Should an SDO (Service Data Object) be adopted in a new project?
I've been programming in Delphi with Midas / DataSnap for quite some time now and am quite happy with it. Moving to .NET. I am more than happy with the ADO.NET DataSet. For a CRUD application, I am very uncomfortable with any ORM. A common data structure with automatic diff / delta handling makes my job easier for me, the average database application developer.
Tried to learn Java a few years ago and couldn't find a similar idea. The closest I could find is SDO (Service Data Object). I thought it should be widely accepted when I saw it, but I am wrong. Even the spec is quite old, I still hardly find many people discussing it or using it extensively. Based on the information I can find on the internet, using SDO is very passive.
I wonder if he's dying What experiences with SDO would you like to share? Is hand coding DTO always better?
a source to share
I would not recommend using SDO unless it is dictated to you by some other part of the project.
The WebSphere process server uses SDO. This is not a bad API once you get to know it. But the spec and documentation are vague. It doesn't specify what happens if you ask for a field that doesn't exist, or it does type conversions when getting or setting up fields to name two scrum.
I don't think the API defines how to define new types, so the part will be implementation specific. Type definitions are based on XSD, so you will be working with these and all related standards.
As others have suggested, the API is not widely used. This means that it will be difficult to find people who can handle it, or to help them.
a source to share