Talk:Collaborative workspace

From Opasnet
Jump to navigation Jump to search

Can the collaborative workspace calculate?

How to read discussions

Fact discussion: .
Opening statement: It is possible to calculate variable results in the collaborative workspace.

Closing statement: Accepted.

(A closing statement, when resolved, should be updated to the main page.)

Argumentation:

⇤--1: . not part of scoping (and not very feasible either I think...) --Anne.knol 16:45, 15 March 2007 (EET) (type: truth; paradigms: science: attack)

⇤--3: . At least some (simple) common calculation methods that nearly everyone uses might be provided. If they are provided directly in the scoping diagram (by clicking on the variables) or not may be decided later. --Alexandra Kuhn 10:20, 19 March 2007 (EET) (type: truth; paradigms: science: attack)
←--2: . It is not directly a part of the scoping, but it puts demands on the scoping tool if this should be possible to do. As for the feasibility I dont know, but KTL are already doing something like this with the wikimedia <-> analytica tool --Sjur 12:04, 16 March 2007 (EET) (type: truth; paradigms: science: defence)

Does uploading take too much resources?

How to read discussions

Fact discussion: .
Opening statement: Uploading data takes too much resources and cannot be handled.

Closing statement: Resolution not yet found.

(A closing statement, when resolved, should be updated to the main page.)

Argumentation:
⇤--1: . Uploading and saving results will cause an unmanageable amount of data that no system will be able to cope with this. So either the variables only contain "cooking receipts" and no results, or only aggregated results should be saved (e.g. number of persons with a certain disease, external costs; not the results of a transport model and all the concentrations values in the whole of Europe). --Alexandra Kuhn 12:28, 10 August 2007 (EEST) (type: truth; paradigms: science: attack)
⇤--2: . Whether the amount is unmanageable or not depends mostly on the information structure and data storage architecture. I have started to think data (more precisely: results of variables) something that we should store in a specific place called result distribution database. This would be a decentralized database, and the actual toolbox software only would contain link information to that database, so that data can be queried from there. Results from very large models should not be saved, unless the data provider is him/herself willing to pay the costs of this. --Jouni 17:48, 14 August 2007 (EEST) (type: truth; paradigms: science: attack)