Issue 9204: CalibratorType (xtce-rtf) Source: (, ) Nature: Enhancement Severity: Critical Summary: SCOS 2K uses another type of calibrators that is not supported by XTCE. XTCE should provide a way to support a kind of logarithmic (or custom) calibrator. The calibration is defined by the following equation: Y = 1/{A0 + A1 ln(x) + A2 ln2(x) + A3 ln3(x) + A4 ln4(x)}. Ln is the natural log (base e), and Ai are coefficients stored for the calibrations. Of course such a calibrator will need, as others, a short description and a name. If this thought as not generic enough, then XTCE should give a way to describe non standard calibrators. For example, coefficient, base, exponent in a term sum used as a calibrator. Resolution: Revised Text: Actions taken: December 1, 2005: received issue Discussion: Resolution: Submit potential new calibrator as a Schema modification Disposition: Deferred End of Annotations:===== me: XTCE Document #: formal/05-08-01 Section: CalibratorType Version #: 1.0 Revision d.: 01/08/2005 Page: Nature: Enhancement Severity: Critical Full Description: SCOS 2K uses another type of calibrators that is not supported by XTCE. XTCE should provide a way to support a kind of logarithmic (or custom) calibrator. The calibration is defined by the following equation: Y = 1/{A0 + A1 ln(x) + A2 ln2(x) + A3 ln3(x) + A4 ln4(x)}. Ln is the natural log (base e), and Ai are coefficients stored for the calibrations. Of course such a calibrator will need, as others, a short description and a name. If this thought as not generic enough, then XTCE should give a way to describe non standard calibrators. For example, coefficient, base, exponent in a term sum used as a calibrator. From: "Rice, James K. {Kevin} (GSFC-581.0)[ASRC RESEARCH & TECHNOLOGY SOLUTIONS]" To: "Kizzort, Brad" , "Overeem, David T" , Rob Andzik , "xtce-rtf@omg.org" Date: Thu, 13 Jun 2013 12:26:02 -0500 Subject: RE: XTCE 1.2 RTF report due at Berlin Thread-Topic: XTCE 1.2 RTF report due at Berlin Thread-Index: AQHOYT4OOzbzvJt7DkaJy14BgcCqjpkngpkAgAAICYCAAcT1gIAAAcQA//+/HwCAAAGxYIAARBKAgAFWs4CAAACTAIAAA8gAgAABSoCAAAI9AIAABIGAgAAPx4CAA1k7AIAAfVQAgAC6coCAAAhUgIAAAdOAgAACeQCAAAf+AIAAANkAgAAC5QCAAAFHAIAAAcaAgAAFGICAAAEcgP//1+AggAAB9NCAAATFQIAABnLwgAAEJSCAAVPbsIAAg80ggAHqx+CAAGcr4IAAYyGQ Accept-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: acceptlanguage: en-US X-Proofpoint-Virus-Version: vendor=fsecure engine=2.50.10432:5.10.8794,1.0.431,0.0.0000 definitions=2013-06-13_07:2013-06-12,2013-06-13,1970-01-01 signatures=0 X-Virus-Scanned: amavisd-new at omg.org Actually I think the real problem with OMG Issue No: 9204 Title: CalibratorType (xtce-rtf), isn.t so much needing more operators. The real issue is having to bail out of FloatDataEncoding/DefaultCalibrator because you need a custom calibrator . and this forces you to move to BinaryDataEncoding & thus you lose the information that the item is a known float encoding . and it doesn.t help to use AncillaryData to mark it as such much because this is basically a customization & not very transferable. So in my mind there are two approaches that should be adopted: one would be to put custom algorithm under the calibrators list (& then the alternate issue is expanding operators in the mathoperations to a custom alg is not needed) . or alternatively add an attribute to the BinaryDataEncoding to give its data type: int, float, string, or custom -- or something along those lines. This of course is not exactly what the issue says. but that.s the real issue. From: Kizzort, Brad [mailto:bkizzort@harris.com] Sent: Thursday, June 13, 2013 8:35 AM To: Rice, James K. {Kevin} (GSFC-581.0)[ASRC RESEARCH & TECHNOLOGY SOLUTIONS]; Overeem, David T; Rob Andzik; xtce-rtf@omg.org Subject: RE: XTCE 1.2 RTF report due at Berlin I will likely submit the final report in OpenOffice, but I can cut-and-paste into it from any number of formats, including Word. The two templates, Word and OpenOffice have different styles and once you have entered a lot of data you have to re-enter it or restyle each paragraph if you switch (and we are half done) in order to get the automated tables to work. My method in the past has been to circulate a subset of issue sections in the same format as the final report to the RTF for a vote. I use Mercurial for CM, so I can send you a zip of that and we can exchange changesets if you prefer to work that way. I try to commit each of the issue polls as a different changeset to the schema, so that I can pull a difference report and roll it back if the RTF doesn.t accept a disposition or modifies the disposition. I would like to continue working sequentially through the issues as they appear on OMG.s web page, to make sure that none get missed. I certainly don.t have a problem, if Kevin or Dave want to tackle the next set of 10-25 issues and circulate a proposal to the RTF. I am beginning to get some free time again as my last program winds down, but I do welcome help. I have attached Poll3Issues.odt (last approved set of dispositions) as an example of the last polling document and the ProposedSchema for 1.2 as of poll 3. The next issue on OMG.s list is 14450 Clear Up Calibrated/Uncalibrated Values in Schema. The issues list is freely viewable at http://www.omg.org/issues/xtce-rtf.open.html, no membership required. I am not averse to submitting new issues to the OMG and addressing them in 1.2, but the RTF has to disposition all of the outstanding issues. Brad