Document the origin of database files

[tpikonen]

I could not find any information in this repo on how the trie database files are produced. I'm thinking of packaging these files to other distros (than SailfishOS) and files with unknown origin are unlikely to be accepted.

Ideally, this repo should contain a build system from a freely distributable corpus to database for each language.

[rinigus]

Its a good idea to document it. Used files were very large and it would not make sense for them to be uploaded. At least one of them I cannot distribute either. But I will document as much as I know and would try to ask for pointers if I haven't made them myself.

[ichthyosaurus]

hey guys :) I'd like to suggest going one step further and making reproducible builds for all databases. I don't know if that's feasible, but at least providing build scripts to rebuild the databases as close to binary identically as possible.

At least one of them I cannot distribute either.

Maybe source corpi (corpusses?) could be kept in the Internet Archive? Or contact info if its kept private by a university.

Since the hostile fork of Whisperfish yesterday, I was a bit alarmed to see binary blobs begin merged which were provided by the same bot (?) without sources. I think we should find ways to make Sailfish more auditable.