[css-position] Clarification on effect of 'position: sticky' on table elements

A recent Chromium issue[1] revealed disagreement regarding the behavior of
position: sticky on table elements. The root of the disagreement is the
following wording within the CSS 3 positioning spec[2] for position: sticky:

"The effect of position: sticky on table elements is the same as for
position: relative"

The position: relative text is:

"The effect of position: relative on table elements is defined as follows
(emphasis mine):

  * table-row-group, table-header-group, table-footer-group and
table-row *offset
relative to its normal position within the table*. If table-cells span
multiple rows, only the cells originating in the relative positioned row is
offset."

Based on this text, an argument was put forward that for these table
elements position: sticky should offset not relative to its flow root, but
instead relative to its normal position within the table. The basis of this
argument is that the spec specifically says the effect is "the same" as
relative in this case, not "like relative but with respect to the flow
root".

My belief is that position: sticky should apply to table-row-group,
table-header-group, table-footer-group and table-row identically to how it
does for other elements; offset with respect to its flow root, not the
normal position.

I note that back in Oct 2014 there was a similar discussion[3] in which an
intention to update the wording was declared - but this seems not to have
happened? (Or the wording was updated but is still unclear?)

Thanks,
Stephen

[1] https://bugs.chromium.org/p/chromium/issues/detail?id=690896#c10
[2] https://www.w3.org/TR/css-position-3/#position-property
[3] https://lists.w3.org/Archives/Public/www-style/2014Oct/0301.html

Received on Thursday, 23 February 2017 12:49:12 UTC