W3C home > Mailing lists > Public > www-style@w3.org > April 2015

Re: [css-grid] Clarifying "all lines in the implicit grid" text (which probably wants to exclude explicit grid)

From: Daniel Holbert <dholbert@mozilla.com>
Date: Fri, 24 Apr 2015 15:50:00 -0700
To: Manuel Rego Casasnovas <rego@igalia.com>, www-style list <www-style@w3.org>
Cc: "Tab Atkins Jr." <jackalmage@gmail.com>, mats@mozilla.com
message-id: <553AC898.9020701@mozilla.com>
On 04/24/2015 02:43 PM, Manuel Rego Casasnovas wrote:
> Just to share my point of view, I interpret like the implicit grid only
> the tracks implicitly added, not all the tracks. From [3]:
> "While grid-template and its longhand properties define the explicit
> grid, grid items can be positioned outside of these bounds. This causes
> the grid container to generate implicit grid tracks, forming the
> implicit grid.
> So the "implicit grid tracks" are only the new ones, and not all the
> tracks of the grid (so IMHO it doesn't include the explicit grid).

I agree that "implicit grid tracks" is unambiguously defined.

However, the term "implicit grid" itself is ambiguous. In particular:
 (1) The spec sentence you quoted about "forming the implicit grid" is
ambiguous -- it can just as easily be interpreted as saying: "once the
grid container has added the implicit tracks, this larger structure is
the implicit grid."

 (2) There are at least two chunks that use "implicit grid" to
*unambiguously* mean the *full grid*:
  - Section 9.5 step 2 (mentioned in my previous post, RE "number of
  - Section 5.1.4 part b: "the number of tracks is determined by the
size of the subgrid’s implicit grid". This is talking about the number
of tracks in the *full subgrid*, not the number of implicit tracks.

 (3) Note that the area you're describing -- the area outside of the
explicit grid -- could be L-shaped.  A term like "implicit grid" seems
like a weird name for an L-shaped area, since "grid" implies something
rectangular, to me at least.

Anyway -- I think we agree that some or all of the language here needs


> Probably the wording in the auto-placement algorithm section should be
> improved to avoid this confusion.
> Regarding your question about the undefined named lines, you can check
> the a previous thread [4] in which we discussed about the expected behavior.
> Just my 2 cents,
>   Rego
> [3] http://dev.w3.org/csswg/css-grid/#implicit-grids
> [4] https://lists.w3.org/Archives/Public/www-style/2014Dec/0242.html
Received on Friday, 24 April 2015 22:50:31 UTC

This archive was generated by hypermail 2.4.0 : Monday, 23 January 2023 02:14:50 UTC