|
I believe the resolution of this issue goes too far. There are two concerns:
1. The attribute is optional, but we do not say what the default behavior is. I suspect it is "false." 2. We say more than the original about what it means for the variable to not be fixed. Something closer to the original language of ODF 1.0 would be the following: """ The defined values of the text:fixed attribute are: * true: value of the field element where this attribute appears is preserved in all future edits of the document where the field element is retained. * false: (default) the value of the field element may be replaced by a new value when the document is read or updated """ This may say too much also, but editing is actually too narrow. Also, in the errata document, text:fixed, true, and false must appear in a monospace (Courier New) font.
Suggestion:
true: value of the field element where this attribute appears is preserved. false: value of the field element where this attribute appears may be changed. No default because this is an optional attribute, that is it only has an impact if present and we define that impact here. Note that "false" only means that it may be changed. I could be editing some other field and despite that edit, another field remains the same. Note that "true" for a dynamic field, such as date (you have seen the various drafts with the same incorrect dates?), means that on reload the date does not change. If False, it would. BTW, there are no "future edits." There are only edits that are occurring while the value of this attribute, if present, is true or false. Thanks to Dennis for spurring this line of thinking! I see that "Needs Discussion" is removed but it is not clear that there has been discussion toward resolution.
Clearly there is a default fact with regard to whether a text field is fixed or not? What other state can a text field be in? [I have no idea whether there is an inter-dependency with whether the field can be protected or not, although that is something certainly worth discussing.] Oh, and "may be changed" is too vague. |
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
|
Bug
Applied
Major
ISO/IEC JTC 1/SC 34 N 1078 : DEFECT REPORT NUMBER JP2-2
That is to say, if I set the value of a field, set this attribute to "true," edit the file multiple times, the value remains the same. If I then give the file to Rob, if he resets the value of this attribute to "false," his edits of the text, which are "future edits," then can result in changes to that field.
Suggestions welcome on how to re-word the suggested prose.