W3C home > Mailing lists > Public > www-style@w3.org > January 2010

Re: Making pt a non-physical unit

From: Jonathan Kew <jonathan@jfkew.plus.com>
Date: Sun, 10 Jan 2010 23:24:09 +0000
Message-Id: <A9671722-2A8F-4200-B539-524697375DC7@jfkew.plus.com>
To: www-style W3C Group <www-style@w3.org>
On 9 Jan 2010, at 14:02, Christoph Pper wrote:

> Boris Zbarsky:
>> [physical units are] clearly nonsense for anything where you don't control the device;
> 
> Thats right.
> 
> That said, people may be using CSS Values in instances where they do know the environment perfectly (or good enough with Media Queries). Those guys could safely use absolute, physical units (i.e. mm, cm, in, pc and pt currently).
> 
> Everyone else should simply not use them. Yet some are using but one of them, namely pt. Noone (mis)uses the other four, so leave them as they are. Fix only the pt problem:
> 1. Redefine it in relation to px.
> 2. introduce a substitute for the empty space it leaves in physical units, e.g. pp.
> 
> Can we leave it with that?

No.

I don't believe it's true that "noone (mis)uses the other four". Yes, people probably don't use things like "mm" or "in" to specify font sizes; but what about other properties such as text-indent? A quick Google search for "text-indent .25in", for example, returns well over a million hits (this is of course searching page content, not stylesheets; but if that many people are writing examples online in this form, it's a fair bet that at least some stylesheets end up containing such units). If we need to redefine "pt" in order to get appropriately-sized text on extremely small (or large) displays, surely we need to do the same for such dimensions.

As I've said before, if we're going to change the basis of "pt", we need to change ALL the current "physical" units in the same way. Anything else is a recipe for even more confusion -- besides the risk of making CSS look somewhat ridiculous.

And then if we do need "more truly physical" units, create them as a clearly-distinct set.

JK
Received on Sunday, 10 January 2010 23:24:44 GMT

This archive was generated by hypermail 2.3.1 : Tuesday, 26 March 2013 17:20:23 GMT