Proposed process for finalizing incubating specs

[Matthew Marum]

During the OpenSocial Summit back in February, there was wide agreement that new features going into the specification would be incubated until there is significant implementation experience with them.

In the specification roadmap, it says "all major specification changes will be incubated until such time that the community has made significant implementations of these incubating changes and provided feedback. While there is wide agreement in principal, the Community still needs to define and vote on this new process."

http://docs.opensocial.org/display/OSD/OpenSocial+Specification+Roadmap 

As we close out OpenSocial 2.5, I think we are past due on defining this process.

In OASIS, one of the requirements in order for a Committee Specification to mature into an OASIS Standard is for "three Statements of Use explicitly referencing the Committee Specification".

https://www.oasis-open.org/policies-guidelines/tc-process#OASISstandard 

So using OASIS as a guide, I propose that one requirement for finalizing incubating specification features is for there to be 3 'users' of that feature with all the opportunity for those users to provide feedback.  The other requirement is another vote requiring 5 +1 votes and no -1 votes to pass.

Any thoughts, comments?  If there aren't any concerns, I can formalize this proposal into a wiki so we can socialize it better and vote on it as being our incubating process from now on.

[Ryan Baxter]

+1 as long as 'users' also means implementers.

[Matthew Marum]

Yes, by users I mean any party using the relevant section of spec. An implementation that goes into a product would count. A half finished implementation that goes unused does not.

--
You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.
To post to this group, send email to opensocial-an...@googlegroups.com.
To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/0fMnr483FdsJ.
For more options, visit https://groups.google.com/groups/opt_out.
 
 

[James M Snell]

I would add a requirement that all discussion about the statements of use and voting be done publically on the list, giving everyone a chance to review and discuss. But otherwise +1

--
You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.
To post to this group, send email to opensocial-an...@googlegroups.com.
To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/-gPn-ck994AJ.

[Matthew Marum]

So in the past we've done basically all proposal voting on the revisions within the specification project which is disconnected from this mailing list.  Are you proposing that we do all voting via this mailing list or just in the cases where we are finalizing incubating spec?

[Matthew Marum]

What do people think about making it a requirement that in order for an incubating feature to become finalized that one of the implementations/users must be Apache Shindig?

[Henry Saputra]

You get +1 from me of course :)

On Tuesday, July 31, 2012, Matthew Marum <mgm...@gmail.com> wrote:
> What do people think about making it a requirement that in order for an incubating feature to become finalized that one of the implementations/users must be Apache Shindig?
>

> --
> You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.
> To post to this group, send email to opensocial-an...@googlegroups.com.
> To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.

> To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/KJWUxurlG-EJ.

[Dan Dumont]

+1

--

You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.
To post to this group, send email to opensocial-an...@googlegroups.com.
To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/KJWUxurlG-EJ.
For more options, visit https://groups.google.com/groups/opt_out.

[Ryan Baxter]

As much as I am in favor of keeping Shindig the main reference implementation I am a little concerned about how specific this is.  Is this something we can easily update if we end up with other implementations?

On Monday, July 23, 2012 4:38:42 PM UTC-4, Matthew Marum wrote:

[Henry Saputra]

Hey Ryan, what do you mean by other implementations?

- Henry

> --
> You received this message because you are subscribed to the Google Groups
> "OpenSocial and Gadgets Specification Discussion" group.
> To post to this group, send email to
> opensocial-an...@googlegroups.com.
> To unsubscribe from this group, send email to
> opensocial-and-gadg...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/2hB0WpdnqhkJ.

[Matt F.]

I am also hesitant to tie process language to any specific implementation; including Shindig.  I think Shindig needs to be 100% spec compliant; but, that is really the Shindig community's decision and not something that the OpenSocial Foundation can directly impose. 

[James M Snell]

+1

--
You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.
To post to this group, send email to opensocial-an...@googlegroups.com.
To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/e0YPsK3iRGIJ.

[Matt F.]

That is a good question.  If we went with this process, what counts as an implementation?  IMO, Rave, Jive, Lotus Notes & Connections are all different users; all of which use Shindig.

> opensocial-and-gadgets-spec@googlegroups.com.

> To unsubscribe from this group, send email to

> opensocial-and-gadgets-spec+unsub...@googlegroups.com.

[Matthew Marum]

I agree, I think we need to be clear about what counts.  Is a user a vendor, product, or a code base?

Matt

[Matthew Marum]

Could we say that 1 implementation must be open source so that the community has an option they can freely adopt?  Without saying it must be Shindig, Rave, or any other project in particular?

Matt

[Franklin, Matthew B.]

+1

 

--

You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.

To post to this group, send email to opensocial-an...@googlegroups.com.
To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/LABiF42H0moJ.

[Henry Saputra]

Well, it does not have to be in Shindig I suppose.

We could add the language to the spec such as the implementation has
to be open source.

So that would include Apache Shindig, or Negroni or Apache Rave/

- Henry

> --
> You received this message because you are subscribed to the Google Groups
> "OpenSocial and Gadgets Specification Discussion" group.
> To post to this group, send email to
> opensocial-an...@googlegroups.com.
> To unsubscribe from this group, send email to
> opensocial-and-gadg...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/e0YPsK3iRGIJ.

[Ryan Baxter]

Sounds good Matt.

[Matthew Marum]

I've written up the proposed OpenSocial Incubating process that we've discussed here.  I've posted it to the OpenSocial documentation wiki.

http://docs.opensocial.org/display/OSD/OpenSocial+Incubating+process

Let me know what you think.  Everyone should feel free to provide input.

[Ryan Baxter]

This looks pretty good Matt, although I am not sure Open Source needs to be capitalized.

[Matthew Marum]

Changed that.  Thanks Ryan.

Any other comments?

Matt

[Franklin, Matthew B.]

I do that all the time… Too much OpenSocial on the brain.

 

--

You received this message because you are subscribed to the Google Groups "OpenSocial and Gadgets Specification Discussion" group.
To post to this group, send email to opensocial-an...@googlegroups.com.
To unsubscribe from this group, send email to opensocial-and-gadg...@googlegroups.com.

To view this discussion on the web visit https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/3ZlbCpK3CvEJ.

[Henry Saputra]

+1 from me.

- Henry

> --
> You received this message because you are subscribed to the Google Groups
> "OpenSocial and Gadgets Specification Discussion" group.
> To post to this group, send email to
> opensocial-an...@googlegroups.com.
> To unsubscribe from this group, send email to
> opensocial-and-gadg...@googlegroups.com.
> To view this discussion on the web visit

> https://groups.google.com/d/msg/opensocial-and-gadgets-spec/-/276dqFR7ijgJ.