Re: The "resolution" media query is a misnomer IMHO

[Most of this message (before the last few paragraphs) can be considered
 off-topic except to the extent that it's relevant to how acceptable the
 existing name is.]

On Thu, Feb 06, 2014 at 01:55:11PM +1100, Behrang Saeedzadeh wrote:
> This is strange. Before Apple's introduction of Retina displays I can't
> remember that many, if any at all, discussions about pixel density of
> screens.

The first such discussions I recall used the word "resolution".  I didn't hear
the phrase "pixel density", or the word "resolution" in a way inconsistent with
"density", until some years later.

> Any while in the context of images it might refer to pixel
> density, here this media query queries the pixel density of the screen.
> 
> Anyway, it is a misnomer to me, but I can live with it I guess.

If it helps: it is a misnomer in the sense that the meaning you describe is now
the more common meaning of "resolution" in the context of displays, but at the
same time, the density meaning is still one of the meanings of "resolution"
even in the context of displays, and the meaning you describe is more of a
misnomer in the sense that it is inconsistent with how "resolution" is used of
instruments outside of displays and cameras.

Traditionally, it's a measure of the ability to _resolve_ fine details (whether
visual, audio, or other sensors), and is a measure of measurement density or
(in telescopes, microscopes or lenses) lack of blurriness.

Hence, in bitmapped printers and monitors, the term was used to mean pixel
density.  I encountered this meaning some years before the first time I saw it
used in the sense you describe, and this meaning is still in use today (as a
web search for ‘resolution’ and anything like dpi or ppi or 96dpi etc. will
attest).  To my knowledge, "resolution" still always relates to density in
printers and scanners (as distinct from cameras).

In some cases, such as a bitmap image of a particular scene or object,
or when telling one's operating system's display software to switch to a
different resolution for a given display, different resolutions correspond
to different numbers of pixels; and in fact the most natural way for software
to refer to a given display resolution is by the number of pixels, since the
software doesn't necessarily know the physical size of the display.  Perhaps
because of this, the term has come to be understood as the number of pixels
rather than their density or the extent to which the display allows resolving
fine detail.

I don't know when this meaning became common; I'd guess some time during the
1990s (which would match what http://en.wikipedia.org/wiki/Multisync_monitor
says about when monitors that could change resolution became common).


> So, "resolution" in this context, looks like a misnomer and I believe it
> should be deprecated in favor of a better name (e.g. "pixel-density").

Renaming of existing well-established identifiers is not without precedent, 
but I think it's fair to say that such renames aren't universally welcomed.

If you still think that it would be better if there were a different name,
and would prefer if you could convince the relevant people of this:
although I've no say myself, my perception is that a rename is more likely to
be accepted if there's evidence of the existing name causing authors extra
debugging time due to being misled by the existing name.  More generally,
I think it would help to show that the benefit to the Web would outweigh the
cost of changing the specification, test suite and implementations.

Hope that helps,

pjrm.

Received on Thursday, 6 February 2014 05:06:27 UTC