Page MenuHomePhabricator

Improve usability of rank selector
Open, LowPublic

Description

Statements have a selector for the rank. It is currently implemented as dropdown that is triggered when clicking on an icon.

rankvalueicons.png (65×70 px, 1 KB)

Problems:

  1. The Icons are non-standard
  2. The terms »rank« and »value« may be unclear
  3. When which value needs to be chosen is unclear

Possible improvements

  • Make it more visible e.g. via a standard dropdown or optiongroup-toggling-buttons
  • add help
  • show rank via order
  • rank via drag and drop

related: T139578, T137781

Event Timeline

possible wording for rank: 1) currently most accurate 2) accurate 3) deprecated (or just "wrong")

In usertest, nobody stumbled over this, so far – while in some cases this is important, for many property/values it is not relevant and if users only edit those they never need to think about the (possibly unintuitive) icons.

Jan_Dittrich moved this task from Review to Doing on the WMDE-Design board.

possible wording for rank: 1) currently most accurate 2) accurate 3) deprecated (or just "wrong")

  1. currently valid
  2. valid, put (possibly) outdated
  3. know to be wrong, but (used to be) commonly believed

Yea, too long...

  1. currently valid
  2. valid, put (possibly) outdated

Thanks –this is a much clearer wording!

However, I wonder: The standard seems to be 2) "normal". However, let's say "Jaws [directed by] Spielberg" would now be normal, I assume, but it is currently valid and can't outdate (only be wrong if it turns out to be actually directed by his twin brother or so), so should it be 1) "preferred"?

Yea, too long...

Thats a problem that we might be able to solve. Better a long text than one that is not understood.

However, I wonder: The standard seems to be 2) "normal". However, let's say "Jaws [directed by] Spielberg" would now be normal, I assume, but it is currently valid and can't outdate (only be wrong if it turns out to be actually directed by his twin brother or so), so should it be 1) "preferred"?

The distinction between normal and preferred doesn't really apply in such cases. That's why we have the notion of statements having the "default" rank: if there are preferred statements, they are the default. If there are no deferred statements, the normal statements are the default.

Statements with the default rank are the once we use per default in queries and infoboxes.

Basically, the interpretation of the normal rank changes when a preferred statement is present. That's a bit confusing to model nicely, but in practice, it's quite intuitive to work with. I haven't heard any serious complaints about this semantics. Few people know about it explicitly, but few people seem to see anythign wrong with how this behaves.

OK, thats good to know.

What could be an easy-to-implement thing is to put your text in a "title" attribute, so they appear on hover on the setting menu. This is not ideal, since it does not make the reason and mechanism obvious, but it makes it at least more learnable, which would be good.

titleonrank.png (142×420 px, 16 KB)

Jan_Dittrich renamed this task from Improve usability of value specifier and rank selector to Improve usability of rank selector.Jul 7 2016, 9:57 AM
Jan_Dittrich updated the task description. (Show Details)
Jan_Dittrich updated the task description. (Show Details)

Yeah let's do the title attribute for now.

As a future step let's work on ordering/collapsing based on rank.

Reviewing the suggestions and what we got as suggestions, it seems to be really hard to "fix" 3 states ranks and that it is also not viable to redo the data model.

So to recap: We added titles to give some context of the use of these.

As a viable change that keeps the current structure and makes it more understandable, we could separate "preferred"/ "normal" and "deprecated" from each other. This could mean having a "flag" function (that either is active or not to the user) and a drop-down (of which normal or preferred is chosen) .

Reasoning: Normal and preferred are meaningful values. They might change, though, particularly a "preferred" might be demoted when a new "preferred" comes (See also the title-text of normal: "valid value but possibly historic"). Deprecated is different: It is not a correct value and stays this way, usually.

This task has been assigned to the same task owner for more than two years. Resetting task assignee due to inactivity, to decrease task cookie-licking and to get a slightly more realistic overview of plans. Please feel free to assign this task to yourself again if you still realistically work or plan to work on this task - it would be welcome!

For tips how to manage individual work in Phabricator (noisy notifications, lists of task, etc.), see https://phabricator.wikimedia.org/T228575#6237124 for available options.
(For the records, two emails were sent to assignee addresses before resetting assignees. See T228575 for more info and for potential feedback. Thanks!)