On Wed, Aug 01, 2018 at 09:42:30AM +0200, Abdussalam Baryun wrote: > On 7/31/2018 10:14 AM, Ted Lemon wrote: > > And I agree with you that the IETF last call review process is badly > broken. The problem is that it becomes an N^2 algorithm when in > principle every IETF participant is on the hook for every document. Its the perfect DHT with a random self-elected hash function on each member. Scales O(rnd^rnd). Aka: Problem really is that IETF is very lazy in drafting members to do reviews. If we wanted to adopt any of Lloyds ICO ideas, then it should be coins you collect for reviews and have to spend for documents you write. Everybody wants to come to IETF with their document, nobody wants to review (from 50,000 feet perspective). > CP> I really did not mean to suggest that. A lot of IETF participants > would not be on the hook for *any* documents. But at the time a document > is adopted by a WG, a group of reviewers should be known for that document (in > addition to the document authors). > > AB> I agree too with both, and I draft-readiness should be managed/directed > by the AD of the WG. The AD and WG chair must be responsible when the draft > is ready for WGLC. https://trac.tools.ietf.org/tools/ietfdb/ticket/2563 > AB> Even WG discussions should be assisted, some times it is not easy to > discuss with some authors they think the draft is their work only. This is easily resolved by WG chairs. Unless you have to fear collusion between WG chairs and authors, in which case you need to escalate to AD, etc. pp. > > AB> I suggest that the ADs appoint one participant to be against authors > per draft per developing it per WG, to make discussions on the list > (against does not mean always against but means that he/she is responsible > to say no/why-no when needed). I remember n the graduation days we use this > way to make discussions and better work. Sure, if we start to hand out IETF Diplomas or coins collected for opposing, otherwise you will only get the feedback from participants that they are personally or through their company interested in expressing. > AB> The AD should find a way to make participants discuss more deep into > each work. The AD should work efficiently and use all IETF resources, to > help optimisation and performance. We're working on cloning technologies for ADs to bring available AD cycles to the level needed to support this requirement. Cheers Toerless > Best Regards > AB -- --- tte@xxxxxxxxx
