Personal vote: doesn't need to be in the protocol if we can't think of a good use beyond the backend of a service. I have stuck this on the agenda for Thursday where we can last-call it for good. Thanks; -jason On 5/5/10 2:02 AM, Jeff W. Boote wrote:
Right, and that is not in the protocol. Does anyone have an issue with not allowing merge-chaining in protocol messages?
thanks, jeff
On Apr 19, 2010, at 7:30 AM, romradz@man.poznan.pl wrote:
So I see that the only usage for merge chaining is store metadata file (for example store metada file of MA)
Roman
On Fri, 16 Apr 2010, Jeff W. Boote wrote:
On Apr 16, 2010, at 10:09 AM, Roman Lapacz wrote:
Yes, we can say that we are removing merge chaining from the NMC but in fact (in real examples) it will be still there as NM structures will be used in messages defined by NMC.
Ok, this was definitely not my impression. I was saying we should specifically define NMC protocol messages to use a sub-set of NM-WG schema that does NOT use merge chaining.
jeff
Roman // Roman Lapacz, PSNC Poland // phone: (+48 61) 858 20 24 On Fri, 16 Apr 2010, Aaron Brown wrote:
To alleviate some confusion on my end, are we removing merge chaining from the NMC protocols? I thought that was what Jeff meant when he talked about removing merge chaining, but based on the emails, I'm not sure if that was how others interpreted it. I'd would be wholly in favor of its removal. Cheers, Aaron On Apr 16, 2010, at 3:04 AM, Roman Lapacz wrote:
I propose to continue the discussion on merge chaining (and examples) on NM-WG mailing list (not to leave it for far future). I'll be waiting there for your opinions I asked ealier.
Roman
On Thu, 15 Apr 2010, Jason Zurawski wrote:
All;
On the call today everyone was in agreement that merge chaining examples
do not need to be in the NMC document. They will continue to be
described in NM however. If anyone disagrees, speak up now.
Thanks;
-jason
On 4/14/10 3:07 PM, Jason Zurawski wrote:
Hi Roman;
I agree. I will add this to the agenda for tomorrow.
-jason
On 4/13/10 7:12 AM, Roman Lapacz wrote:
Hi Jason& all,
I'm reading merge chaining examples in the base doc and I'm not sure
that keeping partial metadata elements, which in my opinion are useless
after chaining oparation, is a good approach. I don't believe that they
might be used for any further processing. Keeping them in the examples
presenting merged structures may be confusing for a reader. I would
remove them. What do you think (see examples in the attachements)?
Roman