OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.


Help: OASIS Mailing Lists Help | MarkMail Help

office message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]

Subject: OpenDocuemnt TC Coordination Call Minutes 2007-01-15

Yue Ma, IBM
Helen Yue, IBM
Robert Weir, IBM
Patric Durusau
David Faure, KDE
Jody Goldberg, Novell
Florian Reueter, Novell 
Michael Brauer, Sun Microsystems
Lars Oppermann, Sun Microsystems
Oliver Wittmann, Sun Microsystems
David A Wheeler, OpenDocument Foundation
Gary Edwards, OpenDocument Foundation

Minutes from Dec. 18th
The attending TC members unanimously accepted the minutes from the last
coordination call.

No pending action items

Michael: today is the 15th and the ODF 1.1 Ballot will open today.
- uploaded draft version for 1.2 on Friday
- adopted to ISO formatting rules
- easier to make changes now, notes and examples are marked up according 
to ISO rules and this is now handled by fields in the document
- also did some consistency checks and improvements
- didn't marl minor editorial changes explicitly as such
- changes that might be regarded at non-trivial are maked up with yellow 
background color
- especially schema changes
- everyone should review these changes, so that the markup can be 
removed and we can than use the marks for new things that we introduce 
with 1.2
- maybe we can aim for approving that draft version in two weeks and use 
that as a baseline
- next step will be splitting the document into main and package part
- will also get together with DavidW on adding formulas as the third part
- I will also start to integrate the agreed upon  proposals

Michael: For today we have a couple of proposals that we discussed 
during past work calls but that have no formal approval yet.
- I only got to update the wiki pages today which I will try to do 
earlier in the coming weeks
- so we only look into proposals that are out for some time today

Protection keys (4.4.3...):
- discussed before holiday break
- open issue: inclusion of attribute value describing MS excel algorithm
Michael: Jody, do you know whether there is a URL which can be used to 
refer to that algorithm
Jody: next meeting is in Febuary, we will discuss it there
Michael: there is no rush, as we can allow any URL there
Jody: we should have on in the standard anyway, otherwise implementors 
wouldn't know what to put
Rob: we corrently say SHA-1 is the default and it has been attacked 
several times
... maybe we should use SHA-256 as the default
Lars: a secure container even for this not secure feature, because the 
user might use a password that is used for other more sensitive data. 
Hence the possibility to break the protection key password would open 
other vectors of attack on the user's other data
Michael: the schema default does not have an impact on the application 
Patrick: is there any good reason for the schema default different from 
the proposed application default
Michael: this is the fall back to use when the algorithm is not 
specified in the digest attribute. So we would want to keep the default 
for backwards compatibility
DavidW: the attribute should be mandatory in 1.2 (you can also tell 
SHA256 from SHA1 by its length)
- but it is probably most sensible to leave it optional for old 
implementation but recommend to use SHA254 in the spec.

pp/ff in index example (
Michael: proposal is to change the text for combining and make clear, 
that the combination of index entries (when the index is updated) is 
application (language) dependent
- no objection

Numbered paragraphs and lists
Michael: we had two proposal which Oliver proposed to merge by adding 
Oliver: for numbered paragraphs, the current spec has nothing to say 
about the numbering domain
- with style override, application has the possibility to group ...
- also fulfills David's requirement of defining different styles on the 
same livel within a single list (e.g. 1., 2., C., D., 5...)
- style override is a better match for Lists
- addition is that by default all numbered paragraphs belong to one list
Florian, I don't want to have two different list models in ODF, I don#t 
think that the two ways of specifying lists (lists/num-para) are equal
Oliver: the current spec is somewhat imprecise
... I think was is indented when you have five numbered paragraph is 
that by default they belong to one list
.. I will post a description of how you can convert between the two ways 
of writing a list
David: Florian should post an example of a list where there is a problem 
with conversion between the two presentations
David: all of the numbered paragraphs are not always part of the same 
counter-domain. If you use a heading on a certain level, it sets up the 
counter domain for numbered paragraphs in that section...

Next call: TC Work call, 2007-01-22, 7am PT (3pm GMT).

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]