[controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread Colin Dixon
During last week's Kernel projects call [0], I asked if and when we wanted to deprecate the config subsystem. During the conversation, I think everyone agreed that we should strongly discourage people from building new projects based on it and encourage people to move toward Blueprint, which sounds

Re: [controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread Colin Dixon
Yes. Sorry, that should have been Nitrogen. The pushback was even to removing it in Nitrogen. --Colin On Tue, Nov 15, 2016 at 12:20 PM, FREEMAN, BRIAN D wrote: > It seems like you need at least one release for deprecation before > removing a function. I know I have code that uses the Config su

Re: [controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread FREEMAN, BRIAN D
It seems like you need at least one release for deprecation before removing a function. I know I have code that uses the Config subsystem so it would be a real pain to move in one release and creates an upgrade nightmare for me that would slow down my migration to Carbon. Brian From: controll

Re: [controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread thomas nadeau
I also thought we had an official project policy for deprecating functions only after several major releases to avoid the scenario Brian describes. tom > On Nov 15, 2016, at 12:20 PM, FREEMAN, BRIAN D wrote: > > It seems like you need at least one release for deprecation before removing a >

Re: [controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread Alexis de Talhouët
I think the options are the following ones: Option 1: Deprecation notice in Carbon Removal in Nitrogen Option 2: Deprecation notice in Carbon Adaptation in Nitrogen Removal in Oxygen Option 3: Deprecation notice in Nitrogen Adaptation in Oxygen Removal in Fluorine Or Option 4 … being whatever w

Re: [controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread Robert Varga
On 11/15/2016 10:08 PM, Alexis de Talhouët wrote: > > Option 2: > Deprecation notice in Carbon > Adaptation in Nitrogen > Removal in Oxygen > > During the kernel meeting I was more after option 3, but I think option > 2 make sense so downstream consumer can be notified earlier in the > process an

Re: [controller-dev] deprecating the config subsystem in Carbon?

2016-11-15 Thread FREEMAN, BRIAN D
I think Option 2 is a reasonable approach. Brian -Original Message- From: controller-dev-boun...@lists.opendaylight.org [mailto:controller-dev-boun...@lists.opendaylight.org] On Behalf Of Robert Varga Sent: Tuesday, November 15, 2016 4:46 PM To: Alexis de Talhouët; thomas nadeau Cc: con