+Simulation data Model version 1.0

Here is a discussion page for Theory IG and Data Model WG about the specification of the Simulation Data model. See http://wiki.ivoa.net/twiki/bin/view/IVOA/IVOATheorySimDMspec.

  • Proposed recommendation SimRFC
  • List of current releases during the WG review

* release v1.0-20110427: doc pdf

  • see also the Appendix Document available on the Volute page

* release v1.0-20110314: doc pdf

  • Comments [MC ...]
  1. Appendix? I miss the appendix of the document. The address provided is not correct and the only document I find in Volute is in a Draft status.
  2. Wording Although yet discussed in the TIG, I just reopen the discussion on the DM group about the conflicts of some wording in the document: in particular the use of Protocol referring to simulations runs and its conflict with the use of Protocol in the IVOA contex which refers to data access\
  3. Drawback and forward compatibility with other DM efforts How this data model includes and it is compatible with other datamodel efforts? In particular, if there is yet a reusable data model to define spectra or observational data, which parts of such data-models that are in common to the proposed one?
How such datamodels can be included in the SIMDM?
(note that the final goal of theoretical data is to be tested with observations, so theoretical data MUST SHARE some of the characteristics of observational data)

  • Replies [GerardLemson 2011-04-27]
  1. Indeed links were wrong. I have created a version of the appendix document that corresponds to the link in the WD. I have updated some other links in the WD document. Results in release v1.0-20110427 (see above).
  2. First, SimDB:simdb/protocol/Protocol does not correspond to a Simulation run, but to the "way in which an experiment can be run", in our model often the simulaiton code. The use of the word Protocol for this concept is directly extracted from Martin Folwer's book Analysis Patterns (see ref [27]). The fact that it may be confusing as we also talk about other kinds of Protocols in the IVOA does not do away with the "appropriateness" of its usage here. So I will vote for keeping it, but if people come up with a better term this seems a good place to add them.

  1. Section 6 deals explicitly with overlap with other efforts. As far as data models are concerned we there explicitly address correspondences with the Registry data model and with Characterisation. We have here a model that aims to describe results of simulations and other computational protocols. Not a model for observations and their results. And the original aim was that it would serve the purpose of discovering potentially interesting results. Not to provide a way for expressing results of simulations in a format useful for messages. That said, we do have a way in which one can describe that the result of a simulation is a spectrum for example. But regarding the aims of our model it is impossible to do that in the same explicit form as the spectrum data model. Linking different models will in general require more sophisticated methods than a simplistic copy-and-paste. A proper discussion of this issue, within the DM WG one would think, is long overdue (in my opinion).

* Comment [MireilleLouys 2011-05-09] About Protocol,

I think this is an appropriate name to express the description of actions that should be executed in order to accomplish an Experiment and produce Results. This very general term is valid for many domains : manufacturing, chemistry, diplomacy, etc. and more recently computer science. The text of the standards reads (experimental) Protocol which helps to clarify the fact that we use a general meaning here and not a technical word borrowed from computer science.

Topic attachments
I Attachment History Action Size Date Who Comment
Microsoft Word filedoc WD-SimulationDataModel-v.1.00-20110314.doc r1 manage 840.5 K 2011-03-14 - 14:25 HerveWozniak Simulation Data Model Working Draft 1.00
PDFpdf WD-SimulationDataModel-v.1.00-20110314.pdf r1 manage 613.3 K 2011-03-14 - 14:26 HerveWozniak Simulation Data Model Working Draft 1.00
Edit | Attach | Watch | Print version | History: r9 < r8 < r7 < r6 < r5 | Backlinks | Raw View | Raw edit | More topic actions
Topic revision: r9 - 2011-05-09 - MireilleLouys
This site is powered by the TWiki collaboration platform Powered by Perl This site is powered by the TWiki collaboration platformCopyright © 2008-2024 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback