Repository navigation
docs(mypolitics): quiz spec changes from the QA round decisions - #86
Conversation
…z specs Decided by the Technical Leader on 2026-10-09, after the QA round on the result modules in mypolitics-app: - universal axis: a comparison no longer hides the taker's value; a side keeps its number wherever the band and the other party's image leave it uncovered (gi-org-pl/mypolitics-app#134) - double axis chart: on a card too narrow for both names a tie title shows the word for a tie, never one pole alone (gi-org-pl/mypolitics-app#137) - horizontal bar chart: a category with one entry has no control (gi-org-pl/mypolitics-app#139) Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
The spec said that whatever falls outside the map is cut off. The module was delivered with the dot drawn in full over the edge, because a data mark is never cut (AGENTS.md of mypolitics-app, section 3.4; decision recorded in gi-org-pl/mypolitics-app#82). The sentence is brought in line with that. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: aa1276c6fb
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
From review: - double axis chart: the tie title switches on a fixed breakpoint of the title slot, not on whether the names fit, which could not be known without measuring; names too long for a wide slot are truncated one by one, so both poles stay visible - universal axis: when the whole track is hatched, no value is shown on either side - both specs say that the frame at the top of the page predates the rule on values next to a comparison and is to be redrawn Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e7b3565719
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
…ll as drawn From review: with values that exceed the track the fills are scaled, and the spec did not say whether the band follows the value as given or the scaled fill. It follows the fill, as the implementation does, which is what makes the position of the other party enough to decide whether a number stays visible. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 2d74756ed7
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
…shows the tie word From review: with one name missing, naming "both" poles would show the surviving one alone, which is exactly what the tie title must never do; with both missing the chip would be empty. In both cases the chip shows the word for a tie, at any width. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5b4433ee39
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
… missing pole name From review: the fallback to the word for a tie covered the visible chip only. The card's accessible name follows it. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 84f35a988a
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
…on the dot at an edge From review: the picture on the page shows the corner dot differently from the rule. It is an export of the Figma frame, which is to be redrawn; the spec says so until then, as the other two changed specs do. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
@codex review |
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: 5660c03cd3
ℹ️ About Codex in GitHub
Your team has set up Codex to review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
Codex can also answer questions or update the PR. Try commenting "@codex address that feedback".
| | Other is at 0 or 100 | The image is clamped so it stays fully inside the track | | ||
| | Double-sided bar | The band is measured against the start entry on the same shared track | | ||
|
|
||
| The band always ends at the taker's fill as it is drawn. When two values exceed the track and the fills are scaled, the band follows the scaled fill, not the value as given, so it never runs past the fill it belongs to. |
There was a problem hiding this comment.
When a double-sided bar exceeds 100, ending the band at the scaled fill can reverse the documented comparison direction. For example, start/end values of 90/90 put the rendered start fill at 50, while another party's value of 70 is behind the taker's raw 90; the resulting band from 50 to 70 lies past the fill rather than over it, contradicting the “other is behind” rule and visually presenting the other party as ahead. Define comparison ordering and positioning in the same scaled coordinate system, or otherwise handle cases where scaling moves the taker's fill across the comparison value.
Useful? React with 👍 / 👎.
There was a problem hiding this comment.
Right about the fact: with 90 / 90 the start fill is drawn at 50 while the other party's 70 is not scaled, so the band lands past the fill although 70 is behind the value as given. This is how the bar already degrades for input that is wrong to begin with (two sides of one axis exceeding 100), and this pull request does not change it. Not fixed here, because scaling the other party's position as well is a change of behaviour that nobody has decided. d2e3516 makes the spec say that the result is degraded and the band can then be on the other side of the fill, and the question is recorded as an open point in gi-org-pl/mypolitics-app#134.
…aled bar From review: on a bar whose fills were scaled the other party's position is not scaled, so the band can sit on the other side of the fill than the values as given would put it. That is how the bar degrades today for input that is already wrong; the spec now says so instead of implying a correct result. Whether the position should be scaled too is left open in gi-org-pl/mypolitics-app#134. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
|
@codex review |
|
You have reached your Codex usage limits for code reviews. You can see your limits in the Codex usage dashboard. |
Brings in the questionnaire specs and tasks (#85). universal-axis.md was changed on both sides and merged without conflicts. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com>
1 participant
What
Spec changes that follow the decisions taken after the QA round on the result modules of
mypolitics-app(epic gi-org-pl/mypolitics-app#144). Specs only - no docs or task files change.universal-axis.mddouble-axis-chart.mdhorizontal-bar-chart.mdnolan-chart.mdNotes for the reviewer
Verification
universal-axis.md), is not reviewed: the review bot answered that its usage limit is reached. To be requested again when the limit is back.#with-a-comparison) points at a heading added in the same file.🤖 Generated by Gieniek with Claude Code