Agenda+ CSS Round Display Issues

Previous Topic Next Topic
 
classic Classic list List threaded Threaded
9 messages Options
Reply | Threaded
Open this post in threaded view
|

Agenda+ CSS Round Display Issues

Jihye Hong
I would like to discuss about integrating polar positioning of CSS Round Display to Motion Path.

I wrote the draft about that:
   Spec Draft: https://drafts.csswg.org/css-round-display/#positioning-content 
and related thread is here:
   Thread: https://lists.w3.org/Archives/Public/www-style/2016Jun/0077.html 

The topics for the discussion are:
[css-round-display][motion-path] Integrate polar positioning to the motion path spec
   1. Need for 'offset-origin'
   2. Initial value of 'offset-rotation'

= Jihye


Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Lars Knudsen-2
Hi,

this is the first I hear of CSS Round Display (been living under a rock).

One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?

One could imagine that a circle is just a specialization of some closed bezier curve (or other).

just a thought

br
Lars 

On Mon, Jun 20, 2016 at 4:41 AM, Jihye Hong <[hidden email]> wrote:
I would like to discuss about integrating polar positioning of CSS Round Display to Motion Path.

I wrote the draft about that:
   Spec Draft: https://drafts.csswg.org/css-round-display/#positioning-content
and related thread is here:
   Thread: https://lists.w3.org/Archives/Public/www-style/2016Jun/0077.html

The topics for the discussion are:
[css-round-display][motion-path] Integrate polar positioning to the motion path spec
   1. Need for 'offset-origin'
   2. Initial value of 'offset-rotation'

= Jihye



Reply | Threaded
Open this post in threaded view
|

RE: Agenda+ CSS Round Display Issues

Jihye Hong
Hi Lars,

When LG suggested ideas of CSS Round Display, we wanted to apply the spec to the smart watch of LG with rounded display.
So we added some features to support the rounded display, especially for circular display at first.

But while developing the spec, we started to consider various shapes of displays.
The shapes like oval, ellipse can be dealt with CSS Round Display because they are included in the boundary of "round display".
Some features in the spec cover other than the rounded shape, such as the rounded rectangular shape.

Also, I'm working on figuring out the other ideas for CSS Round Display which can be applied for various shapes.

If you have any other thoughts about the spec, please feel free to ask me :)

Thanks,
Jihye

> On Mon, Jun 20, 2016 at 5:46 PM, Lars Knudsen < [hidden email] > wrote:
>
> Hi,
>
> this is the first I hear of CSS Round Display (been living under a rock).
>
> One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to
> consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?
>
> One could imagine that a circle is just a specialization of some closed bezier curve (or other).
>
> just a thought
>
> br
> Lars
>
>> On Mon, Jun 20, 2016 at 4:41 AM, Jihye Hong <[hidden email]> wrote:
>> I would like to discuss about integrating polar positioning of CSS Round Display to Motion Path.
>>
>> I wrote the draft about that:
>>   Spec Draft: https://drafts.csswg.org/css-round-display/#positioning-content
>> and related thread is here:
>>   Thread: https://lists.w3.org/Archives/Public/www-style/2016Jun/0077.html
>>
>> The topics for the discussion are:
>> [css-round-display][motion-path] Integrate polar positioning to the motion path spec
>>   1. Need for 'offset-origin'
>>   2. Initial value of 'offset-rotation'


Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Florian Rivoal-4
In reply to this post by Lars Knudsen-2

On Jun 20, 2016, at 17:46, Lars Knudsen <[hidden email]> wrote:

Hi,

this is the first I hear of CSS Round Display (been living under a rock).

One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?

I think we need to walk a fine line between designing a mechanism that makes enough sense in general that we won't be stuck next time something else than a rectangle or a circle becomes popular, but at the same time refrain from over engineering things for devices that don't actually exist and maybe never will.

I think the polar positioning / motion path does that: a polar coordinate system is clearly most useful for round things, but the 'contain' / 'closet-side' / ... keywords work in a generic fashion, which makes it suitable (if not fine tuned) for arbitrary other shapes.

(although I think that these keywords are currently attached to the wrong property, but I'll bring that up separately).

 - Florian
Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Lars Knudsen-2
I have one good example of current tech with odd shaped displays (or masks):  Car dashboards.  Apart from that, we will most probably see other than round (circle, ellipse) and rectangular displays soon in appliances.  I am just saying that one could at least take a step back and consider if there could be another slightly more generic way of representing the shape than limiting it to a circle (and probably ellipse).  E.g. using NURBS or the like.

br
Lars

On Tue, Jun 21, 2016 at 9:12 AM, Florian Rivoal <[hidden email]> wrote:

On Jun 20, 2016, at 17:46, Lars Knudsen <[hidden email]> wrote:

Hi,

this is the first I hear of CSS Round Display (been living under a rock).

One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?

I think we need to walk a fine line between designing a mechanism that makes enough sense in general that we won't be stuck next time something else than a rectangle or a circle becomes popular, but at the same time refrain from over engineering things for devices that don't actually exist and maybe never will.

I think the polar positioning / motion path does that: a polar coordinate system is clearly most useful for round things, but the 'contain' / 'closet-side' / ... keywords work in a generic fashion, which makes it suitable (if not fine tuned) for arbitrary other shapes.

(although I think that these keywords are currently attached to the wrong property, but I'll bring that up separately).

 - Florian

Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Lars Knudsen-2
In reply to this post by Jihye Hong
Hi Jihye,

that sounds great, thanks!  Does LG with with displays of other shapes for e.g. appliances, cars, kiosk displays, etc?

br
Lars

On Tue, Jun 21, 2016 at 3:16 AM, Jihye Hong <[hidden email]> wrote:
Hi Lars,

When LG suggested ideas of CSS Round Display, we wanted to apply the spec to the smart watch of LG with rounded display.
So we added some features to support the rounded display, especially for circular display at first.

But while developing the spec, we started to consider various shapes of displays.
The shapes like oval, ellipse can be dealt with CSS Round Display because they are included in the boundary of "round display".
Some features in the spec cover other than the rounded shape, such as the rounded rectangular shape.

Also, I'm working on figuring out the other ideas for CSS Round Display which can be applied for various shapes.

If you have any other thoughts about the spec, please feel free to ask me :)

Thanks,
Jihye

> On Mon, Jun 20, 2016 at 5:46 PM, Lars Knudsen < [hidden email] > wrote:
>
> Hi,
>
> this is the first I hear of CSS Round Display (been living under a rock).
>
> One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to
> consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?
>
> One could imagine that a circle is just a specialization of some closed bezier curve (or other).
>
> just a thought
>
> br
> Lars
>
>> On Mon, Jun 20, 2016 at 4:41 AM, Jihye Hong <[hidden email]> wrote:
>> I would like to discuss about integrating polar positioning of CSS Round Display to Motion Path.
>>
>> I wrote the draft about that:
>>   Spec Draft: https://drafts.csswg.org/css-round-display/#positioning-content
>> and related thread is here:
>>   Thread: https://lists.w3.org/Archives/Public/www-style/2016Jun/0077.html
>>
>> The topics for the discussion are:
>> [css-round-display][motion-path] Integrate polar positioning to the motion path spec
>>   1. Need for 'offset-origin'
>>   2. Initial value of 'offset-rotation'


Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Lars Knudsen-2

On Tue, Jun 21, 2016 at 9:53 AM, Lars Knudsen <[hidden email]> wrote:
Hi Jihye,

that sounds great, thanks!  Does LG with with displays of other shapes for e.g. appliances, cars, kiosk displays, etc?

br
Lars

On Tue, Jun 21, 2016 at 3:16 AM, Jihye Hong <[hidden email]> wrote:
Hi Lars,

When LG suggested ideas of CSS Round Display, we wanted to apply the spec to the smart watch of LG with rounded display.
So we added some features to support the rounded display, especially for circular display at first.

But while developing the spec, we started to consider various shapes of displays.
The shapes like oval, ellipse can be dealt with CSS Round Display because they are included in the boundary of "round display".
Some features in the spec cover other than the rounded shape, such as the rounded rectangular shape.

Also, I'm working on figuring out the other ideas for CSS Round Display which can be applied for various shapes.

If you have any other thoughts about the spec, please feel free to ask me :)

Thanks,
Jihye

> On Mon, Jun 20, 2016 at 5:46 PM, Lars Knudsen < [hidden email] > wrote:
>
> Hi,
>
> this is the first I hear of CSS Round Display (been living under a rock).
>
> One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to
> consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?
>
> One could imagine that a circle is just a specialization of some closed bezier curve (or other).
>
> just a thought
>
> br
> Lars
>
>> On Mon, Jun 20, 2016 at 4:41 AM, Jihye Hong <[hidden email]> wrote:
>> I would like to discuss about integrating polar positioning of CSS Round Display to Motion Path.
>>
>> I wrote the draft about that:
>>   Spec Draft: https://drafts.csswg.org/css-round-display/#positioning-content
>> and related thread is here:
>>   Thread: https://lists.w3.org/Archives/Public/www-style/2016Jun/0077.html
>>
>> The topics for the discussion are:
>> [css-round-display][motion-path] Integrate polar positioning to the motion path spec
>>   1. Need for 'offset-origin'
>>   2. Initial value of 'offset-rotation'



Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Richard Gibson
I'm pretty sure this can be restricted to genus-0 topologies for quite a while, and cover/contain could be extended to other values quantifying the amount of coverage for less regular shapes.

On Tuesday, June 21, 2016, Lars Knudsen <[hidden email]> wrote:

On Tue, Jun 21, 2016 at 9:53 AM, Lars Knudsen <<a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;larsgk@gmail.com&#39;);" target="_blank">larsgk@...> wrote:
Hi Jihye,

that sounds great, thanks!  Does LG with with displays of other shapes for e.g. appliances, cars, kiosk displays, etc?

br
Lars

On Tue, Jun 21, 2016 at 3:16 AM, Jihye Hong <<a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;jh.hong@lge.com&#39;);" target="_blank">jh.hong@...> wrote:
Hi Lars,

When LG suggested ideas of CSS Round Display, we wanted to apply the spec to the smart watch of LG with rounded display.
So we added some features to support the rounded display, especially for circular display at first.

But while developing the spec, we started to consider various shapes of displays.
The shapes like oval, ellipse can be dealt with CSS Round Display because they are included in the boundary of "round display".
Some features in the spec cover other than the rounded shape, such as the rounded rectangular shape.

Also, I'm working on figuring out the other ideas for CSS Round Display which can be applied for various shapes.

If you have any other thoughts about the spec, please feel free to ask me :)

Thanks,
Jihye

> On Mon, Jun 20, 2016 at 5:46 PM, Lars Knudsen < <a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;larsgk@gmail.com&#39;);" target="_blank">larsgk@... > wrote:
>
> Hi,
>
> this is the first I hear of CSS Round Display (been living under a rock).
>
> One question - if you in the spec state: "Nowadays, devices come in varied shapes of the displays. It needs to
> consider the shape of the display when implementing web pages on devices.", why do you then limit the sperc to focus on *one* more (very specific) shape?
>
> One could imagine that a circle is just a specialization of some closed bezier curve (or other).
>
> just a thought
>
> br
> Lars
>
>> On Mon, Jun 20, 2016 at 4:41 AM, Jihye Hong <<a href="javascript:_e(%7B%7D,&#39;cvml&#39;,&#39;jh.hong@lge.com&#39;);" target="_blank">jh.hong@...> wrote:
>> I would like to discuss about integrating polar positioning of CSS Round Display to Motion Path.
>>
>> I wrote the draft about that:
>>   Spec Draft: https://drafts.csswg.org/css-round-display/#positioning-content
>> and related thread is here:
>>   Thread: https://lists.w3.org/Archives/Public/www-style/2016Jun/0077.html
>>
>> The topics for the discussion are:
>> [css-round-display][motion-path] Integrate polar positioning to the motion path spec
>>   1. Need for 'offset-origin'
>>   2. Initial value of 'offset-rotation'



Reply | Threaded
Open this post in threaded view
|

Re: Agenda+ CSS Round Display Issues

Florian Rivoal-4
In reply to this post by Lars Knudsen-2

> On Jun 21, 2016, at 16:51, Lars Knudsen <[hidden email]> wrote:
>
> I have one good example of current tech with odd shaped displays (or masks):  Car dashboards.  Apart from that, we will most probably see other than round (circle, ellipse) and rectangular displays soon in appliances.  I am just saying that one could at least take a step back and consider if there could be another slightly more generic way of representing the shape than limiting it to a circle (and probably ellipse).  E.g. using NURBS or the like.

Right, these are good examples of oddly shaped screens (although it is questionable how likely a generic purpose web browser on a car dashboard is...).

However, I don't think we are restricting ourselves to a circle or an ellipse with the current proposals:

* Polar coordinates (now under the offset-* properties) work in arbitrary shaped elements/screens, and thanks to the contain keyword, can take that shape into account

* shape-inside:display, viewport-fit and border-boundary are definitely shape agnostic

* For media queries we eventually plan to have two different media queries. One that lets you detect whether any particular location in the viewport would be obscured by the screen shape (whatever that shape is), and one that tells you if the screen is rectangular or round.

That later media query is indeed circle/ellipse specific, but that's for a reason: wanting to opt into a rounded design on rounded screens for stylistic reasons is a clear demand from designers, but it does not generalize well to arbitrary shapes. If you're trying to accommodate the shape of the screen to avoid the corners, slide along the side of the screen, and similar, the other tools are more appropriate.

 - Florian