Etymology Bot

Sending a word in any language and answering in the Telegram language is harder than the English Wiktionary path. Salary already has a clean Latin story on the English page. What happens for a word whose best sources are not English, for example a Chinese or Arabic term that is thin in English Wiktionary but long in its own language? Does the bot quote the original-language page and translate that quote, or only cite English? Folk etymologies are often the lines those pages quote most. If every claim only needs a verbatim quote, a false folk story that Wiktionary itself quotes would still pass. Do you mark those as disputed, or does any quoted line count?

A quote for every claim is exactly what most etymology sites are missing, half of them just repeat each other. Love the idea. Do you pull the sources yourselves or from a set of dictionaries? Congrats on the launch!

The salary page is a good test of the quote-or-drop rule: the Latin salarium story only counts if a Wiktionary line is attached. What is less clear is the free cap. The listing says popular words are cached and shared so they are instant, and that the free plan is 30 lookups a month. Does opening a cached word still spend one of those 30, or only a fresh fetch? If cache hits are free, the limit mostly hits rare words, which is exactly when a missing quote would drop the answer. Pricing Pro in Telegram Stars instead of a card also changes who can pay without another account.

Requiring a verbatim quote for every claim and dropping any claim without a match is a smart guard against hallucinated etymologies. Since answers are cached and shared across users, how do you handle a word whose Wiktionary entry gets edited later? Is the cache refreshed periodically or only on demand?

Hi, I am a solo developer from Ukraine. I built this because etymology answers online are of two kinds. Some are dry dictionary entries. Others are unsourced trivia, where you cannot tell fact from folk tale. I wanted the good part of etymology, the story, with proof next to it. So every claim in an answer must come with a verbatim quote from a cited source, and code removes any claim that has no matching quote. Try "salary" first. I would like to hear which words give a wrong or thin answer. Thank you for looking.

Honestly, Etymology is a really interesting science and now you can get to know a lot of meaning of words with proofs. So, it is not only like a way to be higher educated, but also, a variant for a interesting evening with friend spend with this bot.

Dropping any claim that has no verbatim quote is a strong way to prevent invented word histories. Have you considered adding a short 'confidence' note when a word only has one thin source? Also, how do the 30 free lookups count when a result is already cached from another user?

How does the quote-or-drop rule handle words where Wiktionary lists several competing origins on the same page? I would expect it to keep only the proposal with a matching quote, so I wonder whether the answer shows the alternatives side by side or picks one.

Interesting product!

Enforcing the quote-or-drop rule in code — deleting any claim that has no verbatim quote from the cited source before the answer is composed — is a genuinely strong guard against the invented word histories that fill most etymology sites. Two questions: when Wiktionary lists competing origins for a word, does the bot show all the quoted candidates or pick one? And do you plan to add sources beyond English Wiktionary over time?

The rule that code drops any claim without a verbatim quote from a cited source is a sensible guard against made-up etymologies. When a word has several competing origins in Wiktionary, does the answer show which claims were dropped, or just leave them out silently? A short note like 'one theory omitted, no quote found' would help readers judge how complete the answer is.

The group-chat angle is clever — using /etym in a Telegram group turns a word question into a shared moment instead of a solo dictionary lookup. I also like that answers follow your Telegram language and cover relatives in other languages and how meanings drifted, not just a one-line origin. The cached answers for popular words should make it feel instant in practice. Do you plan to let users save favourite etymologies to revisit later?

Backing every claim with a verbatim quote is exactly how etymology should be done. It's a refreshing change from all the confident myths online. Lovely idea, well executed!

The quote-or-drop rule is the whole product, honestly. In an era where every AI chatbot happily invents word histories, enforcing verbatim quotes in code — deleting any claim without a matching quote — is exactly the right guardrail. I also like that it answers in the user's Telegram language and that popular words are cached for instant replies. One question: how does it handle words where Wiktionary lists several competing origins? Does it present all of them with their quotes, or pick the most cited one? And do the sources go beyond Wiktionary and Wikipedia to, say, the OED or academic etymological dictionaries? Congrats on the launch!

I opened the public salary answer and the Latin salarium line is quoted rather than paraphrased, which matches the drop-if-no-quote rule. Free is 30 lookups a month and Pro is 199 Telegram Stars. Does a cached popular word still count against those 30, or only a fresh fetch? In a group that decides whether /etym salary is free after the first person asks.

Sourcing every claim with a verbatim quote is a smart guardrail against made-up etymologies. One thing I'd find useful: showing which Wiktionary page and language edition each quote came from, so users can check it quickly.

Dropping any claim that has no matching quote is a strong rule, and the salary example shows it well. I like that the answer tells the story of how a meaning drifted, not just a dry root. That is the kind of detail that makes people care about the craft of words.