Re: [dxwg] Conneg editorial (#1132)

Can we get rid of the instances where header-based negotiation is called 
"http"? Query strings are part of http, too. Browsers use http, 
including headers. And query strings are as likely to be used 
programmatically as in a browser, so I would suggest we stick with 
header-based negotiation and query-string-based negotiation. Here I 
suspect you meant to refer to all the non-header-based approaches as a 
single entity. I'm not sure how best to refer to that, but maybe it 
would work to call it URL-based negotiation.

-Annette

On 11/5/19 3:57 AM, Karen Coyle via GitHub wrote:
> I think my edits are in there, although there seem to be line number 
> changes since I made my comments so I may have missed some.
>
> I would really like to fix this run-on sentence:
>
>> This specification provides functional profiles of the abstract model 
>> for two environments: HTTP and human browser (Query String
>        Arguments) as well as a functional profile that specifies a 
> single action - confirming which profile(s) a server response conforms 
> to -
>        that is relevant to multiple environments, such as the HTTP and 
> human browser environments.
>
> and suggest (if the concepts have been captured):
>
> This specification provides functional profiles of the abstract model 
> for two environments: HTTP and human browser (Query String
>        Arguments).  It also provides a functional profile that has the 
> single function to respond with which profile(s) a server response 
> conforms to, and which can be used in both environments.

-- 
Annette Greiner (she)
NERSC Data and Analytics Services
Lawrence Berkeley National Laboratory

Received on Tuesday, 5 November 2019 20:58:06 UTC