The Panose value is said to be used, "so that generating applications using this Office Open XML Standard may determine the closest font type if necessary". However, no font distance metric or font matching heuristic is described.

Describe the intended font matching procedure.

- Part 4, Section 5.1.12.37

te

Proposed Disposition of DIS 29500 Comment PT-0101 (Modified: 2008-01-13) Agreed; we will provide the following clarification around this concept in a new Part 4, §5.1.5.5: 5.1.5.5 Font Substitution If any DrawingML element references a font and an appropriate format of the font is not stored within the document, the process of finding a suitable alternative font is known as font substitution. The following elements specify font information: buFont@panose (§5.1.5.4.6); cs@panose (§5.1.5.3.1); ea@panose (§5.1.5.3.3); font@panose (§4.3.1.10); latin@panose (§5.1.5.3.7); sym@panose (§5.1.5.3.10). The exact algorithm which is used for font substitution is highly dependent on the characteristics which are most desirable when performing the substitution: similar appearance of each glyph (to maximize visual familiarity), similar physical characteristics (to minimize changes in line height and breaking), etc. This specification recommends that applications look for the closest match to the following attribute values (in descending priority) in order to determine a suitable alternative font: panose, charset, pitchFamily, and typeface; however, applications are free to apply higher-order logic in its place. As well, we note that the current text could be clearer by referring to the (now) appropriate ISO standard; accordingly, we will make updates the following locations throughout the spec, in order to improve interoperability: To Part 1, §3, page 5: ISO/IEC 14496-22:2007, Information technology ­ Coding of audio-visual objects ­ Part 22: Open Font Format. To Part 1, §15.2.12, page 154, line 12: application/x-font-ttf specifies that the font shall be stored in the TrueType or OpenType a format conforming to ISO/IEC 14496-22:2007 format . To Part 4, §2.8.2.13, page 756, lines 6­7: This element specifies the Panose-1 classification number for the current font using the PANOSE Classification Guide, Version 1.2 mechanism defined in §4.2.7.17 of ISO/IEC 14496-22:2007 . To Part 4, §2.8.2.16, page 758, lines 16­17: This element specifies information identifying the code pages and Unicode subranges for which the parent font provides glyphs using the mechanism defined in §4.2.7.18 and §4.2.7.28 of ISO/IEC 14496- 22:2007 . To Part 4, §5.4.3.1.10, page 2965, table row “panose”: panose (Panose Setting) Namespace: …/drawingml/2006/main Specifies the Panose-1 classification number for the current font using the mechanism defined in §4.2.7.17 of ISO/IEC 14496-22:2007 Specifies the panose standard setting that will be used to determine the closest matching font by any generating application that employs this method . The possible values for this attribute are defined by the ST_Panose simple type (§5.1.12.37). Part 4, §5.1.12.37, page 3,719, lines 12­13: 5.1.12.37 ST_Panose ( Panose Type Panose-1 Number ) This type specifies the Panose setting for this font so that generating applications using this Office Open XML Standard may determine the closest font type if necessary. This simple type specifies a value formatted according to the mechanism defined in §4.2.7.17 of ISO/IEC 14496-22:2007. [Example: Consider the following information stored for a single font: <w:font w:name="Times New Roman"> <w:panose1 w:val="02020603050405020304" /> … </w:font> The panose1 element specifies its Panose-1 number via its val attribute value of 02020603050405020304 . end example] Similar Comments: CO-0216 , DE-0099 , FR-0362 , GB-0313 , GR-0083 , US-0148 , VE-0066