In the past 20 years, having a digital representation of a calibration certificate has been a topic of discussion. I remember attending my very first NCSLI Workshop & Symposium in 2001 and sitting in on the 141 Automation Committee meeting discussing this very topic. The only thing we could agree on back then was the file format should be XML. But unfortunately, the only thing we have been able to accomplish since then is more calibration labs offer a PDF version of a calibration certificate and call that a digital format.
Well, it would seem the world has come full circle and we are again working to create a standard for digital calibration certifications. Only now there is a worldwide effort of people, companies, and National Metrology Institutes (NMIs) all working on the problem of Digital Calibration Certificates–each with an XML definition, hoping other organizations will adapt their standard.
This was my comment at my first ever NCSLI 141 committee meeting: “Yes, XML is a great technology but who is going to define the XML Schema?” 19 years later we are still asking the same question. But the encouraging news is now there are several published schema that we can evaluate and possibly implement.
The schema, or as it is known XSD file, is a formal definition of how the XML file and its elements are formatted. See, XML files in a sense are mini-database organized in a hierarchical format from the root node to all of its elements. The specified name of each element, parent-child relationships, and attributes are formally defined in the XSD Schema File, thus making it the key or map to all the data in the file.
This week, PTB (Physikalisch-Technische Bundesanstalt), Germany’s NMI, held a virtual conference called the DCC 2020 workshop covering information, adoption, and presentation about digital calibration certificates. Over the past year, there have been a lot of tools and middleware applications built around their XML Schema. If you would like to learn more just visit https://www.ptb.de/dcc.
PTB’s DCC is a great format from the NMI Level, focused on measurement and uncertainty values. But the production calibration labs at the bottom of the metrology pyramid are more interested in Pass/Fail limits with tolerances. They want to know what is their risk, and can they use the instrument to make measurements or not. This was discussed in this year’s DCC virtual conference and may get added in the future. Also, the topic of accredited vs. non-accredited calibration data was addressed. I didn’t know this, but they mentioned some NMIs are not accredited like most calibration laboratories.
Another great paper from Miguel Marques, with INEGI (Institute of Science and Innovation in Mechanical and Industrial Engineering) of the University of Portugal, presents a great schema layout but it brings up an interesting problem: What language should the schema be writing in, what date and time format, and what number format? There are a lot of things that need to be definitely addressed in the schema, such as Americans like the Month/Day/Year format, while Europeans like the Day/Month/Year format. NATO addressed this problem with the 01-Jan-20 format that is non-ambiguous, as well as the worldwide adoption of Zulu Time or UTC (Coordinated Universal Time).
Earlier this year, at NCSLI’s 2020 virtual conference, Collin Delker with Sandia National Labs presented a paper “Exploration of a Data-Enhanced Calibration Certificate as Part of a Complete Measurement Information Infrastructure.” This paper was quite interesting in that it doesn’t solve the international language or formation problems, but provides the best of both worlds by combining XML and PDF in one file. Instead of representing everything in an XML format, the XML data is embedded into the PDF. You can download and read this paper from NCSLI’s 2020 Conference proceedings at https://www.ncsli.org.
Whether it is embedded in the PDF or a standalone XML file, we have a lot of ground to cover as we work to create an International Digital Calibration Certificate schema representation of the calibration data that is transportable from system to system. I think we all need to look at all of the proposed standards, compare and contrast them, and look at what expansion we need to make to our data formats. We also need to think about how we can provide feedback on what is missing from each format that is out there.
11/16/20 Correction: Miguel Marques’ name was misspelled.
