Skip to end of metadata
Go to start of metadata

http://personal.sirma.bg/jana/search-prototype/search.html

Basics

  1. User starts with a single autocomplete field. They search across multiple thesauri.
  2. After they select a value, users select FR.
  3. They click the plus sign at end and empty field appears. They search again. If they choose value from same thesaurus, FR stays the same. Otherwise, they select FR for next section
  4. Values in each section are joined either with ORs or ANDs. (Same for all values)
  5. After FR is fixed for a section, only values from the appropriate thesauri could be entered there
  6. We need buttons: 
    1. to continue sentence (the pink plus button at end)
    2. to add/remove value to a section (the plus/minus buttons that appear when mouse pointer goes over a section)
    3. to remove a section (not shown)

Screenshots

3 FRs

Edit term

Add new FR

Labels:
None
Enter labels to add to this page:
Please wait 
Looking for a label? Just start typing.
  1. May 18, 2012

    Notes from Vlado:

    Changes we both agree about
    - and/or are small between terms, uppercase between FRs
    - how to delete terms (where are the [-])? When the last term is deleted, also delete the FR
    - this mockup uses drop-downs, which leave some empty space after the term. The implementation won't have that space
    Usability notes
    - a linear "sentence" does not elucidate the structure, especially when it wraps to a second line
    - you cannot use this UI with the keyboard alone: click, type, click, type...
    Changes that Vlado suggests
    - don't single-out "object type" as the first box. If the user wants to speify the object type, he can do it at any place
    - let's get rid of WHICH that causes "FR should be verbs" and makes the sentence overly long
      INSTEAD OF: WHICH are from Vienna[place] and are made of Bronze[material]
      WE'LL HAVE: from Vienna[place] and made of Bronze[material]
    the sentence becomes shorter and just as readable!
    - currently the last term is always from all thesauri (not the thesauri of the last FR)
    Changes we both agree about

    - and/or are small between terms, uppercase between FRs

    - how to delete terms (where are the [-])? When the last term is deleted, also delete the FR

    - this mockup uses drop-downs, which leave some empty space after the term. The implementation won't have that space

    Usability notes

    - a linear "sentence" does not elucidate the structure, especially when it wraps to a second line

    - you cannot use this UI with the keyboard alone: click, type, click, type...

    Changes that Vlado suggests

    - don't single-out "object type" as the first box. If the user wants to speify the object type, he can do it at any place

    - let's get rid of WHICH that causes "FR should be verbs" and makes the sentence overly long

      INSTEAD OF: WHICH are from Vienna[place] and are made of Bronze[material]

      WE'LL HAVE: from Vienna[place] and made of Bronze[material]

    the sentence becomes shorter and just as readable!

    - currently the last term is always from all thesauri (not the thesauri of the last FR)