Address 100 Barr Harbor Drive Phone 610.832.9500
PO Box C700 Fax 610.832.9555
W. Conshohocken, PA e-mail service@astm.org
19428-2959 | USA Web www.astm.org
Committee E13 on MOLECULAR SPECTROSCOPY AND CHROMATOGRAPHY
DRAFT Minutes for E13.15 Subcommittee Working Group DRAFT
10:35 am – 12:12 pm EDT
Friday, August 25, 2005
Virtual Meeting
I. Introductions and Welcome: Gary Kramer, NIST, E13.15 Chair called the meeting to order at 10:36
am EDT. Karen Wilson was introduced as the official ASTM liaison.
II. Review of Agenda and Minutes. The agenda was approved without change.
A motion was made to approve the minutes for the August 4, 2005 meeting as circulated. Second. The
motion passed unanimously.
III. Attendees:
Michael Boruta, ACD Peter Linstrom, NIST
Stuart Chalk, Univerity of North Florida Dave Martinsen, ACS
Tony Davies, Waters Mark Mullins, SSI
Bob Dragoset, NIST Anh Dao Nguyen, NIST
Maren Fiege, Waters Burkhard Schäfer, Univ. Kaiserlautern
David Hogenson, EPM Ed Svastits, CH2M
Gary Kramer, NIST Karen Wilson, ASTM
IV. Location of Schemas. Mark Bean is not present today. There is nothing to report on the location of
schemas.
Burkhard Schäfer presented a discussion on the requirements for a repository for schemas and technique
definitions, considering how to best distribute the AnIML core and technique schemas and documents.
The powerpoint slides are available on the AnIML web site.
At this time, ASTM does not have the capability to make parts of the standard publicly available in
machine-readable form prior to approval as a standard, or to make parts of the standard available
without access control. In order for AnIML to work, vendors and users need to be able to get to the
schema files. The best solution is to have a central location or repository where these could be stored.
The problem is that there are two kinds of validation, syntactic (an AnIML document or technique
definition is valid according to the schemas) and semantic (all data defined in the core and technique
files have the correct data, and the required information is there). Using a fixed location allows files
with locations of schema and technique definitions and extensions hard-coded to be transferred between
servers without the need to change the locations. The problem with a fixed location is that it is hard to
guarantee its availability.
A fixed URL namespace would allow us to move away from a fixed location, but allow the repository
to be set up at one location now, and then moved it to a more appropriate location in the future. As long
as the url remains the same, the actual location of the files may change. Burkhard noted several domains
or subdomains that could be updated as needed to point to the repository location. In addition, multiple
hosts could be configured to handle the same url.
Discussion concerned the need for a fixed and permanent location. In practical terms, it means that
applications won’t work unless one is connected to the Internet, and most vendors will implement a
local copy of the files. Burkhard noted that when you open a document, an application will try to
download the referenced schemas, and fall back to a local copy if the referenced version cannot be
accessed. It was noted that mirror sites might also be desirable, with vendors possibly getting involved
in supporting mirror sites.
There are some open questions. Should schemas and technique definitions be stored in the same
location? Technique documents will be updated and extended more frequently than the core. Other
groups may set up techniques independent of ASTM E13.15. Will only official documents be allowed,
or would these independent techniques also be allowed? It was noted that ASTM and/or IUPAC would
probably not allow 3rd party techniques which didn’t have some sort of ASTM approval.
Another discussion centered around ASTM’s policy on the storing and distribution of ASTM standards
files. ASTM policy would need to be considered on whether mirrors at IUPAC would be allowed, or
whether mirrors at vendor sites would be allowed. It was noted that E13.15 could control the
proliferation of mirrors by simply providing an approved list of mirrors.
It was noted that we need to consider the needs of developers as well as end users when deciding how
best to store the schema and technique definition files. Vendors will probably want to store vendor
extensions on the own site. We should think about this now, so that we have a plan for how this will
work. The core and technique schemas need to be publicly available, and we don’t want competing
versions. If we have approved mirrors, they will need to be checked with some periodicity, to ensure
they have the latest version. There should be something in the standard to specify an area where people
can put their own definitions.
V. AnIML Workshop at Pittcon. Abstracts are due tomorrow for Pittcon. Speakers who haven’t already
done so should submit abstracts, and also send a copy to Dave Martinsen for loading on the AnIML web
site.
Other meetings include: a business meeting and a working group meeting at EAS in November; a Lab
Automation session on interoperability, which may accept a talk on AnIML; Burkhard Schäfer will be
submitting a poster or talk for Lab Automation; Gary Kramer will be giving an invite talk at IQPC. The
abstract for that talk will be loaded on the AnIML web site.
VI. Formal AnIML Requirements Document. Mark Bean, Tony Davies, and Alex Mutin are not
available at this time, so this discussion will be deferred.
VII. UnitsML. Bob Dragoset, from the Physics Department at NIST, presented a discussion of UnitsML.
Gary Kramer, Peter Linstrom, and Anh Dao Nguyen are on the UnitsML working group.
One of the important issues in Unitsml is the distinction between quantity and unit. Quantities (length,
mass, etc.) are measured in units (meter, kilogram, etc.) Units are defined in terms of a unit symbol, a
quantity name, and a unit name. The unit symbol is internationally accepted regardless of the unit name,
which may be different in different languages.
There are seven SI base units, and 22 special derived units, special in the sense that they have a special
name associated with them. Any special derived unit can be expressed in the seven base units. One of
the unresolved questions is how to deal with unitless quantities, for example, radians, with a unit of 1.
Absorbance may also cause problems.
SI derived units can be expressed in other SI units, providing for an unlimited number of derived units.
One question is how many of these to include in an SI database. Should square meter be included? The
NIST task group is looking to the various communities to decide which units should be included in the
database.
The following screen shots are from the Bob Dragoset’s demo of UnitsML:
Example 1. Select meter as the SI Unit Name:
Display XML output for the Unit Name meter:
Example 2: Show the XML record for the SI derived unit pascal:
Example 3. Select the SI derived unit “square meter”:
Display the XML record for the SI derived unit “square meter”:
Example 4. Select the SI multiple/submultiple “gram”:
Display the XML record for the SI multiple/submultiple “gram”:
Example 5. Select the non-SI Unit “fermi”:
Display the XML record for the non-SI unit “fermi”:
Example 6. Display the XML record for the SI derived unit “magnetic flux density”:
There was some discussion about what should be done with UnitsML within AnIML. Since UnitsML is
not a standard itself, there was some concern that having something which is not a standard as part of
AnIML would be a barrier to acceptance of AnIML. There was a suggestion that the units syntax in
AnIML could be modified to mirror the syntax of UnitsML. Also expressed was the viewpoint that
AnIML should be completely self-contained. There was no decision on how bests to incorporate
UnitsML in AnIML.
VIII. Agenda Items for Next Virtual Meeting.
1. Review of AnIML requirements document.
IX. Action Items:
Requirements document: Tony Davies to update his requirements figure with new items from
Mark Bean’s document.
Mark Bean to determine whether or not xml.org will satisfy the requirements for the AnIML
schema location.
Core Group will update the core schema.
X. Future Meetings:
Working Group Teleconferences
September 16, 2005, 10:30 am – 12:30 pm EDT
IQPC/ALI (January 30 – February 1, 2006, San Diego) and Lab Automation (January 21-25, 2006,
Palm Springs, CA)
EAS, November 13-17, 2005, Somerset, NJ, Doubletree Somerset, Room 211
Working Group Meeting, November 13, 2005, 3:00 – 5:00 pm
Business Meeting, November 14, 2005, 5:00 – 7:00 pm
PittCon, March 12-17, 2006, Orlando, FL
Business Meeting
Symposium/Workshop
XI. Adjournment. Motion to adjourn. Second. Carried unanimously at 12:31 pm EDT.
Submitted by David Martinsen, ACS, ASTM E13.15 Secretary