Talk:Opasnet structure

From Testiwiki
Revision as of 11:30, 18 February 2010 by Mikko Pohjola (talk | contribs) (Opasnet development discussion 18 February 2010)
(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Opasnet development discussion 18 February 2010

Hello Marko,

Good points. Here are some quick responses to the issues you brought up.

1. Definitely, the most that can be required from an Opasnet user is to have a web browser and web access. Whether it is Opasnet itself or some external resource that Opasnet communicates with that actually does the calculations, the user (any of the myriad types of users imaginable) must be able to communicate with the system through the web browser interface only. No external applications can be required. This principle is well understood, but there is still plenty of work to do to implement it smoothly to full extent. Despite several nice features of Analytica, its not being open source significantly reduces its usability in open collaboration type settings. I believe the solution can be found by developing the interactions between Opasnet (wiki) and an external calculation/simulation/modelling engine(s) or embedding such into the Opasnet system. The co-operation between Opasnet and Simupedia (simupedia.com) may bring about some very useful insights to this issue.

2. I have also often been thinking about the possibility to graphically display the interrelations between objects in Opasnet, and thereby "being able to see one's location in the information jungle". Good solutions to this are yet to be found. E.g. traditional tree structure navigation does not work as the structure of information does not follow a strict hierarchy. Instead, there could be possibilities to map the relationships according to defined dependencies between variables, but that does not take account of other object types (is it necessary?). Assessments and assessment models, especially if represented as causal diagrams, bring some help to seeing bigger wholes and the interrelations between their parts, but that does not really solve the problem (for instance because assessments are discrete and variables are, in principle, eternal). This issue of graphical mapping of Opasnet contents must definitely be brought back into explicit discussions on near future system development.

Thanks for bringing up these important issues, Marko!

BR, Mikko


Hi All,

Couple of ideas came to my mind regarding the functionalities of Opasnet and I decided to write them down (Jouni: yes, I know, Opasnet would be right place but I like to use email).

  1. Jouni has expressed an idea that Opasnet is the place where Indian engineer can do risk assessment. The idea is noble but I don’t think that current Opasnet makes it possible. The problem is that this engineer might have access to internet but no access to any software’s to conduct the calculations. It’s even likely he/she is contacting Opasnet through the mobile phone or through some other mobile device unable to perform any complicated calculations.There is one solution for this: Opasnet must be able to do calculations by itself. Thus, if you would put some equation to Opasnet, then Opasnet server would calculate the results. This way all what that Indian engineer needs is the web browser. How to do previous one is a big and open question. One option is to attach to Opasnet some open source program to do the calculations or attach Opasnet to some calculation program. Or program a tool from scratches…
  2. Second idea is not as broad as first one. Now Opasnet has plenty of variables that are connected to other variables. Would it be possible to add Opasnet a graphic interface that would show with graphics how different variables are connected? This would make it easier to see from where the data to different variables are coming and in where the data from one variable is used.

In the case that these ideas have already been discussed in YMAL, or they already exists in Opasnet, I apologize the disturbance.

Marko