Thanks for your input. Because I chosed Net/C# as my development
platform, it seemed for me at the first place that a C# port would bring
this Data Storage interopeability benefit to deveopers of standalone
desktop application. My app is an ETL (Extract Transform Load) utility,
that's why Apache MetaModel seems so worthy for me (and that's for long
that I waited for a versatile Data Storage solution).
The motivation is to develop a standalone ETL (Extract Transform Load)
desktop application which uses Apache MetaModel as a Data Storage API in
order to be get the interoperability benefit within the range of
Your advice to use gRPC (which is great, did'nt knew this solution in
fact) and the concern about maintaining another codebase are both very
pertinent. In that case, would it be a better idea to create a 'gRPC'
subproject ? My concern is that I would prefer a "Javaless" AND
"Serverless" solution but if Apache Metamodel with gRPC extension is
very straigthforward to deploy for a .Net standalone desktop app (and if
a server is required, that it is an "in process one") then your proposal
seems too make more sense than mine.
Last remark is that when progressing on this "C# port" prototype, I had
the feeling that C# is now a more modern language (less clumsy I mean)
than Java (IMHO, I may be completely wrong of course) and also maybe
that this prototype may reveal some "design / extensibility /
portability issues" in Apache MetaModel, especially because using a
programming language often bring its own set of biases because of the
design choices which are reified by its idioms.
On 05-09-17 20:40, Henry Saputra wrote: