W3C home > Mailing lists > Public > public-i18n-geo@w3.org > September 2003

Minutes from FTF, Atlanta

From: Richard Ishida <ishida@w3.org>
Date: Wed, 10 Sep 2003 14:24:42 +0100
To: "'GEO'" <public-i18n-geo@w3.org>
Message-ID: <002001c3779e$e55e81a0$10a554c7@w3c40upc3ma3j2>

These are the raw notes I took in instant messenger during the meeting
(used for video and audio link with Andrew).  For most of the meeting
the only participants on this IM session were myself and Andrew
Cunningham.  I have pruned out side conversations (eg. about audio or
webcam problems).

Attendees:
Richard Ishida (chair)
Martin Dürst
Russ Rolfe
Tex Texin
Lloyd Honomichl
Addison Philipps (came part way through)


richardishida: plan: look through document for gaping holes
richardishida: check that we don't have anything outright wrong
richardishida: check with andrew what he wants to do early on
richardishida: in terms of timing for publication this will become clear
as we progress today
richardishida: md: we should aim to publish asap after this meeting
richardishida: so we'll pull up the doc and start working through it
andj_c: i though about adding 6 just to show something of the future
richardishida: we'll look at 2-5 now, and then evaluate whether we have
time to do forms later
richardishida: we also have some stuff in 13, so we should check whether
that's ok to leave there
richardishida: should we include xhtml 1.1? if we include xhtml served
as xml
richardishida: md: lets say both because there's no real difference
richardishida: but does include ruby
richardishida: md: need intro to tell people how to use the doc and
mention this stuff
richardishida: need to add info about base browsers etc too
richardishida: md: could mask that out with style sheets
richardishida: tex, need to make clear that we've tested and what we
haven't
andj_c: application/xhtml+xml tends to be unreliable, focus on text/html
and add notes latter on application/xhtml+xml after testing
andj_c: what little testing i've done on application/xhtml+xml sort of
works in mozilla, although i still have trouble with style. IE no luck
todate
andj_c: just as long as the next ie has an uupdated usp10.dll
andj_c: :)
richardishida: some of the xhtml stuff served as xml may be theoretical
- we should make that clear
richardishida: we need to add to the intro information about server side
setup issues
andj_c: how to server docs could initially be an FAQ until we have built
up sufficient examples and notes
richardishida: in particular content type
richardishida: md: explain in the intro the structure and organisation
of the doc, including why material is repeated
richardishida: we're reading beginning of 2.1
richardishida: agreed, tone down the visibilty of editorial comments and
add something ed: before but keep them in
richardishida: 2.1 tech4 good header for previous points
richardishida: change content to editorial comment
richardishida: remove link one under impln guidelines
richardishida: make tbd ed note
andj_c: could add a link to the online ed of unicode4
andj_c: if memeory serves me, if an xhtml doc is served as text/html and
does not have an xml declaration, then the doc should be utf-8 or utf-16
richardishida: change reference to Unicode to v 4.0
andj_c: richard what was the url for the web standrads group doc on this
stuff?
richardishida: http://www.webstandards.org/learn/askw3c/sep2003.html
andj_c: have a look at http://www.w3.org/TR/xhtml1/#guidelines
andj_c: section C.1
andj_c: C.1. Processing Instructions and the XML Declaration

Be aware that processing instructions are rendered on some user agents.
Also, some user agents interpret the XML declaration to mean that the
document is unrecognized XML rather than HTML, and therefore may not
render the document as expected. For compatibility with these types of
legacy browsers, you may want to avoid using processing instructions and
XML declarations. Remember, however, that when the XML declaration is
not included in a document, the document can only use the default
character encodings UTF-8 or UTF-16.
andj_c: cascade the options maybe: 
richardishida: check the webstandards page to see if a pointer to it
will serve our purposes
andj_c: someone drive it and groups discussion via email
andj_c: what can i say, its after midnight :p
richardishida: action: go through 2.2 with a fine tooth comb to check
implications of mime type and declarations
andj_c: http://www.w3.org/TR/xhtml-media-types/
richardishida: action: martin to do this in damage limitation mode -
with an eye to faq
richardishida: action: ri, do spellcheck

Addison arrives

richardishida: nuke ed about serving as html vs xhtml
richardishida: change content type charset declaration to http charset
declaration
richardishida: remove "does the title element have to come before the
meta element..." it doesn't
richardishida: replace tbd tbd with link to hints and tips page
richardishida: 2.3 tech2, delete 'when there is one'
richardishida: also "when using escapes..."
richardishida: lowercase the SHOULDs
richardishida: this is highly discouraged -> use of private use
characters is highly...
andj_c: no sound
richardishida: add editorial note to say: use is ok for invisible
characters or others - "explain what the exceptions might be"
richardishida: nuke don't use entities in xhmtl?
richardishida: replace "something about inline characters" with ed note
richardishida: kill the ednote about title and meta position
richardishida: add to introduction that we have empty sections

LUNCH BREAK

richardishida: section 3
richardishida: very clearly state that  is not in the standard
richardishida: don't use font to represent other scripts - cheating on
charset
andj_c: don't use fonts esp since older versions of fonts may not
suppport target language
richardishida: md: add some text to say that these are not yet baked
andj_c: each font has different versions, the set of characters may
change between versions
richardishida: 3.1 'on the localized web page' -> 'on the client'
richardishida: make this a separate point "rely on text just fitting in
a space"
richardishida: 'on the localized page' -> also 'don't assume that the
font you've chosen will contain the characters needed for localized
pages"
richardishida: make last tech into an ednote
richardishida: don't assume that all versions of a font cover the same
characters
richardishida: andrew, happy with that?
andj_c: yep
richardishida: section 4.1
richardishida: split lang and xml:lang for html and xhtml
richardishida: 'declaring the language of a whole document" ->
'declaring the overall language for a doc" and explain that this is a
default
richardishida: declaring -> specifying
richardishida: section 4.2: 
richardishida: title: specifying (a different) language for part of a
document
richardishida: specify lang = html, xml:lang = xhtml
richardishida: section 4.3: demote technique to ednote
richardishida: add precedence rules
richardishida: section 4.4: 'wherever possible' only relates to 639 only
richardishida: also add to 'interoperability' that the standard says
this too!
richardishida: 5.1 'use css instead' -> 'do it from a LINKED stylesheet'
richardishida: make it clearer that this is not rtl and ltr
richardishida: 'you can have a different style sheet for a language'
richardishida: editorial comment: add mention that before and after
applies to css3
richardishida: sec5.2 make references to Opera consistent
richardishida: incorrect -> bad example
richardishida: update the example
richardishida: incorrect -> bad practise
richardishida: remove sentence "any user agent that supports Unicode
properly..."
richardishida: do something with the the arabic fonts
richardishida: section 5.3:
richardishida: reword 'a whole page'
richardishida: tweak the 'does not look right'
richardishida: fix text about visual orderingin arabic
richardishida: 5.5
richardishida: RLM and LRM -> expand
richardishida: remove ednote: I think XML ought to have meaningful....
andj_c: maybe a random link into parts of the document, rather than a
contents page
richardishida: red text: sometimes you can help the problem by adding
new lines in the code - though be careful about spaces

Russ leaves

richardishida: ** add a technique to say that you should test that the
text looks right in different widths - to ensure that ordering of
directional runs is ok
richardishida: ** might not be just window resize - could be affected by
font size changes
richardishida: ** change order of stuff about control codes for martin
richardishida: put the bidi space link in
richardishida: 5.6: 
richardishida: add explanation
richardishida: add that this is for character names
richardishida: section 13.1
richardishida: change four-digit to full, and remove always
richardishida: replace 'local calendars' by 'multiple calendars' twice
richardishida: change time and date to date and time
richardishida: change 'these' to by more explicity
richardishida: iso 8061 to iso 8601
richardishida: ** eventually refer to XML Schema
richardishida: redo description for sources
richardishida: ** add time zones 
richardishida: section 14.1
richardishida: add pointer to faq - 'see this'


============
Richard Ishida
W3C

contact info: http://www.w3.org/People/Ishida/ 

http://www.w3.org/International/ 
http://www.w3.org/International/geo/ 

See the W3C Internationalization FAQ page
http://www.w3.org/International/questions.html
Received on Wednesday, 10 September 2003 09:25:20 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 8 January 2008 14:12:37 GMT