Comments on use case "intelligently previewing CSV files"

Hi Davide - more good work on the intelligently previewing CSV files use case<http://w3c.github.io/csvw/use-cases-and-requirements/#UC-IntelligentlyPreviewingCSVFiles>

I really like the fact that you've brought out the ability to _sort_ the data in the preview as a key concern!

Paragraph 2; "Note that  the underlying data begins with" ... I downloaded the file myself, and was surprised to find that the header row indicating the column names is not present in the downloaded CSV until row number 3!

(e.g. Department Family               Entity    Date      Expense Type    Expense Area    Supplier               Transaction Number                 Amount in Sterling)

I imagine that there are all sorts of interesting details in this example, such as use of controlled vocabularies for "expense type" and "expense area", that could be mentioned.

Also, I see that the title of the table has been included in the CSV ... you've already picked up on this with the reference to requirement about annotation. Additionally, I think that it would be worth noting the metadata that is present on the web-page but _not_ included in the downloaded CSV ...


-          Resource title

-          Publisher

-          License

-          Abstract / description

-          Date last updated

Evidently this information is useful but cannot currently be included. This should be added to the use case.

Regards, Jeremy