43 | | (organized calls or meetings before GEC13?) |
44 | | |
45 | | Review conclusions in pre-meeting at GEC13 |
46 | | |
47 | | Review with working team at GEC13 |
| 43 | |
| 44 | Review with working team at GEC13 [[BR]] |
| 45 | |
| 46 | Summary of current status (Giridhar Manepalli): [[BR]] |
| 47 | [http://groups.geni.net/geni/attachment/wiki/GEC13Agenda/InstrumentationAndMeasurement/T5%29%20%20MDOD%20Status%20-%20CNRI.pptx slides] [[BR]] |
| 48 | |
| 49 | Conclusions: [[BR]] |
| 50 | |
| 51 | Good things: [[BR]] |
| 52 | Excellent start [[BR]] |
| 53 | Collaborative Specification [[BR]] |
| 54 | Great Coverage [[BR]] |
| 55 | Nicely broken down into elements [[BR]] |
| 56 | Mandatory vs. optional elements identified [[BR]] |
| 57 | Genuine Use Cases: Gathering, transferring, and sharing [[BR]] |
| 58 | |
| 59 | Jensen's proposal (NetKarma): [[BR]] |
| 60 | Current: Identifiers, Descriptors, Holders [[BR]] |
| 61 | Proposed: Identification, Lineage/Provenance, Constraints/Security, MDO Description [[BR]] |
| 62 | |
| 63 | Zurawski's comments: [[BR]] |
| 64 | Too many secondary identifiers [[BR]] |
| 65 | Descriptors should be contextualized [[BR]] |
| 66 | Variations based on the type of object [[BR]] |
| 67 | GENI specific descriptions should be clearly marked and separated [[BR]] |
| 68 | Slight changes to names & enclosing elements recommended [[BR]] |
| 69 | |
| 70 | Comments/suggestions based on metadata practices: [[BR]] |
| 71 | Too many optional elements [[BR]] |
| 72 | Too many choices given to users [[BR]] |
| 73 | Users bound to take the path of least resistance [[BR]] |
| 74 | Keep the scope restricted to only mandatory elements – at least in the beginning [[BR]] |
| 75 | Try those out. Implement them. [[BR]] |
| 76 | One size fits all ---- No! [[BR]] |
| 77 | Capturing descriptions, formats, policies, transactions, etc. in a monolithic fashion [[BR]] |
| 78 | Register individual components separately [[BR]] |
| 79 | E.g., Capture legal formats & interpretations in their own records, and reference them here [[BR]] |
| 80 | E.g., Same with accepted policies [[BR]] |
| 81 | Identifiers cannot be semantic [[BR]] |
| 82 | Domain, sub-domain, and object-type are part of an ID [[BR]] |
| 83 | World view changes frequently [[BR]] |
| 84 | Non-semantic Ids are worth every penny [[BR]] |
| 85 | Search engines & registries mask the opaqueness [[BR]] |
| 86 | After all, IDs are just entities behind the scenes [[BR]] |
| 87 | Object Type controlled vocabulary enumerates apples and oranges [[BR]] |
| 88 | Collection, flow, directory, file, database, gui are not mutually exclusive [[BR]] |
| 89 | Doesn’t help the recipient make any decision looking at the descriptor [[BR]] |
| 90 | Bundle type & format into format interpretation method [[BR]] |
| 91 | Covers too many corner cases, e.g., flow-rate [[BR]] |
| 92 | Expects too many details, e.g., locator (type, access method, etc.) [[BR]] |
| 93 | |
| 94 | |