Re: [css-page-floats][css-multicol] column-span: <integer> (Was: [css-page-floats] First cleanup)

Hey

On Thu, Mar 26, 2015 at 10:55 AM, Florian Rivoal <florian@rivoal.net> wrote:

> On 26 Mar 2015, at 00:48, Johannes Wilm <johanneswilm@vivliostyle.com>
> wrote:
> >
> > Hey,
> > I have spent the last few weeks looking at the manuals of PrinceXML, the
> Antennahouse Formatter, Håkon Wium Lie's CSS Figure Spec [1] and have
> discussed with the Vivliostyle team what they would need from a page float
> spec.
>
> A bit tangent to what you'd like the review to be about, but you have also
> removed a bunch of sections which were not about page floats.
>
> When it comes to the removal of "Regions" and "Exclusions", since these
> were experimental counter proposals to the path the WG is following, I
> think this was appropriate.
>
> You also removed "column-span: <integer>". While this is arguably the
> wrong spec to host this, I believe this is actually something that should
> be preserved. It may have been underspecified, but the concept is sound and
> useful. (I am a lot less convinced about column-span: page).
>

Yes, I totally agree. The idea of column span makes a lot of sense. It just
seemed to be somewhat separate issue and not part of the page float
concept.


>
> I am not sure I can invest the time to be the editor of a new level of
> multi-col if it should cover more than this addition, but for this specific
> feature, I can put in the time needed to make sure it is specified properly
> and to respond to feedback.


> Should we restore it to the css-page-floats spec until a more appropriate
> spec to host it is ready?
>
> Should I start a multi-col level 2 (as a diff spec) for this feature?
>
>
I would say put this is in a new mult-col spec, but I am new to your guys'
work processes. If it's complicated/difficult to start a new spec that
requires a lot of time, we can temporarily add a section entitled something
like "unrelated specifications" at the end of the page float specification
until it can be moved to its own specification.


>  - Florian
>