I'm not sure I understand the logic behind this argument.
It sounds like this is suggesting that we ignore the segment of users who adjust font sizes as an accessibility concern, because browser vendors also provide other methods of zooming?
Besides arbitrarily deciding a user segment with real needs and concerns who are adjusting font sizes isn't worth our time, this is questionable because using REM and EM units are functional in development beyond accessibility concerns (e.g. adjusting the size of a UI component based on its parent's size, etc.).
And since EM units are related to pixel values anyway, zooming functions that do change the reference pixel scale aren't really affected by their use.
Am I missing something?
In Safari 9 and Firefox 42, the "Zoom Text Only" options work the same whether you use
px- they appear to be scaling the calculated
font-sizeof every element as its displayed, rather than scaling the root element like the article describes IE as doing. I don't know how far back that behaviour goes or what IE does (I couldn't find a "Zoom Text Only" option in Chrome at all), but it wouldn't be difficult to find out.
I'm with you on the usefulness of
emin some cases, but I know there've also been times when I've been thinking in
pxin my head and it's been a real pain to have to multiply everything by my root font size; I think the point the article's making is that you no longer have to use
emin places where
pxbetter matches your mental model.
Enough already with these titles.
At least it isn't "REM considered harmful"!