Re: Geolocation privacy implementation

Thanks, Marcos, for copying me on this one.

I like Richard's idea about documenting more of the UI considerations in a way that people actually understand, but will also offer some pieces of the "Web Security Context" story as a cautionary tale:  In that group, we made an early decision to write a spec on usability aspects that, essentially, consists of hard conformance clauses.  The result is a very abstract and very dry specification that *doesn't* include many of the good ideas that came up during discussion (and even found their way into implementations), simply because those are good, but not universally good ideas, and because the overall context for user interfaces and interactions is actually quite a bit in flux right now.

The lesson here is that you get much further in this space by just documenting "soft" best practices, usability considerations, and the reasons around them than you'd get with "hard" conformance criteria.

A corollary from that would be that a separate piece of text that documents usability considerations as best practices might be more useful than additions inside the API spec; it would also fit better with where the API spec is in the process.

(You may also want to review the mailing list discussions around some suggested UI considerations for the API spec about a year ago.)

Regards,
--
Thomas Roessler, W3C  <tlr@w3.org>  (@roessler)







On 8 Jul 2010, at 08:54, Marcos Caceres wrote:

> A colleague at Opera suggested that perhaps we could look at the "Web Security Context: User Interface Guidelines" for inspiration on what a BCP document for Geolocation UI guidelines could look like. There is certainly a degree of overlap between [1] and what we have been discussing. See, for example, the section on indicators.
> 
> Thoughts?
> 
> [1] http://www.w3.org/TR/2010/PR-wsc-ui-20100622/
> 
> On 7/6/10 1:58 PM, Marcos Caceres wrote:
>> 
>> 
>> On 7/6/10 1:27 PM, Richard L. Barnes wrote:
>>> Thanks for that contribution. As far as iOS, I haven't tested version
>>> 4, but as of version 3, there was a similar allow/deny, but it
>>> automatically remembered, with no way to revoke.
>>> 
>> 
>> FYI, I wrote a blog post a few weeks ago about "Privacy issues in Mobile
>> Safari on iPhone OS 3.2".
>> 
>> http://datadriven.com.au/2010/06/privacy-issues-in-mobile-safari-on-iphone-os-3-0/
>> 
>> 
>> 
>>> On Jul 6, 2010, at 4:49 AM, Angel Machín wrote:
>>> 
>>>> Hi Richard,
>>>> 
>>>> On Tue, Jul 6, 2010 at 4:09 AM, Richard L. Barnes <rbarnes@bbn.com>
>>>> wrote:
>>>>> It seems like it could be helpful to users if this group could
>>>>> discuss some
>>>>> of these areas where there are differences, and maybe produce a BCP
>>>>> for how
>>>>> they should be resolved (probably in the form of some privacy UI
>>>>> recommendations in the API spec).
>>>>> 
>>>> 
>>>> I have tested the web app in Android, I am attaching a screen snapshot.
>>>> 
>>>> Regards,
>>>> Angel
>>>> <_android_geo.png>
>>> 
>>> 
>> 
> 
> -- 
> Marcos Caceres
> Opera Software
>

Received on Thursday, 8 July 2010 11:00:35 UTC