Re: [office-formula] Groups -OpenDocument-part2-draft18-editor-revision(OpenDocument-part2-draft18-editor-revision.odt) uploaded

[Eike Rathke <erack@sun.com>]

Hi,

On Friday, 2010-01-29 21:22:26 +0000, Patrick Durusau wrote:

> Another draft of the OpenFormula document for your review!

Could everyone please take a look and communicate objections, if any,
until 2009-02-02 10:00 UTC? Thanks.

Btw, Patrick, please don't use the "Add New Document" button to upload
a new revision of a draft, instead use "Replace" for the "OpenDocument
Formula Specification ..." document that currently has 77 revisions.
This prevents a growing list of apparently different documents that in
fact are only different revisions. Thanks.


> 1) Evaluator (instead of application/implementation) language. (I did not
> replace language in the test cases/notes or in the portable language
> material that may be about to either go away or be re-worded anyway.)

Would had been fine in the portable language material as well, so it
doesn't get overlooked.

> 2) A number of comments that have Office-* links embedded in the text.
> Usually with suggested proposals for revision.

Currently these are now part of the normative text. Making them "TODO"
paragraphs might had been cleaner and easier to spot. Anyway, they are
searchable as "Office-[0-9]+", we wave to keep in mind they have to be
reviewed and removed.

> 3) Extensive revisions/comments in chapters 3 and 4. I am still unhappy
> with complex number and reference. 

There may also be a problem with removing entire blocks of paragraphs
and replacing them with nothing. I didn't check, but do we have any
"empty cell" definition now, for example?

Apart from that there's a problem with removing the word "typically".
For example, a sentence like

"A currency is a subtype of number which typically is displayed with
a currency symbol."

just said that. Removing typically there it suddenly becomes
a definition "A currency is a subtype of number which is displayed with
a currency symbol.", but that's not true, the actual display depends on
the number format assigned to the cell. The same with Percentage,
whether or not the number is multiplied by 100 and a percent sign is
displayed is not defined by the type.


> 5) I think I have caught all the must -> shall instances but need to make
> sure they are all true shall cases. 

Remark: all normative keywords (shall, should, ...) ought to have the
predefined "ISO Keyword" character attribution.

  Eike

-- 
Automatic string conversions considered dangerous. They are the GOTO statements
of spreadsheets.  --Robert Weir on the OpenDocument formula subcommittee's list.

PGP signature