Filed Chromium issue to fix this before we ship. However Safari is shipping the scroll-snap-align with inline/block order so there is interop risk here.

Is there any indication that they will switch as well? otherwise we need to evaluate the interop risk here.

Originally posted here.

I ran a query against httparchive database to better understand the compat risk. My exact query is below if anyone wants to replicate [1].

Only 4 were using the scroll-snap-align with two arguments. This is out of 450k pages so usage and breakage is very low AFAICT. This gives me more confidence to make the switch. We should ensure Safari is willing to make the switch to avoid interop issues.

Here is the result:

Note that one of the cases is actually using it incorrectly.

[1] BigQuery query:

SELECT bodies.page, REGEXP_EXTRACT(bodies.body, r'(scroll-snap-align:\s*\w+?\s+\w+?\s*;)') as match, bodies.url, 
  FROM [httparchive:har.2018_02_01_chrome_requests_bodies]
  AS bodies
JOIN (
  SELECT 
    page, url,
    JSON_EXTRACT(payload, '$._contentType') AS contentType
  FROM [httparchive:har.2018_02_01_chrome_requests]
) AS requests ON requests.url=bodies.url AND requests.page=bodies.page
WHERE (requests.contentType CONTAINS 'html' OR requests.contentType CONTAINS 'css')
AND REGEXP_MATCH(body, r'(scroll-snap-align:\s*\w+?\s+\w+?\s*;)')
LIMIT 10000

@smfr do you have any opinion on this change one way or another? It seems a safe change to me so I plan to match the spec in Chromium implementation.

https://lists.w3.org/Archives/Public/www-style/2018Jul/0018.html

FWIW, I filed a webkit bug since Safari is still shipping the old ordering:

I've also added a note to the MDN docs for scroll-snap-align