| Totto |
Joining data in memory beats joining adta out of process/in db any day.. |
|
why is this so hard to understand.. |
| Totto |
When ever was pulling a large undetermenistic context-bound graph into your application ever a good idea.. |
|
|
| Emil |
Johannes is wrong |
|
|
| Emil |
Integration through a shared db breaks layering => high coupling |
Resistant to change |
|
| Emil |
Shared db => group ownership of schema |
Fucked up shit |
|
| Borud |
Please do not fondle my data |
|
|
| Johannes |
The database is functionally equivallent to a service layer (or can be made to be) |
|
|
| Johannes |
When exposing a database, you have to design the interface schema, you muppet |
|
|
| Dan |
If you're doing data integration, you're fucked |
|
|
| Kjetil |
Any data integration pattern axiom may or may not be applicable to you |
|
|
| Kjetil |
Two data integration pattern axioms may be at least one too many |
|
|
| Ørjan |
We need one and only one model for each service, and how data is persisted is nobodys business but the service itself |
|
|