Hi all,
Our application has about 80 interfaces - small pages/lists/dialogs to do various things.
We also defined about 25 record types with some that have reached 100 fields (max limit)
What would be the recommended approach?
1 - try to keep 1 to 1 relation between interface to record type - many small record types with only the fields that are needed directly by the interfaces and related rules
OR
2 - try to have only a few massive record types - with max amount of columns/fields and reuse across many interfaces - generic massive rules that selects anything with a million filtering options
thank you
Discussion posts and replies are publicly visible
octavian said:What would be the recommended approach?
Really I have no way of saying one way or the other, without more information about your use case / user requirements / etc.
What is your end goal? What design concept(s) will allow your end users to most easily see what they need to see, and complete what they need to complete?
This is not about user experience, rather maintainability of the code, avoiding too many layers
So , how to use it as it was intended, for example in C++ it would be a no-no to make one class with 10000 lines of code..
My goal is to simplify maintenance of the code behind
I think that I do not understand you design approach. What do you mean with layers?
My world with Appian records became very simple. Each DB table becomes a record. And this is completely disconnected from any interfaces. Interfaces are build to support s specific user interaction in a business process. It will read/write any data that is required.
And I try to keep design objects simple. Even at the cost of having multiple doing similar stuff.
ok, I am leaning in this direction too - thank you for confirming the approach