Article on personalisable interfaces...
tony at raisingthefloor.org
Fri Sep 22 17:03:59 UTC 2017
There was a lot of discussion when Steve Githens was working on the
DevPMT. UI generation was very relevant to his work. We discussed it in
several "PCP API" meetings. I posted lots to the architecture list in May
I also wrote about UI and JSON Schema validation in general at length to
help with the PCP API discussions:
I will certainly copy the wider group on additional notes, especially
regarding this topic.
On 22 September 2017 at 16:24, Harnum, Alan <aharnum at ocadu.ca> wrote:
> A late comment on this one- I am curious about the discussions regarding
> "generating Uis from JSON schemas" but couldn't find details looking
> through mailing lists / wikis, are those taking place as part of the GPII
> Architecture meetings?
> From: fluid-work <fluid-work-bounces at lists.inclusivedesign.ca> on behalf
> of Tony Atkins <tony at raisingthefloor.org>
> Date: Thursday, September 14, 2017 at 4:15 AM
> To: Fluid Work <fluid-work at fluidproject.org>
> Subject: Article on personalisable interfaces...
> Hi, All:
> This article from the Paciello group came up in my feed recently:
> They specifically mention the GPII, but also discuss the SUPPLE project
> <http://www.eecs.harvard.edu/~kgajos/research/supple/>. There's a video
> produced by the SUPPLE team <https://www.youtube.com/watch?v=B63whNtp4qc> that
> demonstrates measuring speed and accuracy of using a pointing device for
> two users with very different needs, and then shows how the system
> generates two different interfaces that can be used to input the same data.
> Although the SUPPLE project doesn't seem to have much we can directly
> reuse, we can certainly use their ideas to inform our discussions
> regarding generating UIs from JSON schemas. We have previously discussed
> how breaking down "what can be entered" into declarative form gives us the
> ability to create one mechanism to edit a range of arbitrary (in our case)
> It may already have been obvious to everyone else, but this article
> reminded me that describing "what needs to be entered" separately from
> "how it's entered" also gives us the option to generate multiple interfaces
> tailored to individual user's needs.
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the fluid-work