OASIS Mailing List ArchivesView the OASIS mailing list archive below
or browse/search using MarkMail.

 


Help: OASIS Mailing Lists Help | MarkMail Help

tosca message

[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]


Subject: RE: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0


I don't think anybody is considering anything other than full compatibility. The entire discussion is over what version to assign to this perfectly compatible thing. :-)

Thanks,
Paul 


-----Original Message-----
From: Newman, Matt [mailto:matt.newman@hp.com] 
Sent: Sunday, April 07, 2013 5:56 PM
To: Lipton, Paul C
Cc: Frank Leymann; Doug Davis; Probst, Richard; tosca@lists.oasis-open.org
Subject: Re: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0

A v2 that does not provide a smooth and seamless enhancement of existing content developed to v1 would be very problematic for adoption of v1

Content is king and the investment in content by us and our customers dwarfs the investment in the runtimes themselves....

Are we really considering that?

-- Matt - (408) 827-8194


On Apr 7, 2013, at 1:34 AM, "Lipton, Paul C" <Paul.Lipton@ca.com<mailto:Paul.Lipton@ca.com>> wrote:

Hi TOSCA-ers,

Now, responding to Frank's comment in near real-time:


A.    I would agree with him that v2 does constitute big changes.

B.    V2 does not by any means imply breaking changes.

I would also add:
C: Regardless of how we would distribute those big changes across normative and non-normative documents, we'd want those changes to sync, to have the same version number...

D. All those changes will "encourage" us to consider calling the release, likely many months from now, v2.

So, we can either punt the decision down the road and temporarily call the changes v.next, or we can decide sooner, but I'd hate to have the conversation twice. :-) I will put this on the agenda for next week. We will see where it takes us. I may not be on the call so you all can surprise me. :-)

In any case, I know this TC and I know a good decision is always reached.

Thanks,
Paul


From: tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org> [mailto:tosca@lists.oasis-open.org] On Behalf Of Lipton, Paul C
Sent: Sunday, April 07, 2013 10:22 AM
To: Frank Leymann; Doug Davis
Cc: Probst, Richard; tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>
Subject: RE: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0

Hi TOSCA-ers,

Please see Frank's comment below. He mentioned to me that  the Kavi email reflector didn't work for him, so I'm just acting as a repeater/proxy on this email.

Thanks,
Paul


From: Frank Leymann [mailto:Frank.Leymann@iaas.uni-stuttgart.de]
Sent: Saturday, April 06, 2013 5:27 PM
To: Doug Davis
Cc: Probst, Richard; Lipton, Paul C; tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>
Subject: Re: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0

Doug,

BIG changes in my understanding do not necessarily imply to "throw everything away and start from scratch", but to add "something significant" while fixing bugs in what exists in the former spec.

What we are discussion for "v.next" ( take this as a variable for the name of the succeeded spec(s) we produce) in fact is significant:  the plan portability API enabling portable management of cloud applications (in addition to provisioning and decommissioning in a declarative way we are focussed on in V1.0), the definition of basic types (required for portable declarative templates of cloud applications) - to name but a few.

To me, this is like adding features of active databases (like triggers or stored procedures) to SQL, which produced a new version of the SQL spec - of course without (!) breaking at all the existing functions of SQL.

But I might miss the point why we have this intense discussion about naming "the next spec(s)" we are going to produce in the TOSCA TC.

Gruss/Regards,
Frank


Am 06.04.2013 um 01:32 schrieb Doug Davis <dug@us.ibm.com<mailto:dug@us.ibm.com>>:


Its just the pattern I'm used to.  To me usually a major version number change implies BIG changes and that usually means breaking changes. Not a hard rule though.  :-)

thanks
-Doug
________________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.

<graycol.gif>"Probst, Richard" ---04/05/2013 07:23:10 PM---I am surprised to hear that calling something 'v2' implies incompatibility with v1. Is that common p

From: "Probst, Richard" <richard.probst@sap.com<mailto:richard.probst@sap.com>>
To: Doug Davis/Raleigh/IBM@IBMUS,
Cc: "Lipton, Paul C" <Paul.Lipton@ca.com<mailto:Paul.Lipton@ca.com>>, "tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>" <tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>>
Date: 04/05/2013 07:23 PM
Subject: Re: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0

________________________________



I am surprised to hear that calling something 'v2' implies incompatibility with v1. Is that common practice in OASIS?

If so, I would definitely vote to call these docs v.next or v1.x, because we have not found any reason to be incompatible, and we don't want anyone to assume that we have.

But again, I'm surprised by this assumption.

Best,
  --Richard


On Apr 5, 2013, at 2:21 PM, "Doug Davis" <dug@us.ibm.com<mailto:dug@us.ibm.com>> wrote:

I would actually prefer a version-less title because of the implications that including 'v2' in there will mean to people - mainly a set of breaking changes.  Or maybe I'm reading too much into presence of these docs all together and all we really care about is the new template and the editors will make the conversion using the existing v1 content.  In which case its definitely not a v2 - it more like a v1.errata since there are no normative changes - just an oasis template change.  Was this discussed during the call? Sorry if I missed it.

Here's a possible scenario... we find that the current spec is basically ok as is - there might be some minor tweaks that can be dealt with in an errata but overall its fine.  However, as you said, we need to define some concrete node types.  I'm not convinced that they would go into the current spec. They might go into a new document to highlight the separation of the tosca abstract notions from the concrete node types.  That doesn't require a tosca v2.  Or, perhaps we do stick it into the current doc but because the existing features are untouched, its all non-backwards-breaking-changes, this again doesn't mean its a v2 it could be a v1.1.

The key thing to me is that v2 implies a set of breaking changes and we're not there yet and I don't feel comfortable assuming we'll have any.  Rather than making a premature decision I'd prefer to see how things play out.  I think its far more important that we focus on interop testing of v1 and see where that leads us.

Let's discuss this on the next call.

thanks
-Doug
________________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.

<graycol.gif>"Lipton, Paul C" ---04/05/2013 05:01:58 PM---Hi Doug, OASIS Staff requested that we start with fresh templates for the spec and the primer, in pa

From: "Lipton, Paul C" <Paul.Lipton@ca.com<mailto:Paul.Lipton@ca.com>>
To: Doug Davis/Raleigh/IBM@IBMUS, "tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>" <tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>>,
Date: 04/05/2013 05:01 PM
Subject: RE: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0 Sent by: <tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>>

________________________________



Hi Doug,

OASIS Staff requested that we start with fresh templates for the spec and the primer, in part to make sure the format aligns with the latest OASIS format. We are required to provide a title and rather than opt for a version-less title or a confusing one, I selected a title that was consistent with a version designation that the entire TC has been using freely. This is not necessarily a permanent title by any means.

I suggest we live with v2.0 as a proper, but "working title" for these working drafts, which have no real standing anyway, prior to the first Committee Spec Draft. As you say, we can have that discussion prior to CSD at which point we'll certainly know if we need to change the version.

That said; co-chair hat off now, I think our frequent, albeit perhaps subconscious use of the term v2 was telling. Just look at the lack of base classes in the current spec that we intend to specify, and the already present large volume of documents that have been generated by the Interop SC. It seems extremely hard to believe that all of that constitutes an errata or v1.1, if we are to be honest with ourselves.

I hope that the situation and the lack of "title lock in" makes some sense to you and eases your concern.

Thanks,
Paul


From: tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org> [mailto:tosca@lists.oasis-open.org] On Behalf Of Doug Davis
Sent: Friday, April 05, 2013 4:31 PM
To: tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>
Subject: Re: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0


tosca-chairs,
during the call on Thursday we talked a lot of v2 activities and perhaps I should have given the terminology more consideration (or I apologize if this was discussed during the call and I missed it) but I was interpreting the talks to really be about v.next - meaning "whatever we do next".  Whether that turns out to be a full v2, a v1.x or just an errata is yet to be seen.  So, I was a bit surprised to see these "starter" documents get created with the assumption that we're immediately going to work on a v2.  I would prefer that we didn't make this assumption and waited until the TC made changes to our documents and _then_ we decided how best to release those - ie. as an errata, v1.x or v2.  I think its too soon to jump to the v2 conclusion.

thanks
-Doug
________________________________________________________
STSM |  Standards Architect  |  IBM Software Group
(919) 254-6905  |  IBM 444-6905  |  dug@us.ibm.com<mailto:dug@us.ibm.com>
The more I'm around some people, the more I like my dog.

<graycol.gif>Paul Knight ---04/05/2013 03:32:45 PM---Per the TC's submission request [1], please find the attached starter document for:

From: Paul Knight <paul.knight@oasis-open.org<mailto:paul.knight@oasis-open.org>>
To: tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>,
Date: 04/05/2013 03:32 PM
Subject: [tosca] Starter Document for Topology and Orchestration Specification for Cloud Applications Version 2.0 Sent by: <tosca@lists.oasis-open.org<mailto:tosca@lists.oasis-open.org>>

________________________________




Per the TC's submission request [1], please find the attached starter document for:
Topology and Orchestration Specification for Cloud Applications Version 2.0
WP-abbrev: TOSCA

We expect this Work Product to be published at:
http://docs.oasis-open.org/tosca/TOSCA/v2.0/csd01/TOSCA-v2.0-csd01.doc

Please let me know if anything here fails to meet your expectations.
Further revisions to this Work Product must be maintained in Working Drafts, following procedures detailed in the OASIS TC Administration How-to Guide [2].
Working Drafts should be made available by uploading the document(s) to the TC's Kavi document repository, or (provisionally/temporarily) to the TC's Subversion (SVN) repository, if SVN has been activated for your TC [3].  TCs are encouraged to use ZIP packaging when the WD releases contain multiple files (and directories).

For each WD revision, you will need to:
1) increment the Working Draft revision (number) from 01 to 02, 03, 04 etc., as successive Working Drafts are produced; the revision number needs to be updated at the top of the document in the stage identifier (Working Draft ##) and in the document identifier within the page footer.

2) supply the relevant publication/release/upload date for each successive Working Draft instance, using the  prescribed date format: DD Month YYYY; the date needs to be updated at the top of the document (just below the stage identifier, Working Draft ##) and in the page footer.

3) provide suitable text for a document Abstract, updating this summary with successive revisions to provide an accurate description of the subject matter, goals, scope, etc.

4) keep the Acknowledgments (Appendix A) and Revision History (Appendix C) up-to-date with each WD revision.

5) consult the OASIS Naming Directives document when creating new artifacts that may be part of the Work Product (e.g., image files, XML schemas), observing the rules for name characters in files and proposed directories, and for proposed URIs/namespaces [4].

When the TC votes [5] to approve a Working Draft as a Committee Draft (CSD or CND), the Chair or other designated person must submit a "Committee Specification Draft Creation and Upload Request" accessible on the TC Administration Requests Page [6].

Upon receipt of this form, the TC Administrator will QC and process the Work Product for official publication in the OASIS Library (http://docs.oasis-open.org/) as a Committee Draft, including addition of the requisite cover page metadata and other boilerplate information.

=========== References:
[1] https://tools.oasis-open.org/issues/browse/TCADMIN-1237
[2] Developing Committee Specifications and Notes
  https://www.oasis-open.org/resources/tcadmin/developing-committee-specifications-and-notes
  Starting a Working Draft
  https://www.oasis-open.org/resources/tcadmin/starting-a-working-draft
[3] SVN Version control, via Tools
  https://tools.oasis-open.org/
[4] OASIS Naming Directives
  http://docs.oasis-open.org/specGuidelines/ndr/namingDirectives.html
[5] Approval of a WD as a CD (CSD or CND)
  https://www.oasis-open.org/resources/tcadmin/approving-a-committee-specification-or-note-draft
  https://www.oasis-open.org/policies-guidelines/tc-process#committeeDraft
[6] TC Administration Requests Page, see Committee Specification Draft Creation / Upload Request
  https://www.oasis-open.org/resources/tc-admin-requests

Best wishes,
Paul
--
Paul Knight<mailto:paul.knight@oasis-open.org>  - Tel: +1 781-861-1013 OASIS<http://www.oasis-open.org/> - Advancing open standards for the information society Document Process Analyst<http://www.oasis-open.org/people/staff/paul-knight>
[attachment "TOSCA-v2.0-wd01.doc" deleted by Doug Davis/Raleigh/IBM]
---------------------------------------------------------------------
To unsubscribe from this mail list, you must leave the OASIS TC that generates this mail.  Follow this link to all your TCs in OASIS at:
https://www.oasis-open.org/apps/org/workgroup/portal/my_workgroups.php <graycol.gif>




[Date Prev] | [Thread Prev] | [Thread Next] | [Date Next] -- [Date Index] | [Thread Index] | [List Home]