We don't care about serialization verbosity

[enikao]

Reducing the size of serialized models (and other data) is not a goal for LIonWeb.
We accept more verbose (aka. longer, aka bigger) serialized models in order to reduce complexity.
If size became an issue, we would tackle it on a different level, e.g. compression of transport format.

[sergej-koscejev]

Besides size, there's also human readability. It's easier to read a structure that only contains the important data, compared to a structure with a lot of boilerplate properties set to null values.

Related: the decision to Base64-encode concept IDs makes them non-human-readable in the JSON.

Would it be a problem if we would by default allow to leave out fields that are optional/null, unless there are reasons not to?

[enikao]

Related: the decision to Base64-encode concept IDs makes them non-human-readable in the JSON.

We discussed this as feature, not a bug in #31

[enikao]

Would it be a problem if we would by default allow to leave out fields that are optional/null, unless there are reasons not to?

Arguments at #59

[enikao]

In the meeting at 2023-02-17, we confirmed that we don't care about no stinkin' serialization verbosity.