Closed xrotwang closed 5 years ago
Hi, sorry about the delay in getting back to you. We're happy to help seed the adoption of DataCrate, datapackage conversion should be super easy, and the CLDF stuff looks interesting. If you'd like to start on one I'm happy to help out. Basically that's needed is something that puts a CATALOG.json file in the root of a dataset, as per the spec / examples, then use Calcyte to generate HTML, bag and zip it.
I'll add datapackage to my list of things we can consider supporting. The problem is with it's extensibility - because it doesn't use linked data principles there's no easy way to support arbitrary metadata that people might add, but I don't know how much of an issue that would be in practice.
The package format CLDF uses actually isn't datapackage, but w3c's Model for tabular data and metadata on the web, which we use through (our own) csvw package. So this format does use JSON-LD for metadata.
What makes me interested in DataCrate, is that it may provide a standardized way to put a web catalog on top of dataset collections we currently aggregate using ZENODO Communities.
Anyway, I'll have a play with calcyte (although I'm saddened to see that the python implementation seems to have been discontinued :) ) and see where I get.
We could look at resurrecting the python version of calcyte, at least for going from CATALOG.json to HTML.
Closing this - please feel free to get back to us if needed.
ok. will keep an eye on DataCrates - and the other half of the globe.
DataCrate looks like a useful addition to CLDF (https://github.com/cldf/cldf/blob/master/README.md) a data format for cross-linguistic data, based on CSVW. When specifying CLDF we intentionally left out the "zipped or not" and archiving aspect, considering that there are options like bagit out there. Do you intend to provide support for turning other package formats like CSVW or datapackage into DataCrate semi-automatically? If such support is implemented in python, I could also help with the effort.