Internationalization Teleconference

29 Jul 2009


See also: IRC log


+1.408.790.aaaa, aphillip, andrewc, Richard, +1.650.253.aabb, MarkD
DanC, Felix
Addison Phillips
Addison Phillips




<scribe> Scribe: Addison Phillips

<scribe> ScribeNick: aphillip


Action Items

Addison: supply text to WG of a suitable selectors normalization health warning

Richard: respond on our behalf to CSS on ruby issue

not yet

addison: set up document edit transition with dan for ws-i18n

not yet

richard: look at framework for i18n guidelines and suggest actions to take on document


obsoleted on beta page

not rewritten status and republished

<r12a> http://beta.w3.org/standards/techs/i18n

richard: publish i18n-guide-frameword as (retired, obsolete) WD

auto-rtl/auto-ltr for markup?


(and thread)

richard: changing base direction based on first character (not necessarily the alignment)

mark: basic issue is that if I type (say) Hebrew in a LTR context
... my results will be unitelligible because typing order is "funny"
... users are sometimes confused because UBA in ltr context is not convenient
... and vice versa (ltr input in rtl contexts)
... propsal is to have mechanism that adapts based on content
... mainly on user content
... link in email was to YouTube, where google uses Javascript to alter dir

richard: issue arises when you have bidi text with weak chars (such as punct) or unidirectional text with punctuation at edges


mark: happens when you have some text that matches page direction

richard: wouldn't solve latter problem
... still have "strange things" happening
... in bidi, things still "jump around"

mark: if I type hebrew1-dash-hebrew2

if ltr, you get jumping

richard: but if you have latin and hebrew

addison: maybe provide ways to contorl override better

(addison and mark discuss whether getting browser vendors to implement directinoal controls)

richard: my inclination is that sounds good for like form input
... but looks like something for implementation
... might be useful if writing a table
... but not want this for stuff like list items
... or <h3> or <p> etc.
... if we did use markup, rather than auto-ltr/auto-rtl
... just have 'auto' and that would would be allow content to determine

mark: principal benefit for input

richard: should w3c specificy?
... or default for implementations?

andrew: context when processing form?
... might lose context for directionality
... input some info into form with auto field
... so display right
... based on content
... but then the content is redisplayed badly


richard: perhaps on <input> or <form>
... provide some evidence of user requirement, not just annecdotal

mark: was originated by israeli team
... like addison's popup control suggestion
... will pass on for comment

(exuent mark and markus)

<r12a> @ i18n

Techniques or Best Practices


richard: so (fanfare)... might end up be a w3 discussion too
... best practices is a catchy title
... but are these really best practices
... the bidi might be close to them
... but language tagging is more like technique stuff
... and want to make changes to bidi
... for stuff like "if you find that say scroll bar moves then workaround is X"
... and that's more technique

(meant to mention that title and tooltip might want 'auto dir detect')

(when discussing with Mark, etc.)

(might be useful there....... but I digress)

scribe: so anyway, lots of "techniquey" stuff
... could call "tips" or "guidelines"...
...so: call it "Techniques" ?
... and then search/replace in document

andrew: yeah

addison: yeah

richard: will also introduce a few more techniques
... also matty had some feedback
... found way of getting current browser a bit problematic
... thinking of putting back stuff like IE and Opera switching scroll bar sides
... in a mroe generic way
... toying with solutions to that
... one thing that confused was link took to top of page unless there was something to be said
... and only one time is that the case

Working with Time Zones


richard: not totally clear on target of document
... who is the audience
... focused on datastructures

addison: current doc focused on xmlschema

richard: how far are we opening topic up?
... brief history should be first
... then what is a time zone
... and only then data structures, etc.
... when talking about utc offsets, distance from equator isn't strictly determiner
... some examples in original text were quite useful
... restore those?
... field-based time... the "field-based resprentation of computer time is more complicated" instead of just "the human based...."
... might need a section on "oh and by the way there is the way we write"

addison: as with my original document (but omitted here)

richard: wall time is never explained
... no one understands unless we define
... was written into the document
... if youR application (in past-and-future events)
... needs more examples
... guidelines for ambiguous forms, and calendars, and......
... lots of things could be done
... need to scope well
... looking forward to do/dont's
... say "incremental time" in section on "field based"
... it's a true statement about both


Summary of Action Items

[End of minutes]

Minutes formatted by David Booth's scribe.perl version 1.135 (CVS log)
$Date: 2009/07/29 20:08:27 $

Scribe.perl diagnostic output

[Delete this section before finalizing the minutes.]
This is scribe.perl Revision: 1.135  of Date: 2009/03/02 03:52:20  
Check for newer version at http://dev.w3.org/cvsweb/~checkout~/2002/scribe/

Guessing input format: RRSAgent_Text_Format (score 1.00)

Found Scribe: Addison Phillips
Found ScribeNick: aphillip
Default Present: +1.408.790.aaaa, aphillip, andrewc, Richard, +1.650.253.aabb, MarkD
Present: +1.408.790.aaaa aphillip andrewc Richard +1.650.253.aabb MarkD
Regrets: DanC Felix
Agenda: http://lists.w3.org/Archives/Member/member-i18n-core/2009Jul/0029.html
Got date from IRC log name: 29 Jul 2009
Guessing minutes URL: http://www.w3.org/2009/07/29-core-minutes.html
People with action items: 

[End of scribe.perl diagnostic output]