- From: Shane McCarron via GitHub <sysbot+gh@w3.org>
- Date: Thu, 11 Aug 2016 12:28:18 +0000
- To: public-annotation@w3.org
halindrome has just created a new issue for https://github.com/w3c/web-annotation: == Legal IRI for Annotation Containers == In the Annotation Protocol spec, the language reads: > Annotation Containers may have any IRI, but must end in a "/" character. Now, first this is not great spec text. It should be more precise. But I know what you mean, and any reasonable person will know what you mean. However, an IRI has a lot of components. One way to interpret this is that the literal string representing the IRI must end in a solidus (U+002F). Another way to interpret this is that the path component of the IRI must end in a solidus. I have decided that the most liberal interpretation is required of the spec for 2 reasons. One, we are already in CR and constraining it further would require a restart. Two, it makes it much much easier to implement in the test suite. So, my question for the group is: can I use the following as a container IRI - "client/foo.py?path=annotations/" I could use an interpretation on this as soon as possible. Please view or discuss this issue at https://github.com/w3c/web-annotation/issues/344 using your GitHub account
Received on Thursday, 11 August 2016 12:28:25 UTC