On Mon, Apr 21, 2014 at 6:21 AM, Jamal Hadi Salim <[email protected]> wrote:
> Dean, > > On Mon, Apr 21, 2014 at 9:01 AM, Dean Bogdanovic <[email protected]> > wrote: > > Jamal, > > > > Majority of the people are using today NC/RC/YA (because it is > available) and it provides necessary >functionality. Many people, including > myself, didn't find any issues so far with it and we believe it is >the > right choice. > > You have NOT implemented the spec to justify that technical arguement. > As an example, In your last posting you said latency and throughput > was important. > Jan claims they are irrelevant. Is it a requirement? My answer and your is > yes. > > > If you remember, at the beginning I mentioned two criteria, technical > and business. > > I know "business" reasoning is a BIG driver; unfortunately often made > to appear like the > reasoning is technical (Ive said before i appreciate your sincerity). > Dean, I think you should look at this with "bussiness" outside the picture. > If you were to be given a chance to get this right from a technical > pov - what would > you do? There is an "E" in IETF (no "B"); yes i am trying my best to > be pragmatic > and not ignoring the "B" - but only when all is equal that becomes a > clear decider. > To be noted again: I am not even pushing for ForCES, but it definetely > aint netconf > or restconf - sorry. > > > NC/RC/YA are satisfying both of them for lot of people in this group. > > You are trying to explain why ForCES is better then NC/RC/YA and I > appreciate it, but it is not better >for my case and why should I go ahead > and implement ForCES on the company platform, when >there is no request > from the market and the existing tools are working satisfactory. > > > > sure - so what is wrong then with letting you go and do restconf and > me do ForCES. > Lets just have the "market" decide? All i am asking for is the process > be followed or > have the tyranny of the majority take over. > > It looks to me like the market already has decided. NETCONF started in 2002 and was first published as an RFC in 2004. Now, 12 years later, it is finally being significantly deployed. (Delay is mostly because converting legacy systems from ad-hoc hand-made CLI is hard.) Will it take 10 more years for vendors to deploy ForCES? Are any vendors shipping it now? If not, how soon? Ignoring the tool-chains and the history and claiming that both NETCONF/YANG and ForCES are at the same level of maturity is not really fair. Andy > People do not have to explain their vote, they can just cast it, as it is > their opinion. > > > > There is a difference between a vote and a discussion. I believe we > are in discussion > mode right now. In discussion mode people voice opinions not cast a ballot. > > cheers, > jamal >
_______________________________________________ i2rs mailing list [email protected] https://www.ietf.org/mailman/listinfo/i2rs
