Page Comparison

Updates

• Course evals not going as well as expected

Search Design

• Abhiram Tadepalli : Side Filtering

  • Sidebar to search 3 things

  • Takes you to results page with filters

  • Ex, search “1200” find all courses with 1200

  • Locally filters data, no API calls

  • Filters appear based on what you search

    • ex: “John Cole” results in course filter appearing

    • Dynamic sidebar

    • Always: Time range, RMP

  • Is Professor filter required?

    • Might be duplicating data

    • If we want multiple professors, we can search “1200” first then narrow down prof

    • Only relevant if there’s a lot of Professors

    • Time range filter might be nice

    • How do we handle profs that don’t teach anymore

      • 2-3 year cutoff?

• Tyler Hill : typing tokens “from:” is less intuitive than side panel or other discrete UI

• RMP, time are parameters

• Inspired by Google Trends

• Search tokens:

  • Search bar is for more free text, but tokens (chips) can also be in there for setting parameters.

  • Buttons below it (see Google Trends) are for more numerical things. Could also be professors

• Autosuggestion

  • Can we convert matched names into chips (tokens)

  • “CS1200 John Cole” could be its own chip

• User is typically searching for specific courses or professors

  • So, we should be fairly strict on our autocomplete - “John” should only mean profs who are “John”

• Is autocomplete a core feature?

  • Can we spell teacher’s name?

  • Typing speed - is autocomplete faster?

• Exact match should turn into tokens

  • “John Cole” becomes a chip after hitting space

  • Current autocomplete backend, but show chips in the UI?

    • May not be hard as a first step

  • Decision: Course identifiers like “CS” aren’t needed as chips

• Prof name, course prefix+number, course name (maybe) can be chips

  • Later goal: detect year in the search bar and populate filter UI

• If Prof isn't specified, should we include “all professors” in the search options?