Re: Performance of aggregates in the database

Hi Greg,

I think when you talk with an expert you talk in concepts and give name and
structure to things. This can be captured pretty well in DSL model,
regardless wheter it is implemented as anemic or rich model.

On the other hand, implementing behavior, I agree, is the more interesting
part. Although I would not call that part modelling.

I like to think of the tool as the replacement for all the boilerplate I
need to implement complex behavior. And I want such boilerplate if it gives
me type safety, evolutionability and performance. When there is complex
part of the app which warrants custom objects the tool gives me API and
DTOs for mapping my rich objects to external systems.

Without going into discussion of rich services vs rich objects.

Regards,
Rikard
On Mar 12, 2015 10:57 PM, "Greg Young [email protected]
[domaindrivendesign]" <[email protected]> wrote:

>
>
> Reading https://docs.dsl-platform.com/ddd-real-world
>
> This seems to suggest that the model you should be talking about with the
> domain expert is the anemic model that you are generating with the tool
> (you should be discussing the model but the tool generates an anemic
> model). Normally I find the interesting parts of the model have nothing to
> do with its structure.
>
> On Wed, Mar 11, 2015 at 4:16 PM, [email protected] [domaindrivendesign] <
> [email protected]> wrote:
>
>>
>>
>> I've managed to finish up benchmark of Revenj framework. It's available
>> at: https://github.com/ngs-doo/dal-benchmark
>>
>>
>> I think DDD people might find it interesting since it talks to
>> "Relational database" in non-conventional way and proves that DDD can be
>> practiced without friction outside of JSON DB.
>>
>>
>> It also shows how framework can be faster than manual coding of SQL.
>>
>>
>> If you have questions or comments, let me know.
>>
>>
>> Regards,
>>
>> Rikard
>>
>>
>>
>>
>
>
> --
> Studying for the Turing test
>   
>