The following notes were taken by R. Joyner. Corrections are welcome. IN PROGRESS
Planetary Data System Standards Teleconference
28 February 2007 9-10 AM
Announcements:
Elizabeth gave birth to a baby girl yesterday, 2007-02-27. Both mommy and baby are doing fine (unconfirmed).
Abigail Dorothy Rye
Weight: 7lbs and 3 oz
Length: 19 inches
Birthday: February 27, 2007
1. Standards approval discussion (D.Simpson)
- D.Simpson distributed a marked-up version of a Standards policy from 2002. In the document, he has interleaved a number of questions regarding operation of the Standards process (as proposed in 2002). D.Simpson indicated that the policy posted at the EN site was authored by J.Wilf in 2005; although it claims to have incorporated comments from others, there seems to have been little connection between the 2002 effort and the version that appeared in 2005. One of the issues not addressed in either version is the criterion for approval at the Tech Session -- the narrower question shown on today's agenda. D.Simpson asked for thoughts on how the Tech Session wants to address this. The Management Council approval procedure requires approximately a two-thirds vote, but a certain mix of votes is required. The Tech Session membership is different, but a similar procedure could be drafted. Another option would be a simple majority, but that risks a higher rejection rate by MC. There are other possibilities.
- A.Raugh indicated that there is a statement in the document posted at the EN that concerned her (and the statement is not in the document that D.Simpson distributed). The Systems Engineer (SE) can unilaterally decide that a candidate statement is ready for MC vote.
- D.Simpson indicated that there is also a Keyword Process document posted at the EN site that may not include the comments that were sent in.
- A.Raugh asked what is the starting point for these discusssions. D.Simpson suggested that we use the document posted at the EN.
- M.Gordon commented that the Standards document is simply out of date. Pick a small group to create a current document.
- D.Simpson, A.Raugh, M.Gordon volunteered to form a WG to review the standards process and update the document.
2. SCR 3-1103.v3: Chapter 7: Date Time Format (A.Raugh / T.King)
- A.Raugh indicated that she is ready to incorporate the specific comment written by T.King. However, she would like to add the comment to section 7.3.2 (instead of 7.2). She thinks that the comment is more applicable to section 7.3.2 (under Native times).
- L.Huber asked if this statement isn't already in section 7.3.1; why are we repeating it?
- A.Raugh offered to reword the paragraph to suggest that using Native format is not the preferred format (except in rare instances or a specific exception).
- A.Raugh will send out the current wording and the proposed re-wording for comment.
- A.Raugh will add an annotation to indicate there are special cases which allow a 24th hour or a 60th second (e.g., leap seconds).
- The group agreed to an e-vote that will end within 1 week.
3. SCR 3-1104.v1: Increase length of START_TIME and STOP_TIME values (A.Raugh)
- A.Raugh indicated that the first line in the SCR is incorrect; it should be 23 characters (instead of the stated 24). She also argued that this SCR does not have low impact. These keywords are required in every data product label, are searchable across the entire PDS system, and are part of the external interface to other archives. Approval of the SCR will have a major impact on SBN validators, verifiers, and other tools. Finally, allowing microsecond precision is very close to being ridiculous; that kind of precision can't be justified for a label keyword.
- S.Slavney countered that the increase in precision should be supported because data providers want it. It is not our business to overrule their decisions regarding necessity.
- A.Raugh suggested that the data providers could use Native time instead of the START_TIME and STOP_TIME keywords.
- M.Gordon agreed that the keywords are important for search and for validation. With respect to the search issue, the point is that the increase in precision will not adversely affect searching. Validation may be affected -- specifically VTOOL may be affected.
- S.Slavney indicated that the consequence of not approving this is that the PDS looks very "....".
- M.Gordon proposed that search and validation are two separate issues which should be addressed separately.
- L.Huber also agreed that the SCR understates the impact; but is there really more involved than changing the length of the keyword?
- A.Raugh noted that the SCR does not explicitly identify what has (and has not) been checked for impacts. There may be documentation that will need to be updated -- for example, the PAG and APG.
- M.Gordon asked if the understated impacts are sufficiently significant to stop the SCR from moving forward.
- The group agreed to an e-vote that will end within 1 week.
4. SCR 3-1064.v2: Increase length of PATH_NAME (A.Raugh)
- R.Joyner indicated that this SCR was rejected. Does the group want to propose amendments or let it die?
- A.Raugh volunteered to summarize the effect of the ISO standard on the file specification.
- T.King suggested that the PDS should not be overly concerned about this in moving forward as the PDS is moving away from hard media (at least until the PDS decides what the archive media will be).
- A.Raugh indicated that the 277 character limit should not include the 31 characters required for the filename -- that the number could safely be increased to 255 minus 32, or 233.
- L.Huber suggested that we amend the SCR to indicate 233 characters and vote.
- D.Simpson stated that he doesn't vote on SCR's that don't exist.
- R.Joyner will email A.Raugh the current version so she can make the changes.
- A.Raugh agreed to re-write and submit v.3 for e-vote (also within 1 week).
5. SCR 3-1105.v1: Change BAND_NAME definition to include non-spectral bands (S.Slavney)
- T.King pre-voted: yes
- S.Slavney agreed with A.Raugh's assessment of the previous SCR and submitted this new SCR which includes underlined text to indicate the sections that have changed.
- A.Raugh asked why the BAND_CENTER is still constrained to SPECTRAL contexts.
- S.Slavney indicated that she will make that one generic.
- A.Raugh asked if the MARS ODYSSEY reference is significant.
- P.Garcia indicated that it is not and that it is OKAY to remove it.
- S.Slavney agreed to re-write and submit v.2 for e-vote (also within 1 week).
6. Next Round of E-Votes
- P.Garcia suggested that the e-vote end at noon on the Tuesday preceeding the next Standards telecon so that the Node votes can be discussed, if that is necessary.
7. Second round of "Find the no-brainers'
- T.King agreed to review the list and identify approximately six for consideration at the next telecon (which will be held next we
Agenda for next telecon (20070307):
1. Report the Node votes on the following re-written SCRs and discuss if there is interest:
- SCR 3-1103.v4: Chapter 7: Date Time Format (A.Raugh / T.King)
- SCR 3-1104.v1: Increase length of START_TIME & STOP_TIME keywords (A.Raugh)
- SCR 3-1064.v3: Increase length of PATH_NAME (A.Raugh)
- SCR 3-1105.v2: Change BAND_NAME definition to include non-spectral bands (S.Slavney)
2. Discuss the Node votes on the following SCRs (from the previous round of e-votes):
- SCR 3-1071.v1 - Use of format effectors within attribute assignment statements (E.Rye 2007-07-03)
- SCR 3-1083.v2 - Increase length of SOURCE_PRODUCT_ID keyword
- SCR 3-1070.v1 - ODL lines and records
3. Discuss the T.King list of additional '2nd round of no-brainers'
4. Discuss SCR 3-1073.v1: Update ARCHIVE_STATUS Keyword (B.Sword)