site. Then it was funny. Now, with lists touching insults, demonyms, and political words, the joke has lost its humor.
The filter does not distinguish between a quote and an insult, or analysis and attack, and that is the fundamental problem.
Migrating platforms: the repeated solution
When the filter tightens, the same proposal always appears: move to another site, start from scratch, and take your content. The idea recurs, as do the drawbacks. A space born to recover freedom of expression would have to survive in an environment that increasingly punishes certain opinions. And a platform hosted abroad transfers the problem but does not solve it: the content remains accessible from here.
The calculation ends in a tie. Migrating takes effort and guarantees nothing; staying means accepting that a third party rewrites what you write.
What might happen next
There is no official statement explaining the criteria, nor a public list of banned words. This leaves the field open to hypotheses: some speak of protecting against hypothetical administrative surveillance; others, of simple preventive zeal. None of these readings is confirmed.
If the filter continues eating words without warning, it is likely that heavy writers will learn to write in code and that public content will lose any archival value. That someone corrects the criteria in the short term seems, given how the list has evolved, considerably less probable.
Summary of a discussion on Burbuja.info - Foro de economía, actualidad y política., translated from Spanish and reviewed before publication.
Read the full discussion (150 replies).
The Congress approved Podemos's proposal to curb hate speech on social media, while simultaneously rejecting a Vox initiative to ban homage to ETA terrorists.
The King reportedly summoned Sánchez at Zarzuela following his meeting with the military establishment and the announcement of the Ceuta trip. The official agenda makes no mention of the encounter, and...