Hi all, For Liberty I am working on a specification to update the way we deliver application developer information. [1] I'm incorporating Tom's suggestion to automate API reference information. [2] In revising the API reference specification I wanted to ask a few more questions to shape the spec further. * Should we do a proof of concept for say, Compute, then have the other projects follow? Identity v3 has the most calls in the core with 74, but Compute v2 plus extensions has over 120 calls. Alternatively, we could also limit the scope to "greater than 50% reported on User Survey [2]". * Is there any way to know what versions of an API a user answering the survey is running or do we have to scope to both versions for some of these services? Identity Compute Block Storage Image Networking * Do we continue to enable the project teams to indicate where they want to maintain REST API information? Or does this spec require particulars down to the folder name in the repo due to automation efforts? Feel free to reply here or in the spec itself. Thanks, Anne 1. https://review.openstack.org/#/c/177934/1 2. http://lists.openstack.org/pipermail/openstack-docs/2015-April/006502.html 3. http://superuser.openstack.org/articles/openstack-user-survey-insights-november-2014 -- Anne Gentle annegentle at justwriteclick.com -------------- next part -------------- An HTML attachment was scrubbed... URL: <http://lists.openstack.org/pipermail/openstack-dev/attachments/20150505/f9a3f45d/attachment-0001.html>