97539 - the scaling factor of font-size: <absolute-size> should be 1.2 to be compatable with IE

Status: RESOLVED INVALID

(Core :: CSS Parsing and Computation, defect)

Description

[Frank Tang]

block bugscape bug 7108 and there fore block topembed bug 7970 
the reason the font size different between IE and n6.1 is because font-size:
<ABSLUTESIZE>

in CSS1: http://www.w3.org/TR/REC-CSS1#font-size
5.2.6    'font-size'

Value: <absolute-size> | <relative-size> | <length> | <percentage>
Initial: medium
Applies to: all elements
Inherited: yes
Percentage values: relative to parent element's font size

<absolute-size>
An <absolute-size> keyword is an index to a table of font sizes computed and
kept by the UA. Possible values are: [ xx-small | x-small | small | medium |
large | x-large | xx-large ]. On a computer screen a scaling factor of 1.5 is
suggested between adjacent indexes; if the 'medium' font is 10pt, the 'large'
font could be 15pt. Different media may need different scaling factors. Also,
the UA should take the quality and availability of fonts into account when
computing the table. The table may be different from one font family to another.


also

in CSS2 : http://www.w3.org/TR/REC-CSS2/fonts.html#font-size-props

15.2.4 Font size: the 'font-size' and 'font-size-adjust' properties

'font-size'
    Value:   <absolute-size> | <relative-size> | <length> | <percentage> | inherit
    Initial:   medium
    Applies to:   all elements
    Inherited:   yes, the computed value is inherited
    Percentages:   refer to parent element's font size
    Media:   visual

This property describes the size of the font when set solid. Values have the
following meanings:

<absolute-size>
    An <absolute-size> keyword refers to an entry in a table of font sizes
computed and kept by the user agent. Possible values are:

    [ xx-small | x-small | small | medium | large | x-large | xx-large ]

    On a computer screen a scaling factor of 1.2 is suggested between adjacent
indexes; if the 'medium' font is 12pt, the 'large' font could be 14.4pt.
Different media may need different scaling factors. Also, the user agent should
take the quality and availability of fonts into account when computing the
table. The table may be different from one font family to another.

    Note. In CSS1, the suggested scaling factor between adjacent indexes was 1.5
which user experience proved to be too large.



notice that the SUGGEST scaling facotr is 1.5 in CSS1 and 1.2 in CSS2. Since it
is SUGGEST , not REQUIRED factor, either (1.5 or 1.2) fullfill the CSS
specification.

It looks like IE use 1.2 and mozilla currently use 1.5 as scaling factor.

I think we should change back to 1.2 to be compababile with IE without violate
CSS2 (since it is only SUGGEST in CSS2, there are no specification violation issue)

Comment 1

[Frank Tang]

block bugscape 7108 and therefore 7970. mark as topembed add yokoyama to the cc
list.

Comment 2

[Frank Tang]

Attachment: test case attach

Comment 3

[Frank Tang]

Attachment: attach html again in text/html

Comment 4

[Frank Tang]

reproduce procedure-
1. load http://bugzilla.mozilla.org/showattachment.cgi?attach_id=47581 with
Netscape 6.1 rtm
2. load http://bugzilla.mozilla.org/showattachment.cgi?attach_id=47581 with IE
you can see the Netscape 6.1 use different font size than IE. 

We should make it the same size to be IE compatable. It won't violate the CSS2
(nor CSS1) specification since there are fix size or factor CSS2 defined.
CSS1/CSS2 only *suggest* factors and there are conflict *suggestion* between
CSS1 and CSS2 (CSS1 suggest 1.5 and CSS2 suggest 1.2)

Comment 5

[Pierre Saslawsky]

Bugscape #7108 doesn't seem to be related to Bugscape #7970:
- Per Roy Y. [2001-08-28 11:42], Bugscape #7970 is about a unitless length quirk.
- Per Scott M. [2001-08-22 16:44], Bugscape #7108 is about missing graphics.

What did you see in Bugscape #7108 that made you think it was related to scaling 
factors?  

The present bug is invalid: the scaling factor is already around 1.2. The exact 
values can be found at the following urls:
http://lxr.mozilla.org/seamonkey/source/content/shared/src/nsStyleUtil.cpp#376
http://lxr.mozilla.org/seamonkey/source/content/shared/src/nsStyleUtil.cpp#327
http://lxr.mozilla.org/seamonkey/source/content/shared/src/nsStyleUtil.cpp#298
See bug 18136 for more information.

Closed as Invalid: the description is incorrect and the bug doesn't appear to be 
related to Bugscape #7108 (missing graphics) nor Bugscape #7970 (missing quirk).

See the attached testcase which shows that the ratios are already around 1.2.

Comment 6

[Pierre Saslawsky]

Attachment: testcase showing that the ratios are already around 1.2

Comment 7

[Frank Tang]

reopen:
I know it is 1.2 right now (which CSS2 suggest not required), but I think we
should use 1.5 (which CSS1 suggest not required) so we can compatble with IE.

Comment 8

[Frank Tang]

show your test case in IE and you will see IE display differently. 
I think we should make mozilla display the same factor as IE since neither 1.2
nor 1.5 factor are REQUIRED by CSS1 or CSS2, 1.2 are SUGGESTED by CSS2 and 1.5
are SUGGESTED by CSS1. 

Are there anyway in CSS1/CSS2 to label CSS version number ? If no, I will
suggest we use 1.5 to maintain CSS1 compatability. If yes in CSS2, then I will
sguggest we use 1.5 as default for CSS1 compatability and treat any CSS label as
CSS2 use 1.2 

Comment 9

[Frank Tang]

I think we should compatable with IE at least in those area that CSS1/CSS2 does
not clearly defined. (For example, this one the specification only suggest, not
defined)

Comment 10

[Pierre Saslawsky]

Franck: as a general rule, in quirks mode, the first priority is to be compatible 
with Nav4 and the second priority is to be compatible with IE.  Only in rare 
cases will we reverse the priorities.

Comment 11

[Hixie (not reading bugmail)]

This bug is INVALID. A *huge* amount of work was put into choosing good values
for the absolute-size scaling. IE has MANY bugs with its font size scaling and
should NOT be imitated.

If there are specific sites that break because of this then please list them 
in this bug so that I may take a look at them.

In the worst case scenario that we absolutely must be compatible with IE for 
whatever contrived reasons we can come up with, then it should be limited to 
quirks mode. (I believe quirks mode is already a little quirky in this regard).