Re: Reasons of using Canvas for UI design

On 7/28/2011 8:48 PM, David Singer wrote:
>>>> In your application, the placement of the materials are indeed based on spatial regions. They are path based.
>>> No, they are not.  There are no regions with well-defined borders that contain only one active element. Every point in the space has a mix, and the proportions in the mix vary continuously over the space.  You can pick up from one place, add that mix to the mix in your pipette, and drop some of that mix somewhere else. Where is there a path with well-defined borders?
>> There are regions: your example is certainly distinct from the use cases that I'm actively working to solve, but it does not fall
>> so far outside of existing accessibility APIs, and testing, as to be indescribable with the vocabulary we currently have access to.
>>
>> Your pipette may be in a particular location, for you to grab in the first place.
> The pointer ('mouse') is the pipette when it is over the canvas.

With all due respect, I disagree in all ways possible.
The pipette is pointer('mouse') only when the author describes it as such.

A mouse is a clumsy device; intent is also clumsy, but it's something
that the user, the author, and their technology have some say over.

In this case, pipette is only activated at the user's say so.
Again, I reference WCAG.

If you make something automatic, you're violating an agreement
as ascribed in WCAG.



>
>>   -- even if it just says "Bio-reactor with a large density of green sludge".
> That's less than the sighted person can gain, and not enough to use the application.
>
>> I drop the poison into the reactor, I'm informed that the poison has hit the reactor,
>> I'm informed that the population count has started to drop. I know where I have
>> dropped the poison, in the reactor. Perhaps I dropped it at the edge. I can hit pause
>> on the simulation now. I can move over to the edge that I dropped it on. That's another defined region.
> In a continuous area with no borders there are no regions.
Paths and z-indexing give me more than both the box model
and the UN model (of country borders). Yes, there are regions, if even 
in conflict.

Close your eyes, move your fingers over an iPad,
and you will feel me here. I know it.

> Look, you seem to be saying that all UIs have to fall into the DOM/region metaphor because that's all you have tools to handle. Every problem is a nail because I only have a hammer. I am trying to point out that one of the attractions of canvas to designers is the ability to do something else, outside the old models.

I've been here, and I'm happy to guide you to better places.

There are vocabularies that exist, in the present. They're easiest and 
most politically accessible when described as IAccessible.

Then there's what we would hope for, when we close our eyes, and open 
our minds to what is possible
when our eyes our shut and our ethnocentric capacities are diminished.

My primary attraction to canvas is the development of technologies that 
I've -only- seen displayed in the context of accessibility,
outside of old models. My primary attraction to ARIA is in supporting 
old models when I'm using new ones.

My background is in the anthropology of linguistics, where standards 
have been, tried, failed and resumed.

I'm 100% behind exploring interfaces, and any model they can be 
represented. But the moment you tell
me that I can't translate a visual interface into an eyes-free 
interface, my entire being is motivated to find
ways to prove you wrong.

The DOM metaphor is insufficient. My primary is ARIA, and it is limited, 
but it is also fundamentally linguistic;
so its limitations are a matter of practice.

The very fact that I can say:  role="spreadsheet grid" is enough for me. 
Spreadsheet is undefined,
but as an author, as a communicator, I can say it anyway.

-Charles

Received on Friday, 29 July 2011 03:58:32 UTC