Re: Interoperability between Mutable and Immutable Versioning

Yaron Goland (yarong@microsoft.com)
Thu, 21 Jan 1999 22:01:51 -0800


Message-ID: <3FF8121C9B6DD111812100805F31FC0D08792D7F@RED-MSG-59>
From: Yaron Goland <yarong@microsoft.com>
To: Chris Kaler <ckaler@microsoft.com>,
Date: Thu, 21 Jan 1999 22:01:51 -0800
Subject: RE: Interoperability between Mutable and Immutable Versioning

As Geoffrey pointed out in one of his posts the definition of branch used by
the versioning authors is different than the one I presented in my post
which the authors refer to as forked. I read Geoffrey's definition of branch
a number of times and I still do not understand it. I would appreciate
further clarification so I can understand the arguments being held between
the authors.

			Thanks,
					Yaron

> -----Original Message-----
> From: Chris Kaler [mailto:ckaler@microsoft.com]
> Sent: Thursday, January 21, 1999 9:31 AM
> To: 'Geoffrey M. Clemm'; ietf-dav-versioning@w3.org
> Subject: RE: Interoperability between Mutable and Immutable Versioning
> 
> 
> 
> Personally, I like the THAW/FREEZE idea we discussed because 
> it requires an
> overt action to replace an older revision.  I guess I don't 
> really get what
> it means to "checkout" an old revision that will be replaced 
> on checkin and
> not branched.  To mean they seem like dissimilar things that need to
> interoperate.  Doing a THAW, GET, PUT, FREEZE seems a viable 
> approach for
> DMS systems that need to do this.  I still think the mainline 
> case is that
> you checkout/in and create immutable revisions.  The 
> THAW/FREEZE is just a
> way to support DMS scenarios within the versioning model.
> 
> Chris
> 
> -----Original Message-----
> From: Geoffrey M. Clemm [mailto:gclemm@tantalum.atria.com]
> Sent: Thursday, January 21, 1999 8:42 AM
> To: ietf-dav-versioning@w3.org
> Subject: Interoperability between Mutable and Immutable Versioning
> 
> 
> 
> To clarify my position on this topic, it is my belief that 
> the protocol
> could be designed in a way to make mutable and immutable versioning
> incompatible (i.e. by providing a THAW and FREEZE operations that
> would fail on versioned resources that supported immutable revisions),
> or could be written in a way to make them compatible (i.e. by 
> providing
> a consistent interpretation of CHECKOUT/CHECKIN for both.
> 
> The CHECKOUT/CHECKIN model I posted in my earlier message is the
> latter (i.e. interoperable) protocol.  In this form, I not only
> believe that mutable versioning is acceptable, but I advocate it as an
> interoperable simplification of the more powerful but more complex
> configuration management protocol.
> 
> Cheers,
> Geoff
>