W3C home > Mailing lists > Public > public-multilingualweb-lt@w3.org > November 2012

Re: action-283

From: Felix Sasaki <fsasaki@w3.org>
Date: Wed, 14 Nov 2012 02:18:38 +0100
Message-ID: <CAL58czoZ+n9j75QyKrMqiR5_iNfi+aM+VROUj6Y2LFdgotSBHQ@mail.gmail.com>
To: Dave Lewis <dave.lewis@cs.tcd.ie>
Cc: public-multilingualweb-lt@w3.org
Hi Dave, taking the list into the loop to get some thoughts,

I started editing provenance, but then stopped. Your changes make sense,
but they create a disalignment with other data categories, e.g.
localization quality issues has not the mutually exclusive version of
pointer attributes that you created. I also have problems to understand
nesting like these:

    Exactly one of the following:
            o    At least one of the following:
                          Exactly the following:

How about deleting all pointer attributes, except the pointer
"provRecsRefPointer" for standoff? This would make things easier and follow
the patterns Yves proposed for localization quality issue at

I'm not trying to re-open the general "too many global rules" issue - this
is specific to provenance and trying to avoid things like the "Exactly one
... " nesting above.



2012/11/13 Dave Lewis <dave.lewis@cs.tcd.ie>

> On 12/11/2012 22:47, Felix Sasaki wrote:
>> And another one - can you help me with this:
>> https://www.w3.org/**International/multilingualweb/**lt/track/actions/283<https://www.w3.org/International/multilingualweb/lt/track/actions/283>
>> Thanks,
>> Felix
> Hi Felix,
> thanks for reminding me, I'd not addressed that in the update to the
> provenance text i sent last night. i've added some text (hihglighted in
> green) in the attached update covering;
> - the warning of what the attributes in standoff relate to - copied from
> qualityissues
> - text on the use of script for containing provRecs standoff internally in
> I also fixed the text about the pointers attributes not being used in HTML.
> finally, after a bit of thought, i removed the note about the order of the
> provRec element within provRecs being significant in relation to temporal
> ordering. This information won't always be available but we don't have a
> mechanism for indicating when it is, so better i think just to drop it - it
> was just something i added, not really driven by a partner use case and the
> PROV records can provide this sort of detail if needed.
> cheers,
> Dave

Felix Sasaki
DFKI / W3C Fellow
Received on Wednesday, 14 November 2012 01:19:03 UTC

This archive was generated by hypermail 2.3.1 : Tuesday, 6 January 2015 21:08:24 UTC