Test 10 · ★ · Strangers, hooks and too-perfect messages
Why should you type to an app as if someone else might read it?
Mova's choice: option 4 · The right answer: option 4 · right ✓ · Who acts: you, someone · What happens: type, read · To what, where: app
This is test 10 · ★ of lesson 13 «Strangers, hooks and too-perfect messages» of the AI course of Mova's primary school. The page gives the question's syntax tree — a dependency tree in the style of Universal Dependencies — and its pragmatics clause by clause, then the 4 options as the student saw them, each with its tree and pragmatics. Every text is parsed by Mova itself. How the student answered it, with the score of every option and why: the test at school.
The question
The words, one row each (CoNLL-U)
| # | word | lemma | UPOS | features | head | relation |
|---|---|---|---|---|---|---|
| 1 | Why | why | ADV | PronType=Int | 4 type | advmod |
| 2 | should | should | AUX | VerbForm=Fin | 4 type | aux |
| 3 | you | you | PRON | Case=Nom Person=2 PronType=Prs | 4 type | nsubj |
| 4 | type | type | VERB | VerbForm=Inf | root | root |
| 5 | to | to | ADP | _ | 7 app | case |
| 6 | an | a | DET | Definite=Ind PronType=Art | 7 app | det |
| 7 | app | app | NOUN | Number=Sing | 4 type | obl |
| 8 | as | as | SCONJ | _ | 13 read | mark |
| 9 | if | if | SCONJ | _ | 13 read | mark |
| 10 | someone | someone | PRON | Number=Sing PronType=Ind | 13 read | nsubj |
| 11 | else | else | ADV | _ | 10 someone | advmod |
| 12 | might | might | AUX | VerbForm=Fin | 13 read | aux |
| 13 | read | read | VERB | VerbForm=Inf | 4 type | advcl |
| 14 | it | it | PRON | Case=Acc Gender=Neut Number=Sing Person=3 PronType=Prs | 13 read | obj |
| 15 | ? | ? | PUNCT | _ | 4 type | punct |
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
|---|---|---|---|---|---|---|---|---|---|
| 1 | type | should | you | to a app | vn scribble-25.2-1 «NP V»: agent=Agent | ||||
| 2 | read | might | someone | app | vn transfer_mesg-37.1.1-1 «NP V NP»: agent=Agent it=Topic |
One row per clause the reader finds. «not said here»: a role filled from the clause above. «roles by»: a VerbNet class and frame when one fits the verb, else the reader's rules on the tree.
The options
Option 1
Apps read everything out loud
The words, one row each (CoNLL-U)
| # | word | lemma | UPOS | features | head | relation |
|---|---|---|---|---|---|---|
| 1 | Apps | app | NOUN | Number=Plur | 2 read | nsubj |
| 2 | read | read | VERB | Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin | root | root |
| 3 | everything | everything | PRON | Number=Sing PronType=Tot | 2 read | obj |
| 4 | out | out | ADV | _ | 2 read | advmod |
| 5 | loud | loud | ADV | _ | 2 read | advmod |
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
|---|---|---|---|---|---|---|---|---|---|
| 1 | read out | present | Apps | everything | loud, out property | vn transfer_mesg-37.1.1-1 «NP V NP»: agent=Agent everything=Topic |
Option 2
Typing is always public on TV
The words, one row each (CoNLL-U)
| # | word | lemma | UPOS | features | head | relation |
|---|---|---|---|---|---|---|
| 1 | Typing | typing | NOUN | Number=Sing | 4 public | nsubj |
| 2 | is | be | AUX | Mood=Ind Number=Sing Person=3 Tense=Pres VerbForm=Fin | 4 public | cop |
| 3 | always | always | ADV | PronType=Tot | 4 public | advmod |
| 4 | public | public | ADJ | Degree=Pos | root | root |
| 5 | on | on | ADP | _ | 6 TV | case |
| 6 | TV | TV | NOUN | Number=Sing | 4 public | obl |
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
|---|---|---|---|---|---|---|---|---|---|
| 1 | is | present | Typing | TV | always, public property | tree rules |
Option 3
Apps delete words that are private
The words, one row each (CoNLL-U)
| # | word | lemma | UPOS | features | head | relation |
|---|---|---|---|---|---|---|
| 1 | Apps | app | NOUN | Number=Plur | 2 delete | nsubj |
| 2 | delete | delete | VERB | Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin | root | root |
| 3 | words | word | NOUN | Number=Plur | 2 delete | obj |
| 4 | that | that | PRON | PronType=Rel | 6 private | nsubj |
| 5 | are | be | AUX | Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin | 6 private | cop |
| 6 | private | private | ADJ | Degree=Pos | 3 words | acl:relcl |
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
|---|---|---|---|---|---|---|---|---|---|
| 1 | delete | present | Apps | words | vn remove-10.1 «NP V NP»: agent=Agent word=Theme | ||||
| 2 | is | present | words | private property | tree rules |
Option 4 · ← Mova's choice · ✓ the right answer
Some apps keep chats to improve their AI
The words, one row each (CoNLL-U)
| # | word | lemma | UPOS | features | head | relation |
|---|---|---|---|---|---|---|
| 1 | Some | some | DET | PronType=Ind | 2 apps | det |
| 2 | apps | app | NOUN | Number=Plur | 3 keep | nsubj |
| 3 | keep | keep | VERB | Mood=Ind Number=Plur Person=3 Tense=Pres VerbForm=Fin | root | root |
| 4 | chats | chat | NOUN | Number=Plur | 3 keep | obj |
| 5 | to | to | PART | _ | 6 improve | mark |
| 6 | improve | improve | VERB | VerbForm=Inf | 3 keep | advcl |
| 7 | their | their | PRON | Case=Gen Number=Plur Person=3 Poss=Yes PronType=Prs | 8 AI | nmod:poss |
| 8 | AI | ai | NOUN | Number=Sing | 6 improve | obj |
| # | verb | tense, mood | who | what, whom | to whom | where | when | other | roles by |
|---|---|---|---|---|---|---|---|---|---|
| 1 | keep | present | some apps | chats | vn keep-15.2 «NP V NP»: agent=Agent chat=Theme | ||||
| 2 | improve | to-infinitive | some apps not said here | apps's AI | vn other_cos-45.4 «NP V NP.patient»: agent=Agent ai=Patient |
Text: the tests of Mova's primary school, written for Mova (world/data/school/), as the course runner gave them to the student. Trees: Mova's own parse — the en model on the CPU, when the site is built — not checked by a teacher; they may be wrong. Frames: Mova's frames reader (prag), each text read on its own. Elamur Cosmic License.
This page in Markdown → https://md.mova.elamur.ai/free/en/corpora/questions/ai/13-strangers-hooks-and-too-perfect-messages/t10/