Implicit read in CRUD Update?

[Paolo F Cantoni]

Elsewhere, I've indicated I'm doing some modelling involving CRUD (Create, Read, Update, Delete) relationships between behavioural elements and artifacts.

I've modelled "read" separately from "update" (that is, two relationships) - but I find I'm getting sooo many relationships that the diagrams are somewhat unreadable.

I couldn't find any explicit definition, to this effect, but it seems to me that a normal update implies and implicit read.  So I can retain "read" for a read-only - no consequent update.

This would allow me to cut the number of relationships significantly.

In the unlikely event that I needed a "write-only" function (writing without a preceding read) I could extend the set to "write".

Thoughts?
Paolo

[«Midnight»]

Makes sense.

Could there be confusion between the 'normal' interpretation of "read" and "write" and your context? That is, are the situations where the normal meanings would be applicable in your paradigm? If such were the case you might be obscuring the intent of your models for readers who are not familiar with the specific instances you diagram.

[Paolo F Cantoni]

Makes sense.

Could there be confusion between the 'normal' interpretation of "read" and "write" and your context? That is, are the situations where the normal meanings would be applicable in your paradigm? If such were the case you might be obscuring the intent of your models for readers who are not familiar with the specific instances you diagram.

Yes...  However, in the context in which they are being used, I'm having to develop an entire Instructional/Help subsystem to describe exactly how the modelling is to be performed and understood.

This is just another aspect of that.  I just need to be consistent in how I propose to model the real-world relationships.  That's why I'm asking now (as the proof of concept/prototype is being created)....

Thanks for the input, David.

Paolo

[«Midnight»]

I wanted to provide limits within which the concepts (or their interpretations) might be contextually sound. Within those confines you should be OK.

[jeshaw2]

Paolo;

You seem to be at a very low level of  abstraction, almost at the coding level.  Are you sure that's where you want to be?

I don't know what DBMS you are using.  Does it distinguish between 'read' and 'read for update'?

Jim

[Paolo F Cantoni]

Paolo;

You seem to be at a very low level of  abstraction, almost at the coding level.  Are you sure that's where you want to be?

I don't know what DBMS you are using.  Does it distinguish between 'read' and 'read for update'?

Jim

I'M actually modelling at the business process level.  I'm trying to record if a document is created, read-only, updated (implicit read) or destroyed by a process.  The idea is that in a complex set of business processes, we can ask the questions:

Where is this document created?
Does anyone update the document?

Etc...

Hope this clarifies things...
Paolo

[Geert Bellekens]

Paolo,

That is exactly the same interpretation I used a while ago when documenting the business modelling method for a former client. (i.e. Update implies Read)
Back then I was working in a team of methodologists so I had the chance to check this interpretation with them, and they all agreed. (what a luxury that is sometimes, to be able to discuss those kind of issues with people who know what you are talking about)

Geert