I don't know if this will help, but here it goes... Really not else to say in regards to dataset... Rabi is very correct in saying "When you talk about two datasets...you are talking about two separate containers to represent two different versions (i.e. data) of the same real world object(s)."
In terms of namespaces, it might help thinking that the namespace would be similar to using a reserved range of IDs for your new fields in AR or some particular naming convention for new forms you develop. Namespace is good to identify and separate the "customisattion" in your CMDB. Again, hope this helps... Julio ----- Original Message ---- From: Rabi Tripathi <[EMAIL PROTECTED]> To: [email protected] Sent: Friday, 21 December, 2007 8:24:20 AM Subject: Re: dataset vs. namespace Very similar concepts that apply to different areas of object oriented world. Each one acts like a boundary to define one universe. Sorry, if you're not a fan of SAT, but let me try this-- Namespace : Dataset = Classes & Attributes : Class Instances (or Data) or Namespace : Dataset = Class Definition : Class Data or Namespace : Dataset = World in which data's structure is defined, named: World in which actual data is stored In other words, when you talk about two different namespaces, A and B, you are talking about two separate and independent worlds in which you can define entities (definition of your data's structure) separately. The namespaces are independent of each other in the sense that you can define an entity, say "MyClassName", regardless of whether this name is already defined in the other or not. Because they are separate worlds, for the purpose of naming entities (entities=classes and attributes). When you talk about two datasets...you are talking about two separate containers to represent two different versions (i.e. data) of the same real world object(s). For example, your "Production" dataset for CIs in CMDB contain official CI details as they are recognized now. Your "Discovered" dataset will have information about the same CIs, but as discovered by some tool. They may or may not match. In both datasets, you may have a an instance of a CI (.i.e. data about its state at some point), say your laptop123. In production dataset, data indicates that laptop123 is a healthy system. In discovered dataset, the same laptop123 may be represented as a screwed up system on which your dog last weekend spilled coffee and when you tried punching the dog, you broke the CD drive, instead. Note that the structure definition for both datasets (i.e. the class and attributes definitions) are identical, so they would fall under the same namespace. I feel like writing more, so here it goes. :) My Class Definitions (i.e. definition of my data's structure) ================= Namespace: MyPlayNamespace Class: Person Class Attributes: Name Attractiveness Money Cool Factor Namespace: MySeriousNamespace Class: Person (same name is allowed, because this is a different namespace) Class Attributes: Ethnicity Name Average Volume Average # of Drinks Before Passing Out My Class Instances (i.e. my actual data) ============================= Namespace: MyPlayNamespace Class: Person Dataset: "young" (this is one version of the real world entities, the people; so "young" represents the "version number" for this collection of data) Data (this dataset has 3 class instances): Name: Rabi Attractiveness: High Money: Low Cool Factor: High Name: Labi Attractiveness: High Money: High Cool Factor: Superhigh Name: Babi Attractiveness: Low Money: Low Cool Factor: Superlow Dataset: "old" (this is another version of the same real world entities) Data (3 class instances): Name: Rabi Attractiveness: Low Money: Medium Cool Factor: Medium Name: Labi Attractiveness: Low Money: Low Cool Factor: Low Name: Babi Attractiveness: Low Money: High Cool Factor: Medium ---------- The other namespace "MySeriousNamespace" that also has a class with the same name, "Person", can similarly have its data defined in more than one datasets, with any names. Note that in this namespace, the "Person" instances happens to represent different real world entities. It really defines defines human ethnicities, not individuals like the "Person" class in "MyPlayNamespace" So, namespaces define totally separate universes. Within each namespace, for a given class(es), you can have one or more datasets, each dataset representing data about the different versions (perhaps at different times, or from different experiments etc) of the classes's real world objects. I wrote too much. Now I want to stop. Drew Shuller wrote: > > Hello everyone, > > I'm a little confused as to the difference between a dataset and a > namespace. Any help? Thanks a bunch! > > Drew > Tulsa, OK > > _______________________________________________________________________________ > UNSUBSCRIBE or access ARSlist Archives at www.arslist.org > Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are" > > -- View this message in context: http://www.nabble.com/dataset-vs.-namespace-tp14441063p14445451.html Sent from the ARS (Action Request System) mailing list archive at Nabble.com. _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are" Make the switch to the world's best email. Get the new Yahoo!7 Mail now. www.yahoo7.com.au/worldsbestemail _______________________________________________________________________________ UNSUBSCRIBE or access ARSlist Archives at www.arslist.org Platinum Sponsor: www.rmsportal.com ARSlist: "Where the Answers Are"
