Greetings!
The agenda for our TC call on Monday, 2026-02-23
Time of meeting:
https://www.timeanddate.com/worldclock/meetingdetails.html?year=2026&month=2&day=23&hour=17&min=0&sec=0&p1=25&p2=37The call counts towards voter eligibility.
In a web browser, at meeting time, load the following URL:
https://meet.jit.si/moderated/4cae32628ad7bc11cebf38b633d7a7a66c7981240894785fb969a8d5d5fed9faChat room for meeting is at webconf.soaphub.org/conf/room/odf
ODF TC Holidays remaining:
April 6th - Easter Monday
May 4th - Early May Bank Holiday
May 25th - Memorial Day / Whit Monday / Spring Bank Holiday
August 31st - Summer Bank Holiday
September 7th - Labor Day
October 12th - Columbus Day
December 28 - Monday after Christmas
Please send comments to the mailing list.
1. Dial-In, Roll Call, Determination of Quorum and Voting Rights
2. Motion (simple majority): Approve ODF TC Minutes for February 09,
2026
https://groups.oasis-open.org/higherlogic/ws/public/document?document_id=73192&wg_id=4bf06d41-79ad-4c58-9e8e-018dc7d05da83. Action Items (from prior meetings)
A. weeknum function, Sebastian Schleich - Patrick - partially completed,
acknowledgement sent.
B. various comments, part 4,
arne@arne-thomassen.de - Patrick -
Acknoledgement sent
4. Motion (simple majority): Approve the Agenda
5.
arne@arne-thomassen.de has sent more comments (at end of agenda, I
have pasted them into Office-4189. Notice that JIRA can't be viewed by
the public??? end of his post, I'm pinging Kelly on it.
6. OpenFormula comments from Arne (cont.)
7. Public comment on wording in BIN2DEC and similar functions
https://issues.oasis-open.org/browse/OFFICE-41818. 6.14.8 LOOKUP -
https://issues.oasis-open.org/browse/OFFICE-4164https://groups.oasis-open.org/discussion/proposal-for-xlookup-and-xmatch-terms(proposal by Patrick)
9. We have a new webpage to review from OASIS.
https://www.oasis-open.org/tc-opendocument/Apparently we are going to have a public page and a workgroup page, or
at least it seems so to me.
BTW, Francis has spotted an amusing error on our new page. Do you see
it?
10. Adjournment
Hope everyone is at the start of a great week!
Patrick
*************More comments***********
the OpenFormula specification OpenDocument-v1.4-os-part4-formula.pdf
appears to have some flaws. I'd like to suggest the following changes to
individual functions in addition to the first batch from my e-mail on 19
January 2026:
- 3.5 "When recalculation occurs": the list of volatile functions
mentions RAND, but not RANDBETWEEN.
- 6.4.11 'Infix Operator Reference Range (":")': "cube" should be
"cuboid".
- 6.8.23 IMSECH: the given return type "Number" should be "Complex".
- 6.10.8 EASTERSUNDAY: the algorithm requires "an integer value for
Year", but does not mention the types of the variables from A1 to A7.
Should these be integers too (e.g. in C)? The algorithm rather looks
like using floating-point numbers throughout.
- 6.10.17 SECOND: an algorithm to calculate a number of seconds is
given. But the first line already calculates the value, and the further
lines provide a longer algorithm which apparently leads to the same
value. Is the description text missing a hint that these are two
different, separate algorithms which lead to the same result? Similarly
for the calcultion of minutes in 6.10.13 MINUTE.
- 6.13.3 CELL, "Table 17 - CELL": the short description for the
parameter value "PARENTHESES" seems to make no sense, especially the
word sequence "for positive or all values". Does that mean "for at least
one value"? The word sequence "format settings with parentheses" is also
unclear. None of the formats mentioned in the description of the
parameter value "FORMAT" have parentheses. Please extend the description
of the parameter value "PARENTHESES". Does the word "parentheses" in the
description mean the usual round ones () or rather the brackets []? The
description for the parameter value "FORMAT" calls the round parentheses
() "brackets". (Much more detail was provided for the parameter values
"FORMAT" and "PREFIX", and short but sufficient details were provided
for many others, but not for "PARENTHESES".)
- 6.13.13 INFO, "Table 18 - INFO": for the category "release", the
column "Meaning" requires: "version of the implementation"; but a mere
version number like "1.0" would probably be useless. Should the name of
the program be prepended? Should the separator between the name and the
version number be a space character; or e.g. a slash character "/" like
in HTTP User-Agent request header lines?
- 6.13.13 INFO, "Table 18 - INFO": for the category "origin", the column
"Meaning" requires: "The top leftmost visible cell's absolute
reference". What should happen when a user has several windows/views for
the same document/worksheet open and the "top leftmost visible" cells
are different because the user scrolled the worksheet to different
positions? An implementation cannot always discern for which of the
views the current formula is interpreted.
- 6.13.20 ISNA: the "Syntax" line says "ISERR" instead of "ISNA".
- 6.13.29 ROW: the return type "Number" should probably be
"Number|Array".
- 6.13.29 ROW: "an array of numbers is returned with all of the rows in
the reference"; the word "rows" should probably be "row numbers";
otherwise the numerical _contents_ of the cells of the rows would have
to be stored in the array.
- 6.13.30 ROWS: the notion "number of rows in a given range" is unclear.
If the cuboid encompasses several sheets, should implementations
multiply the number of rows in a sheet by the number of sheets? Or
should only the number of rows within one sheet be returned? (The word
"range" in the quoted text also seems to be wrong.)
- 6.13.33 TYPE: which value should be returned if the parameter results
in an empty cell? Probably 0? (Cf. 6.13.14 ISBLANK)
- In 6.13.33 TYPE and other places, a notion like "the reference is
first dereferenced" is used, but the meaning is not specified. Is
dereferencing the same as implied intersection? Or should dereferencing
attempts lead to an error when a reference encompasses more than one
cell?
- 6.14.2 ADDRESS: when the parameter "Text Sheet" is given but is an
empty string, should that be treated as an error?
- 6.16.54 SIGN: "if N = 0, returns 0." What should happen when N is
-0.0? (In C, should the test for N be iszero() or signbit()?)
- 6.18.50 MODE: "does no contain": "no" should be "not".
- 6.18.56 PEARSON: "are the averages of the given x, y data". The word
"averages" is unspecific. It probably refers to the _arithmetic_ mean?
- The notion "area" is not defined. Does it have the same meaning as
"cuboid" or "matrix" or something else? The "part3-schema.pdf" file
provides various definitions of "area" in different contexts, but none
seems to apply to "recalculated formula" areas.
Please consider these additional suggestions.
You mentioned JIRA in your reply. I'm aware of
<https: tools.oasis-open.org issues browse office>, but this web page
required a login. Would it be useful to make these JIRA pages publicly
readable for all interested people? - Maybe I could separate and
categorize my proposed changes to match the current ODF drafts and not
the possibly outdated ODF 1.4 section numbers.</https:>