)]}'
{"specs/pike/approved/capabilities-kickoff.rst":[{"author":{"_account_id":2750,"name":"Sean Dague","email":"sean@dague.net","username":"sdague"},"change_message_id":"a0c87394469b62d359d72e06e943e0f37de300f4","unresolved":false,"context_lines":[{"line_number":118,"context_line":"                ..."},{"line_number":119,"context_line":"            },"},{"line_number":120,"context_line":"        }"},{"line_number":121,"context_line":"    }"},{"line_number":122,"context_line":""},{"line_number":123,"context_line":"Security impact"},{"line_number":124,"context_line":"---------------"}],"source_content_type":"text/x-rst","patch_set":1,"id":"bff0334d_962d8eff","line":121,"updated":"2017-04-12 14:41:43.000000000","message":"so, I do kind of wonder about the following: http://blog.novatec-gmbh.de/the-problems-with-swagger/ where the suggestion is about using the HATEOS approach of links.\n\nI do realize that links kind of fell out of vogue for us, but in that world this would look like:\n\n{ \n    \"server\": {\n       \"id\": \"some-uuid\",\n       \"links\": [\n            {\n                \"rel\": \"snapshot\",\n                \"href\": \".../action\",\n                \"type\": \"POST\"\n            },\n            {\n                \"rel\": \"attach_port\",\n                \"href\": \".../action\",\n                \"type\": \"POST\"\n            }\n            ...\n       ]\n}\n\nWhile the href links aren\u0027t great at the moment, because we overload them for actions, they would open up some more interesting patterns in the future. It feels like it\u0027s at least worth discussing. \n\nI think we would also want to express in our contract that links are never considered permanent, and that you always need to fetch the resource to ensure fresh links.","commit_id":"7895f7f29948fa57f591e55bd922d7b8c118a063"}]}