Rob, I see that you tried to add a pdf to the "layered API proposal" that didn't make it (looks like size limitations). Can you work with Phil to get this posted?

Thnx,

--dmm

On Wed, May 8, 2013 at 9:11 PM, Rob Sherwood <rob.sherwood@...> wrote:

Hi all,

Apologies if I've been slow to reply, but it's been because I've been heads down reading code and meeting with our top engineers discussing the proposal.  Colin and Dave have done a great job throwing together something that looks good on paper, but given that BSN is going to bet the entire company on this project, it's important that we really drive down to a fine level of detail to convince ourselves that this makes sense and is feasible.

So, while I had talked with Colin a bit before the proposal came to light three days ago, only since then am I really starting to understand the implications of it, and I'm sorry to say that I have an increasing number of concerns about this merging proposal particularly around building on the SAL in its current form.

I'm doing my best to document these concerns and bring them to public discussion ASAP but some of them are very subtle (e..g, "yes, some of the function prototypes do _look_ the same, but the consistency guarantees are different, so this is a huge amount of work to port") and so it's a bit slow going.

The bottom line is that I'm rushing to get feedback on this and I'm very worried about people pre-supposing that "no news is good news".  While we're definitely behind the high level idea of a "best of breed" merged controller, I think we're going to have to weigh in with a different view on how to get there.  

I'll be working on this all night to see if I can get something in front of the TSC by tomorrow.

- Rob

.

On Wed, May 8, 2013 at 2:01 PM, Vijoy Pandey <vijoy.pandey@...> wrote:

Folks,

The Dixon-Erickson (DE) proposal has been built from the ground up looking at the various pros and cons of the two controller bases. I believe its unbiased, has solid technical merit, its open and frankly, its the best proposal we have.

I think we should start using this proposal as the guideline to actually start the code and merge the controller bases into one, and start writing applications around it - Anees, Colin and the team from IBM will start working on this direction right away.

-vijoy



Thomas Nadeau ---05/08/2013 01:51:18 PM---We intend to make our contribution (and start developing code) in the context of Dixon-Ericsson prop

From: Thomas Nadeau <tnadeau@...>
To: Colin Dixon/Austin/IBM@IBMUS,
Cc: "discuss@..." <discuss@...>, "tsc-bounces@..." <tsc-bounces@...>, "tsc@..." <tsc@...>
Date: 05/08/2013 01:51 PM

Subject: Re: [OpenDaylight TSC] [OpenDaylight Discuss]  Next Hackfest Dates
Sent by: tsc-bounces@...

---

We intend to make our contribution (and start developing code) in the context of Dixon-Ericsson proposal because it makes most sense to us. I'd like to move forward with this as soon as possible.

--Tom

I feel a bit sheepish after you saying "to be crystal clear" in the first sentence, but I'm still a bit confused.

Are you saying that you agree with
(1) the proposal that Dave Erickson and I put forward for how to merge the two controllers?
(2) the proposal that Chris Wright made to try to decide the issue before the hackfest?
(3) or both?

Thanks,
--Colin

tsc-bounces@... wrote on 05/08/2013 02:48:09 PM:
> From: Thomas Nadeau <tnadeau@...>
> To: Thomas Nadeau <tnadeau@...>, Abhishek Chauhan
> <Abhishek.Chauhan@...>, Rob Sherwood
> <rob.sherwood@...>, Phil Robb <probb@...>
> Cc: "tsc@..." <tsc@...>,
> "discuss@..." <discuss@...>
> Date: 05/08/2013 02:52 PM
> Subject: Re: [OpenDaylight TSC] [OpenDaylight Discuss]  Next Hackfest Dates
> Sent by: tsc-bounces@...
>
> A few have pinged me asking me to be crystal clear about what I
> said, so here goes:
>
> I personally have been following the discussions and presentations
> on the merge question and believe the proposal on the table from
> Chris et al answers the question in a sufficiently detailed,
> objective and sound manner for us to move forward. Based on that, I
> do not see any benefit in extending the time we need to decide on
> this proposal unless there is additional (substantive) discussion
> that materializes in the very near future.
>
> --Tom
>
> From: Thomas Nadeau <tnadeau@...>
> Date: Wednesday, May 8, 2013 3:26 PM
> To: Abhishek Chauhan <Abhishek.Chauhan@...>, Rob Sherwood <
> rob.sherwood@...>, Phil Robb <probb@...>
> Cc: "tsc@..." <tsc@...>, "
> discuss@..." <discuss@...>
> Subject: Re: [OpenDaylight Discuss] [OpenDaylight TSC] Next Hackfest Dates
>
> Thanks for clarifying my point. *)
>
> Adding discuss list too.
>
> --Tom
>
> From: Abhishek Chauhan <Abhishek.Chauhan@...>
> Date: Wednesday, May 8, 2013 3:22 PM
> To: Thomas Nadeau <tnadeau@...>, Rob Sherwood <
> rob.sherwood@...>, Phil Robb <probb@...>
> Cc: "tsc@..." <tsc@...>
> Subject: RE: [OpenDaylight TSC] Next Hackfest Dates
>
> I should have employed more precision below.
>  
> To be clear – I agree with Tom, that we should keep the current pace
> on the controller unification efforts, and not couple it with the hackfest.
>  
> -abhishek
>  
> From: Abhishek Chauhan
> Sent: Wednesday, May 8, 2013 12:14 PM
> To: Thomas Nadeau; Rob Sherwood; Phil Robb
> Cc: tsc@...
> Subject: RE: [OpenDaylight TSC] Next Hackfest Dates
>  
> I agree.
>  
> --
> abhishek
>  
> From:tsc-bounces@... [mailto:tsc-
> bounces@...] On Behalf Of Thomas Nadeau
> Sent: Wednesday, May 8, 2013 12:06 PM
> To: Rob Sherwood; Phil Robb
> Cc: tsc@...
> Subject: Re: [OpenDaylight TSC] Next Hackfest Dates
>  
>  
> I personally would rather not delay the decisions that need to be
> made until June.
>  
> --Tom
>  
>  
> From: Rob Sherwood <rob.sherwood@...>
> Date: Wednesday, May 8, 2013 1:50 PM
> To: Phil Robb <probb@...>
> Cc: "tsc@..." <tsc@...>
> Subject: Re: [OpenDaylight TSC] Next Hackfest Dates
>  
> Question:
>  
> There was some talk (suggested by Chris Wright, IIRC) about trying
> to coordinate the controller merge proposals evaluation/decision
> process with the hackfest.  Now that the hackfest dates are being
> moved out, do we still want to do this?
>  
> I would suggest yes, but I wonder what others think.
>  
> - Rob
> .
>  
> On Tue, May 7, 2013 at 2:21 PM, Phil Robb <probb@...> wrote:
> Hello TSC:
>  
> As you know we have been working on the dates for the next Hackfest.
> While we were quite close on one upcoming date, we have had multiple
> requests to schedule the next hackfest later.  The new proposed
> date/location for the next hackfest is:
>  
> When: June 6th and 7th, 2013.
> Where: Hotel Valencia at Santana Row (San Jose)
>  
> Pushing back the date provides a variety of benefits including more
> time for participants to make travel arrangements resulting in
> significantly higher attendance.
>  
> My apologies if this has caused anyone significant inconvenience.
>  
> TSC members, please let me know if you will be able to attend.
>  
> If there is reasonable representation from the TSC,  we will lock
> this date down and send a broader announcement to the full developercommunity.
>  
> Phil.
> --
> Phil Robb
> Director - Networking Solutions
> The Linux Foundation
> (O) 970-229-5949
> (M) 970-420-4292
> Skype: Phil.Robb
>
> _______________________________________________
> TSC mailing list
> TSC@...
> https://lists.opendaylight.org/mailman/listinfo/tsc
>  _______________________________________________
> TSC mailing list
> TSC@...

> https://lists.opendaylight.org/mailman/listinfo/tsc_______________________________________________
TSC mailing list
TSC@...
https://lists.opendaylight.org/mailman/listinfo/tsc

_______________________________________________
TSC mailing list
TSC@...
https://lists.opendaylight.org/mailman/listinfo/tsc

_______________________________________________
Discuss mailing list
Discuss@...
https://lists.opendaylight.org/mailman/listinfo/discuss

_______________________________________________
TSC mailing list
TSC@...
https://lists.opendaylight.org/mailman/listinfo/tsc