W3C home > Mailing lists > Public > www-style@w3.org > July 2010

Re: Haptics CSS extension proposal

From: Christoph Päper <christoph.paeper@crissov.de>
Date: Tue, 6 Jul 2010 13:53:46 +0200
Message-Id: <9304FB85-A041-4C31-B384-606B1B0273D6@crissov.de>
To: CSS WWW Style <www-style@w3.org>
<kim.1.gronholm@nokia.com>:

> We have been researching tactile feedback (haptics) support for touch based web user interfaces, and we would like to propose the results of the research to be elaborated as web standards.
> 
> http://portaali.org/~ilkkao/w3c/CSS/css3-haptics.html

I agree with you that haptic style should be introduced to CSS.

There is no media type for tactile devices. They usually are members of the ‘screen’ type, but there should be some kind of detection via Media Queries.

I agree with Brad Kemper – rarely as that happens – that ‘haptic-tap-strength’ in your proposal describes appearance (which always seems close to behavior) and should be handled by the property of that name.

‘haptic-tap-strength’ seems to have one excess word in it, either ‘haptic’ or ‘tap’ – or the name should be totally different like Mikko Rantalainen suggests (‘force-feedback’ or ‘touch-feel’ or ‘tactile-feel’ etc.). Unlike him I believe the absolute keywords are not too much. They may be enough, but perhaps we should add ‘weaker’ and ‘stronger’ from the start; number-unit pairs may be added in a later level of CSS.

As for the “haptic types” I believe we can keep the descriptions, but the keywords should not resemble behavior / appearance. Industry jargon here is probably not as old and established and ambiguous and arbitrary as in other fields of style, such as typography.

Current Nokia proposal:
  button
    gives equal feedback for both element press and release.
  latched-button-down
    gives stronger feedback on element press than release.
  latched-button-up 
    gives lighter feedback on element press than release.
  latched-button-stuck
    gives a light feedback on both press and release.

These suggest something like
  
  <tactile-force>: strong | medium | weak | weaker | stronger | none

(You should either use ‘light’ or ‘weak’ throughout, not both.)

  tactile-force-press:   <tactile-force>
  tactile-force-release: <tactile-force>
  tactile-force: <'tactile-force-press'> <'tactile-force-release'>?
  /* maybe: */   | equal | down | up | stuck | …

Press and release might be better suited for pseudo-classes. So instead of

  button {tactile-force: medium medium;}

we would have

  button {tactile-force: none;}
  button:press, button:release {tactile-force: medium;}

Actually ‘:press’ and ‘:release' could and should be merged with existing pseudo-classes.

Current Nokia proposal (continued):
  link 
    gives standard link feedback …
  checked-checkbox
    … mimics writing a checkmark next to a checkbox.
  unchecked-checkbox 
    … mimics erasing a checkmark from a checkbox.

I don’t know what to expect from ‘link’ and I don’t know how to handle the “checkbox” values.
Received on Tuesday, 6 July 2010 11:54:20 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:29 GMT