TL;DR: Standard A1Z26 covers A–Z English only. Accented letters (é, ñ, ü) usually strip, transliterate, or error—check tool behavior before grading international names.
English-centric baseline
A1Z26 teaching sets assume 26 Latin letters without diacritics. José encoded as JOSE is policy choice, not universal standard.
Stripping accents
Many tools normalize é→e before map. Café → CAFE → same as unaccented if stripping on. Document for students with French vocabulary words.
Transliteration rules
Some assignments map ñ to N, ß to SS. Publish transliteration table beside cipher key for fairness.
Excluded letters
Old English ash or Icelandic thorn won’t map—use replacement or exclude word from bank.
Student name sensitivity
Ask how student prefers legal name encoded before display on board. Never force strip accents if culturally important—pick alternate nickname with consent.
Test tool behavior
Paste café, naïve, Zürich into letterstonumbers.com observe output once per semester—updates may change normalization.
Unicode education moment
Explain composed characters vs base letter—optional extension for HS CS club without overwhelming elementary cipher day.
Puzzle hunt inclusivity
Designers avoid accented answers unless clue specifies strip rule. English-only word list keeps solvers aligned globally online.
Grading policy snippet
“Accents ignored; encode base letter.” One sentence in syllabus prevents appeals.
Future encoding systems
Unicode code points differ entirely from A1Z26—mention so students do not conflate internationalization with this homework.
Transliteration policy publishing
Syllabus line: accents stripped to base letter unless assignment states otherwise—fairness for international names.
Student name sensitivity
Encode preferred classroom name only with consent. Never force strip accents on culturally important spelling.
Test tool normalization yearly
Paste café, naïve, Zürich into https://letterstonumbers.com/ each semester—updates may change behavior.
Unicode extension for HS club
Optional composed character discussion for CS club—beyond elementary cipher day scope.
Puzzle hunt English-only word lists
Global online solvers align when answers avoid accented dependency unless rule specifies strip.
Deciding what Zoë becomes before any numbers
Zoë contains a diaeresis. A1Z26 has no cell for ë. Decide the policy before anyone encodes, and print it: “ë becomes E, then encode.” Under that policy Zoë becomes ZOE, which is 26 15 5. A different class may choose to refuse the item. Both policies are honest. What is not honest is letting each student guess, because then Zoë, ZOE, and ZO are three answers to one question and the key is a coin flip.
Show the decision on the board as a transformation, not as a correction of the person’s name. The name remains Zoë in attendance, on the essay, and on the door card. Only the cipher practice string becomes ZOE, and only if the owner of the name agreed to that example. If they did not agree, use the word ZOE from a dictionary and leave the classmate’s name alone.
François and the cedilla the tool will drop
François brings a cedilla and a vowel the English alphabet does not store as a separate letter. If the published policy is “drop marks, keep the base letter,” François becomes FRANCOIS, or 6 18 1 14 3 15 9 19. Encode that string only after you have written the stripped form where the student can see it. A tool that silently drops the cedilla will produce the same numbers and hide the policy. Silence is how two teachers end up with two keys.
Test the tool with François, Zoë, and a plain word such as FRANC on the same day each term. FRANC is 6 18 1 14 3. If FRANCOIS comes back identical to FRANC, the tool ate more than the mark. If Zoë errors while François succeeds, the tool’s accent support is partial and your policy must be “strip by hand, then paste the ASCII letters,” not “paste the real name and hope.”
A class rule for names with marks
Write the rule in the syllabus in one sentence and keep the examples generic: “Marked letters are stripped to A–Z before encoding; the legal name is never rewritten outside this exercise.” Students can then encode a practice string without wondering whether their grade depends on hiding an accent. The sentence also tells a visiting teacher what the key assumes.
When a puzzle hunt uses player names, do not encode the names at all if any player uses letters outside A to Z. Encode a team nickname you all choose in ASCII, such as NORTH, which is 14 15 18 20 8. The hunt stays fair, and nobody’s name is the puzzle. Fairness here is the decision to change the token, not the decision to approximate a letter until it fits.
NORTH as the team token when a name will not fit
NORTH is the ASCII nickname from the hunt rule, and it deserves its own practice encode so the token feels chosen rather than leftover. N is 14, O is 15, R is 18, T is 20, H is 8. The row is 14 15 18 20 8. Post it next to the team, not next to anyone’s legal name. If the team dislikes NORTH, they may pick another ASCII word of similar length, such as MAPLE, but they pick it before the hunt starts and they do not encode personal names “just for fun” on the same sheet.
MAPLE, if they choose it, is 13 1 16 12 5. Write the choice on the hunt brief in plain letters and in numbers. A token that exists only as numbers will be debated at the start line. A token that exists in both forms is a label. Labels are not puzzles. The puzzle starts after the team knows its own name.
