OASIS Open Document Format for Office Applications (OpenDocument) TC

 View Only
  • 1.  Re: [office] comparing requirementsagainst Thomas'/David's/Oliver'sproposal

    Posted 03-27-2007 09:14
    Hi Michael,
    
    === wrt. to questions regarding F1 ===
    
    We're talking about b).
    
    > What is with a). Is my assumption correct that you believe this 
    > requirement is met?
    
    > What is the case for an ODF 1.2 document that does not use the new the 
    > text:continue-list attribute, but only attributes that ODF 1.1 actually 
    > supports? Will it be displayed correctly in your opinion?
    
    This is very hard to state, since everytime I came up with an example I was told that ODF1.0/ODF1.1 isn't clear about
    this. This is where my F2 requirement comes from.
    
    So when I understood correctly you once summarized in a TC meeting that numbered-paragraph are not specified and that
    text:list are not specified when they are split.
    
    So you concluded that F1) is never an issue for numbered-paragraphs, since they where not specified at all.
    You also conluded that F1) is never an issue for text:list, since they also have specification holes.
    
    Regarding your question: 
    > Will it be displayed correctly in your opinion?
    
    No. It won't be displayed correctly in my opinion. However that would mean arguing about what ODF1.0/ODF1.1 actually
    meant. We decided not to do this.
    So the revised answer is: I have no idea, since it is totally unclear to me what ODF!.0/ODF1.1 says. The only thing I
    can do  is to look at applications and their interpretations. But that's req F2.
    
    === wrt. to questions regarding F4 ===
    
    > So, can you please explain to me why you think a screen reader would come to a different result.
    
    I guess the OOo screen reader will read whats displayed and thus the statement "it looks the same" is true. However
    readers which are not build upon the layout will come to a different result. 
    The concern came into my mind when reading the HTML spec. There is a sample regarding tables and screen reader and the
    screen reader here clearly operated on the XML structure.
    
    Basically (F4) goes far beyond screen readers. It demands that there is exactly one list concept in ODF. Which means
    that applications must not implement two different list concepts. And the list representations should be mapable 100%.
    
    Patrick for example was so kind to summarize this for me:
    http://www.oasis-open.org/archives/office/200703/msg00223.html.
    
    So when rethinking the while ODF1.2 lists this is a very fundamental requirement for me. 
    
    The are two ways to look at it:
    a) The lists share the same concept but have different rpresentations.
    b) The lists have different concepts but are somehow convertable into each other.
    
    I guess this is a TC decision, whether the TC want's a) or b).
    I clearly want a), but there other TC members see b) as sufficient.
    
    === wrt. to questions regarding F5 ===
    
    > I have to admit that I don't understand this. Why does it make a 
    > difference whether the start value is specified by the list style or the 
    > item itself. 
    Right. With Olivers proposal there are two ways to chance the counter of a list:
    a) using the start value specified by the list style 
    b) using the start value specified by the list item
    
    As a matter of fact a) causes problems wrt to my req F5. So I demand that we specify that we only offer b) to change the
    start value in ODF1.2.
    
    > The 


  • 2.  Re: [office] comparing requirements =?iso-8859-1?q?against=09Thomas=27/David=27s/Oliver=27sproposal?=

    Posted 03-27-2007 09:26
    On Tuesday 27 March 2007 11:13, Florian Reuter wrote:
    > However if the TC views lists in the ways that
    > a) The lists share the same concept but have different rpresentations.
    > then this is no option, since the 100% deterministic roundtrip is not
    > given.
    
    Florian,
    what purpose do the numbered-paragraphs have in ODF?
    
    Why do you think they were added?
    -- 
    Thomas Zander
    


  • 3.  Re: [office] comparing requirements againstThomas'/David's/Oliver'sproposal

    Posted 03-27-2007 12:35
    Hi Florian,
    
    please see my comments and questions inline.
    
    Regards, Oliver.
    
    Florian Reuter wrote:
    > Hi Michael,
    > 
    > === wrt. to questions regarding F1 ===
    > 
    > We're talking about b).
    > 
    >> What is with a). Is my assumption correct that you believe this 
    >> requirement is met?
    > 
    >> What is the case for an ODF 1.2 document that does not use the new the 
    >> text:continue-list attribute, but only attributes that ODF 1.1 actually 
    >> supports? Will it be displayed correctly in your opinion?
    > 
    > This is very hard to state, since everytime I came up with an example I was told that ODF1.0/ODF1.1 isn't clear about
    > this. This is where my F2 requirement comes from.
    > 
    > So when I understood correctly you once summarized in a TC meeting that numbered-paragraph are not specified and that
    > text:list are not specified when they are split.
    > 
    > So you concluded that F1) is never an issue for numbered-paragraphs, since they where not specified at all.
    > You also conluded that F1) is never an issue for text:list, since they also have specification holes.
    > 
    I didn't know any "specification holes" for text:list in the ODF1.1. I 
    think there is some stuff that can be stated clearer, but I didn't see 
    any of these "holes"
    Can you please name the "holes", which exist in your view of the ODF1.1?
    
    > Regarding your question: 
    >> Will it be displayed correctly in your opinion?
    > 
    > No. It won't be displayed correctly in my opinion. However that would mean arguing about what ODF1.0/ODF1.1 actually
    > meant. We decided not to do this.
    > So the revised answer is: I have no idea, since it is totally unclear to me what ODF!.0/ODF1.1 says. The only thing I
    > can do  is to look at applications and their interpretations. But that's req F2.
    > 
    > === wrt. to questions regarding F4 ===
    > 
    >> So, can you please explain to me why you think a screen reader would come to a different result.
    > 
    > I guess the OOo screen reader will read whats displayed and thus the statement "it looks the same" is true. However
    > readers which are not build upon the layout will come to a different result. 
    > The concern came into my mind when reading the HTML spec. There is a sample regarding tables and screen reader and the
    > screen reader here clearly operated on the XML structure.
    > 
    > Basically (F4) goes far beyond screen readers. It demands that there is exactly one list concept in ODF. Which means
    > that applications must not implement two different list concepts. And the list representations should be mapable 100%.
    > 
    > Patrick for example was so kind to summarize this for me:
    > http://www.oasis-open.org/archives/office/200703/msg00223.html.
    > 
    > So when rethinking the while ODF1.2 lists this is a very fundamental requirement for me. 
    > 
    > The are two ways to look at it:
    > a) The lists share the same concept but have different rpresentations.
    > b) The lists have different concepts but are somehow convertable into each other.
    > 
    > I guess this is a TC decision, whether the TC want's a) or b).
    > I clearly want a), but there other TC members see b) as sufficient.
    > 
    > === wrt. to questions regarding F5 ===
    > 
    >> I have to admit that I don't understand this. Why does it make a 
    >> difference whether the start value is specified by the list style or the 
    >> item itself. 
    > Right. With Olivers proposal there are two ways to chance the counter of a list:
    > a) using the start value specified by the list style 
    > b) using the start value specified by the list item
    > 
    In my view these two ways already exist in the ODF1.1 specification. As 
    I tried to illustrate in my previous post - see 
    http://www.oasis-open.org/archives/office/200703/msg00306.html
    
    
    > As a matter of fact a) causes problems wrt to my req F5. So I demand that we specify that we only offer b) to change the
    > start value in ODF1.2.
    
    Can you please name these problems?
    
    > 
    >> The 


  • 4.  Re: [office] comparing requirements againstThomas'/David's/Oliver'sproposal

    Posted 03-27-2007 14:15
    Hi Florian,
    
    Florian Reuter wrote:
    > Hi Michael,
    > 
    > === wrt. to questions regarding F1 ===
    > 
    > We're talking about b).
    > 
    >> What is with a). Is my assumption correct that you believe this 
    >> requirement is met?
    > 
    >> What is the case for an ODF 1.2 document that does not use the new the 
    >> text:continue-list attribute, but only attributes that ODF 1.1 actually 
    >> supports? Will it be displayed correctly in your opinion?
    > 
    > This is very hard to state, since everytime I came up with an example I was told that ODF1.0/ODF1.1 isn't clear about
    > this. This is where my F2 requirement comes from.
    > 
    > So when I understood correctly you once summarized in a TC meeting that numbered-paragraph are not specified and that
    > text:list are not specified when they are split.
    
    Sorry, but I cannot remember to have said this. What I can remember is 
    that I have said that the current ODF specification does not specify how 
    numbered paragraphs can be combined to lists (which for me personally 
    means that all numbered paragraphs form a single list), and that lists 
    do not have a feature to specify an individual list style for a single item.
    
    > 
    > So you concluded that F1) is never an issue for numbered-paragraphs, since they where not specified at all.
    > You also conluded that F1) is never an issue for text:list, since they also have specification holes.
    
    I cannot remember to ever have concluded one of these two things.
    
    > 
    > === wrt. to questions regarding F4 ===
    > 
    >> So, can you please explain to me why you think a screen reader would come to a different result.
    > 
    > I guess the OOo screen reader will read whats displayed and thus the statement "it looks the same" is true. However
    > readers which are not build upon the layout will come to a different result. 
    > The concern came into my mind when reading the HTML spec. There is a sample regarding tables and screen reader and the
    > screen reader here clearly operated on the XML structure.
    
    Sorry, but I still do nut understand that. The initial list and the 
    resulting list have the same structure, only the style assignments are 
    different.
    
    Michael
    
    -- 
    Michael Brauer, Technical Architect Software Engineering
    StarOffice/OpenOffice.org
    Sun Microsystems GmbH             Nagelsweg 55
    D-20097 Hamburg, Germany          michael.brauer@sun.com
    http://sun.com/staroffice         +49 40 23646 500
    http://blogs.sun.com/GullFOSS
    
    Sitz der Gesellschaft: Sun Microsystems GmbH, Sonnenallee 1,
    	   D-85551 Kirchheim-Heimstetten
    Amtsgericht Muenchen: HRB 161028
    Geschaeftsfuehrer: Marcel Schneider, Wolfgang Engels, Dr. Roland Boemer
    Vorsitzender des Aufsichtsrates: Martin Haering