Issue 17446: DTV Issue: A calendar day is not a time period (date-time-ftf) Source: NIST (Mr. Edward J. Barkmeyer, edbark(at)nist.gov) Nature: Uncategorized Issue Severity: Summary: Specification: Date Time Vocabulary Version: Beta-1 Title: A calendar day is not a time period Source: Ed Barkmeyer, NIST, edbark@nist.gov Summary: DTV section 9.5.3 defines 'calendar day', 'calendar month' and 'calendar year' to be time points, as shown in figure 9.10. The entry for 'day period', however, contains a Note that reads: "Calendar day is a period that starts and ends as defined by a calendar. Day period starts and ends at any time within a calendar day." And there are similar notes under 'month period' and 'year period'. A calendar day is not a '(time) period'; it is a 'time point'. But a 'day period' time interval can indeed start and end at any time point "within", i.e., on some time scale that subdivides, a calendar day. The Note could be modified to read: "A calendar day corresponds to time intervals that start and end as defined by a calendar." Or, the concept "calendar day period" could be introduced to refer to time intervals that instantiate calendar days. The juxtposition of the two otherwise unrelated sets of concepts in Figure 9.10 suggests that the latter may be what was intended. And in any case, the repair must be applied to year period and month period as well. It appears that a 'day period' is not just a time interval whose duration is one day, because of leap seconds. A note to that effect would be valuable. (It is clear that months and years are of variable duration.) Resolution: Revised Text: Actions taken: June 13, 2012: received issue Discussion: End of Annotations:===== te: Wed, 13 Jun 2012 19:12:50 -0400 From: Ed Barkmeyer Reply-To: "Barkmeyer, Edward J" , Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: "issues@omg.org" Subject: DTV Issue: A calendar day is not a time period X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: q5DNCtnf012499 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1340233976.72906@IkO0G+ZRWBNz+NaC1hoEuA X-Spam-Status: No Specification: Date Time Vocabulary Version: Beta-1 Title: A calendar day is not a time period Source: Ed Barkmeyer, NIST, edbark@nist.gov Summary: DTV section 9.5.3 defines 'calendar day', 'calendar month' and 'calendar year' to be time points, as shown in figure 9.10. The entry for 'day period', however, contains a Note that reads: "Calendar day is a period that starts and ends as defined by a calendar. Day period starts and ends at any time within a calendar day." And there are similar notes under 'month period' and 'year period'. A calendar day is not a '(time) period'; it is a 'time point'. But a 'day period' time interval can indeed start and end at any time point "within", i.e., on some time scale that subdivides, a calendar day. The Note could be modified to read: "A calendar day corresponds to time intervals that start and end as defined by a calendar." Or, the concept "calendar day period" could be introduced to refer to time intervals that instantiate calendar days. The juxtposition of the two otherwise unrelated sets of concepts in Figure 9.10 suggests that the latter may be what was intended. And in any case, the repair must be applied to year period and month period as well. It appears that a 'day period' is not just a time interval whose duration is one day, because of leap seconds. A note to that effect would be valuable. (It is clear that months and years are of variable duration.) -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 Date: Thu, 2 Aug 2012 19:02:55 -0400 From: Ed Barkmeyer Reply-To: Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: OMG DateTimeVoc FTF Subject: DTV Issue 17446 - A calendar day is not a time period - proposed resolution X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: q72N30tR025210 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1344553381.297@DSANxmq7jFQRgM5A+aKe5Q X-Spam-Status: No I attach a draft resolution for DTV Issue 17446 - A calendar day is not a time period. This is what I understood to be the directions from the conference call: rewrite the Notes to get the relationships right, and split the class diagram. -Ed -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 DTV Issue 17446 - a calendar day is not a time period-d1.docx Disposition: Resolved OMG Issue No: 17446 Title: A calendar day is not a time period Source: Ed Barkmeyer, NIST, edbark@nist.gov Summary: DTV section 9.5.3 defines 'calendar day', 'calendar month' and 'calendar year' to be time points, as shown in figure 9.10. The entry for 'day period', however, contains a Note that reads: "Calendar day is a period that starts and ends as defined by a calendar. Day period starts and ends at any time within a calendar day." And there are similar notes under 'month period' and 'year period'. A calendar day is not a '(time) period'; it is a 'time point'. But a 'day period' time interval can indeed start and end at any time point "within", i.e., on some time scale that subdivides, a calendar day. The Note could be modified to read: "A calendar day corresponds to time intervals that start and end as defined by a calendar." Or, the concept "calendar day period" could be introduced to refer to time intervals that instantiate calendar days. The juxtaposition of the two otherwise unrelated sets of concepts in Figure 9.10 suggests that the latter may be what was intended. And in any case, the repair must be applied to year period and month period as well. It appears that a 'day period' is not just a time interval whose duration is one day, because of leap seconds. A note to that effect would be valuable. (It is clear that months and years are of variable duration.) Resolution: The Notes are incorrect and are reworded, essentially as suggested. The FTF agrees that Figure 9.10 suggests a relationship between two sets of concepts that was not intended, and that two separate diagrams are wanted. The proposed Note about leap seconds is broadened to include any changes in time offset that affect the definition of local time of day on consecutive days. Revised Text: 1. In clause 9.5.3, REPLACE Figure 9.10 (Calendar time points and time periods) with: and DELETE the words "and time periods" from the caption, so that it reads: Figure 9.10 Calendar Time Points 2. In clause 9.5.3, immediately before the entry for 'year period', INSERT a new figure: Figure 9.11 Time periods based on calendars 3. In clause 9.5.3, in the entry for 'calendar year', REPLACE the Note: Note: Calendar year is a period that starts and ends as defined by a calendar. Year period starts and ends at any time within a calendar year. with: Note: A calendar year corresponds to time periods that start and end as defined by a calendar. A year period starts at any time within an instance of a calendar year. 4. In clause 9.5.3, in the entry for 'calendar month', REPLACE the Note: Note: Calendar month is a period that starts and ends as defined by a calendar. Month period starts and ends at any time within a calendar month. with: Note: A calendar month corresponds to time periods that start and end as defined by a calendar. A month period starts at any time within an instance of a calendar month. 5. In clause 9.5.3, in the entry for 'calendar year', REPLACE the Note: Note: Calendar day is a period that starts and ends as defined by a calendar. Day period starts and ends at any time within a calendar day. with: Note: A calendar day corresponds to time periods that start and end as defined by a calendar. A day period starts at any time of day within an instance of a calendar day. Note: A day period is defined by starting and ending at the same local time of day. When the local time of day is affected by a change of time offset between the starting and ending time intervals, the day period can have a duration that is not 24 hours. The duration of a month period or a year period may also be affected by changes in the time offset for the local time of day. Disposition: Resolved To: date-time-ftf@omg.org Subject: Re: DTV Issue 17446 - A calendar day is not a time period - proposed resolution X-KeepSent: 87C40983:93B0F07B-85257A4F:0044AB51; type=4; name=$KeepSent X-Mailer: Lotus Notes Release 8.5.3 September 15, 2011 From: Mark H Linehan Date: Fri, 3 Aug 2012 08:33:10 -0400 X-MIMETrack: Serialize by Router on D01ML604/01/M/IBM(Release 8.5.3HF266 | January 13, 2012) at 08/03/2012 08:33:13, Serialize complete at 08/03/2012 08:33:13 X-Content-Scanned: Fidelis XPS MAILER x-cbid: 12080312-5930-0000-0000-00000A7E758C This looks good to me. But I noticed what may be another issue, or alternatively could be handled in this issue: there is an inconsistency between figure 9.10 and the text. The figure shows 'calendar day' (etc.) as <> but the corresponding glossary entries are missing "Concept Type: concept type". I think the figure is right and the glossary entries should be updated. Do we agree? And should we do it in this issue? -------------------------------- Mark H. Linehan STSM, IBM Research From: Ed Barkmeyer To: OMG DateTimeVoc FTF , Date: 08/02/2012 07:04 PM Subject: DTV Issue 17446 - A calendar day is not a time period - proposed resolution -------------------------------------------------------------------------------- I attach a draft resolution for DTV Issue 17446 - A calendar day is not a time period. This is what I understood to be the directions from the conference call: rewrite the Notes to get the relationships right, and split the class diagram. -Ed -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 [attachment "DTV Issue 17446 - a calendar day is not a time period-d1.docx" deleted by Mark H Linehan/Watson/IBM] Date: Fri, 3 Aug 2012 14:54:12 -0400 From: Ed Barkmeyer Reply-To: Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: Mark H Linehan CC: "date-time-ftf@omg.org" Subject: Re: DTV Issue 17446 - A calendar day is not a time period - proposed resolution X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: q73IsIp2012264 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1344624861.35413@LHj0WPtLapb5ihoXqcn7XA X-Spam-Status: No I attach a revision of 17446 that adds "Concept Type: concept type" to each of the time point categories. -Ed Mark H Linehan wrote: This looks good to me. But I noticed what may be another issue, or alternatively could be handled in this issue: there is an inconsistency between figure 9.10 and the text. The figure shows 'calendar day' (etc.) as <> but the corresponding glossary entries are missing "Concept Type: concept type". I think the figure is right and the glossary entries should be updated. Do we agree? And should we do it in this issue? -------------------------------- Mark H. Linehan STSM, IBM Research From: Ed Barkmeyer To: OMG DateTimeVoc FTF , Date: 08/02/2012 07:04 PM Subject: DTV Issue 17446 - A calendar day is not a time period - proposed resolution -------------------------------------------------------------------------------- I attach a draft resolution for DTV Issue 17446 - A calendar day is not a time period. This is what I understood to be the directions from the conference call: rewrite the Notes to get the relationships right, and split the class diagram. -Ed -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 [attachment "DTV Issue 17446 - a calendar day is not a time period-d1.docx" deleted by Mark H Linehan/Watson/IBM] -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." DTV Issue 17446 - a calendar day is not a time period-d2.docx Disposition: Resolved OMG Issue No: 17446 Title: A calendar day is not a time period Source: Ed Barkmeyer, NIST, edbark@nist.gov Summary: DTV section 9.5.3 defines 'calendar day', 'calendar month' and 'calendar year' to be time points, as shown in figure 9.10. The entry for 'day period', however, contains a Note that reads: "Calendar day is a period that starts and ends as defined by a calendar. Day period starts and ends at any time within a calendar day." And there are similar notes under 'month period' and 'year period'. A calendar day is not a '(time) period'; it is a 'time point'. But a 'day period' time interval can indeed start and end at any time point "within", i.e., on some time scale that subdivides, a calendar day. The Note could be modified to read: "A calendar day corresponds to time intervals that start and end as defined by a calendar." Or, the concept "calendar day period" could be introduced to refer to time intervals that instantiate calendar days. The juxtaposition of the two otherwise unrelated sets of concepts in Figure 9.10 suggests that the latter may be what was intended. And in any case, the repair must be applied to year period and month period as well. It appears that a 'day period' is not just a time interval whose duration is one day, because of leap seconds. A note to that effect would be valuable. (It is clear that months and years are of variable duration.) Resolution: The Notes are incorrect and are reworded, essentially as suggested. The FTF agrees that Figure 9.10 suggests a relationship between two sets of concepts that was not intended, and that two separate diagrams are wanted. The proposed Note about leap seconds is broadened to include any changes in time offset that affect the definition of local time of day on consecutive days. Revised Text: 1. In clause 9.5.3, REPLACE Figure 9.10 (Calendar time points and time periods) with: and DELETE the words "and time periods" from the caption, so that it reads: Figure 9.10 Calendar Time Points 2. In clause 9.5.3, immediately before the entry for 'year period', INSERT a new figure: Figure 9.11 Time periods based on calendars 3. In clause 9.5.3, in the entry for 'calendar year', after the Definition, INSERT: Concept Type: concept type 4. In clause 9.5.3, in the entry for 'calendar year', REPLACE the Note: Note: Calendar year is a period that starts and ends as defined by a calendar. Year period starts and ends at any time within a calendar year. with: Note: A calendar year corresponds to time periods that start and end as defined by a calendar. A year period starts at any time within an instance of a calendar year. 5. In clause 9.5.3, in the entry for 'calendar month', after the Definition, INSERT: Concept Type: concept type 6. In clause 9.5.3, in the entry for 'calendar month', REPLACE the Note: Note: Calendar month is a period that starts and ends as defined by a calendar. Month period starts and ends at any time within a calendar month. with: Note: A calendar month corresponds to time periods that start and end as defined by a calendar. A month period starts at any time within an instance of a calendar month. 7. In clause 9.5.3, in the entry for 'calendar day', after the Definition, INSERT: Concept Type: concept type 8. In clause 9.5.3, in the entry for 'calendar day', REPLACE the Note: Note: Calendar day is a period that starts and ends as defined by a calendar. Day period starts and ends at any time within a calendar day. with: Note: A calendar day corresponds to time periods that start and end as defined by a calendar. A day period starts at any time of day within an instance of a calendar day. Note: A day period is defined by starting and ending at the same local time of day. When the local time of day is affected by a change of time offset between the starting and ending time intervals, the day period can have a duration that is not 24 hours. The duration of a month period or a year period may also be affected by changes in the time offset for the local time of day. Disposition: Resolved To: date-time-ftf@omg.org Subject: Re: DTV Issue 17446 - A calendar day is not a time period - proposed resolution X-KeepSent: CB52C233:DD173CB6-85257A4F:0068BD15; type=4; name=$KeepSent X-Mailer: Lotus Notes Release 8.5.3 September 15, 2011 From: Mark H Linehan Date: Fri, 3 Aug 2012 15:51:32 -0400 X-MIMETrack: Serialize by Router on D01MC604/01/M/IBM(Release 8.5.3 ZX853HP5|January 12, 2012) at 08/03/2012 15:51:33, Serialize complete at 08/03/2012 15:51:33 X-Content-Scanned: Fidelis XPS MAILER x-cbid: 12080319-3534-0000-0000-00000B178BB2 Thanks. But looking further, I see the <> stereotype used for a number of other kinds of time points, but not present in the text. Not just for 'time of day' and 'calendar week' (which are in figure 9.10 but not in this revised resolution) but also in figure 9.12, 'Gregorian year' and several others. Which makes me think again about the criteria for when these should be marked <>. It seems to me that a concept is a concept type when its extension contains concepts. By that criteria, all of these are concept types, but there are more. Such as 'day of week' (which has 'Monday', 'Tuesday', etc. in its extension). Net: I wonder if we should deal with this whole topic in a separate issue. -------------------------------- Mark H. Linehan STSM, IBM Research From: Ed Barkmeyer To: Mark H Linehan/Watson/IBM@IBMUS, Cc: "date-time-ftf@omg.org" Date: 08/03/2012 02:55 PM Subject: Re: DTV Issue 17446 - A calendar day is not a time period - proposed resolution -------------------------------------------------------------------------------- I attach a revision of 17446 that adds "Concept Type: concept type" to each of the time point categories. -Ed Mark H Linehan wrote: This looks good to me. But I noticed what may be another issue, or alternatively could be handled in this issue: there is an inconsistency between figure 9.10 and the text. The figure shows 'calendar day' (etc.) as <> but the corresponding glossary entries are missing "Concept Type: concept type". I think the figure is right and the glossary entries should be updated. Do we agree? And should we do it in this issue? -------------------------------- Mark H. Linehan STSM, IBM Research From: Ed Barkmeyer To: OMG DateTimeVoc FTF , Date: 08/02/2012 07:04 PM Subject: DTV Issue 17446 - A calendar day is not a time period - proposed resolution -------------------------------------------------------------------------------- I attach a draft resolution for DTV Issue 17446 - A calendar day is not a time period. This is what I understood to be the directions from the conference call: rewrite the Notes to get the relationships right, and split the class diagram. -Ed -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 [attachment "DTV Issue 17446 - a calendar day is not a time period-d1.docx" deleted by Mark H Linehan/Watson/IBM] -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [attachment "DTV Issue 17446 - a calendar day is not a time period-d2.docx" deleted by Mark H Linehan/Watson/IBM] Date: Fri, 3 Aug 2012 18:39:36 -0400 From: Ed Barkmeyer Reply-To: Organization: NIST User-Agent: Thunderbird 2.0.0.23 (Windows/20090812) To: Mark H Linehan CC: "date-time-ftf@omg.org" Subject: Re: DTV Issue 17446 - A calendar day is not a time period - proposed resolution X-NISTMEL-MailScanner-Information: Please contact postmaster@mel.nist.gov for more information X-NISTMEL-MailScanner-ID: q73Mdg6b029828 X-NISTMEL-MailScanner: Found to be clean X-NISTMEL-MailScanner-SpamCheck: X-NISTMEL-MailScanner-From: edbark@nist.gov X-NISTMEL-MailScanner-Watermark: 1344638383.51276@tDpGX8KelBw3BLOSORrvfw X-Spam-Status: No Mark H Linehan wrote: Thanks. But looking further, I see the <> stereotype used for a number of other kinds of time points, but not present in the text. Not just for 'time of day' and 'calendar week' (which are in figure 9.10 but not in this revised resolution) but also in figure 9.12, 'Gregorian year' and several others. Which makes me think again about the criteria for when these should be marked <>. It seems to me that a concept is a concept type when its extension contains concepts. By that criteria, all of these are concept types, but there are more. Such as 'day of week' (which has 'Monday', 'Tuesday', etc. in its extension). Net: I wonder if we should deal with this whole topic in a separate issue. Well, yes. But these changes sort of fit with the observation that 'calendar day' is a category of time points, not intervals. Do you want to just go back to draft 1? Either way, we should be done. Just pick one. -Ed -------------------------------- Mark H. Linehan STSM, IBM Research From: Ed Barkmeyer To: Mark H Linehan/Watson/IBM@IBMUS, Cc: "date-time-ftf@omg.org" Date: 08/03/2012 02:55 PM Subject: Re: DTV Issue 17446 - A calendar day is not a time period - proposed resolution ------------------------------------------------------------------------ I attach a revision of 17446 that adds "Concept Type: concept type" to each of the time point categories. -Ed Mark H Linehan wrote: This looks good to me. But I noticed what may be another issue, or alternatively could be handled in this issue: there is an inconsistency between figure 9.10 and the text. The figure shows 'calendar day' (etc.) as <> but the corresponding glossary entries are missing "Concept Type: concept type". I think the figure is right and the glossary entries should be updated. Do we agree? And should we do it in this issue? -------------------------------- Mark H. Linehan STSM, IBM Research From: Ed Barkmeyer __ To: OMG DateTimeVoc FTF __ , Date: 08/02/2012 07:04 PM Subject: DTV Issue 17446 - A calendar day is not a time period - proposed resolution ------------------------------------------------------------------------ I attach a draft resolution for DTV Issue 17446 - A calendar day is not a time period. This is what I understood to be the directions from the conference call: rewrite the Notes to get the relationships right, and split the class diagram. -Ed -- Edward J. Barkmeyer Email: _edbark@nist.gov_ National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 [attachment "DTV Issue 17446 - a calendar day is not a time period-d1.docx" deleted by Mark H Linehan/Watson/IBM] -- Edward J. Barkmeyer Email: _edbark@nist.gov_ National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800 "The opinions expressed above do not reflect consensus of NIST, and have not been reviewed by any Government authority." [attachment "DTV Issue 17446 - a calendar day is not a time period-d2.docx" deleted by Mark H Linehan/Watson/IBM] -- Edward J. Barkmeyer Email: edbark@nist.gov National Institute of Standards & Technology Manufacturing Systems Integration Division 100 Bureau Drive, Stop 8263 Tel: +1 301-975-3528 Gaithersburg, MD 20899-8263 Cel: +1 240-672-5800