Timeline for What are the benefits of dependency injection in cases where almost everyone needs access to a common data structure?
Current License: CC BY-SA 3.0
6 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Sep 22, 2015 at 22:12 | comment | added | corsiKa | Reminds me of a Dilbert where PHB asks for 500 features and Dilbert says no. So PHB asks for 1 feature and Dilbert says sure. Next day Dilbert gets 500 requests each for 1 feature. If something doesn't scale, it just doesn't scale no matter how you dress it up. | |
| Sep 22, 2015 at 20:58 | comment | added | Telastyn | @RobertHarvey - sure, but is the interface they depend on "can access a database" or "some persistent data store for foos"? | |
| Sep 22, 2015 at 19:59 | comment | added | Robert Harvey | Applications that have many objects which routinely access some database are not exactly unprecedented. | |
| Sep 22, 2015 at 19:57 | comment | added | vsz | "This is a design nightmare" - I know it is. But if the requirements are that all those events must be able to perform the changes in the data, then the events will be coupled to the data even if I split and hide them under a layer of abstractions. The whole program is about this data. For example, if a program has to perform a lot of different operations on an image, than all or almost all of its classes will somehow be coupled to the image data. Just saying "let's write a program which does something different" is not acceptable. | |
| Sep 22, 2015 at 19:50 | comment | added | vsz | sorry for the confusion with "non-mutable", I wanted to say mutable, and somehow didn't recognize my mistake. | |
| Sep 22, 2015 at 13:57 | history | answered | Telastyn | CC BY-SA 3.0 |