Review of Track Issues for Best Practices (Part VII)

After a long gap, starting where I left off...

http://www.w3.org/International/track/issues/187


BP: Use U+hhhh syntax for Unicode code points. Be sure to always use at least 4 hex digits (up to six hex digits)
BP: Choose wisely the escape syntax for Unicode. Be aware of supplementary characters when using \uxxxx notation
BP: If you have a choice of escape mechanism, use one consistent with other specifications related to yours. If none is particularly relevant, \u{xxxx} variable width escaping is recommended.

http://www.w3.org/International/track/issues/188


// need to examine Eric's eventual response to this issue for IRI handling BPs in specs

http://www.w3.org/International/track/issues/189


BP: refer to obs-language-tag in BCP47 in preference to creating your own version

http://www.w3.org/International/track/issues/190

BP: Do not write requirements to remove all combining marks from text values.
// Needs further consideration. The use of PN_CHARS_BASE from XML probably merits further attention.

http://www.w3.org/International/track/issues/191


BP: When a specification is defined in terms of Unicode text (a sequence of Unicode code points), do not refer to the text as being "in UTF-8" or confusing the serialization scheme (character encoding scheme) UTF-8 with the fact that the file should be considered as a sequence of code points.

#192 refers to BP already...

http://www.w3.org/International/track/issues/193


BP: A specification must define when escapes are evaluated: before document processing or as part of the processing.

#194-#201 are a three year old comments on ruby; these should probably be evaluated in light of more recent developments

http://www.w3.org/International/track/issues/202

BP: use before/after and start/end rather than positional names (top/bottom; left/right) for properties. 
BP: positional names (see previous comment) may also be provided, but should not be the default and should have a health warning about text-direction and writing-mode.

http://www.w3.org/International/track/issues/204


// does the 'attirbute prefix' comment apply to other specs in addition to ITS?

[[ stopped here where the comments are about ITS]]


Addison Phillips
Globalization Architect (Amazon Lab126)
Chair (W3C I18N WG)

Internationalization is not a feature.
It is an architecture.

Received on Thursday, 28 May 2015 00:51:55 UTC