Re: Running code (Was: I-D Action: draft-ietf-6man-ipv6only-flag-03.txt)

["Joel M. Halpern" <[e-mail]>]

The IDR working group does have requirements on having implementations 
of drafts before they complete their work on the draft.

I would prefer not to extend that as a requirement to other working 
groups.  Treating it as desirable, something we ask for, etc. is fine 
and makes sense.

In my case, I am not in a position (and have not been for many years) to 
implement these things myself.  I do work with the implementors.  While 
I can not commit for my product line folks who set schedules, the 
primary reason i have for working on a draft is that it is something we 
want to put in our product.  Getting resources to put it in even a 
demonstration version NOW however is MUCH harder.  Sometimes it happens. 
  Usually, I can't do that.  And even if I can get it implemented early 
to check things, I may not be permitted to talk about that implementation.

Yours,
Joel

On 10/24/18 9:52 AM, Job Snijders wrote:
> Dear Jordi,
> 
> On Wed, Oct 24, 2018 at 1:08 PM JORDI PALET MARTINEZ <[e-mail]> wrote:
>> Then the vendors that have the power to implement everything they
>> want, definitively win, only because they are able to implement it,
>> not necessarily because it is good or bad, and the argument of the
>> implementation will be a tool to reach consensus, even if it takes
>> longer, more versions, etc. This is real life.
> 
> I wouldn't position yourself as 'powerless': you can fork any Linux or
> BSD IP stacks yourself and make modifications - or you can hire people
> to do so, or work with students. We are all vendors if we choose to be -
> we all have this "power". And even if it takes longer, if drafts need to
> go through more revisions - it usually is a sign the quality and
> readability are increasing.
> 
>> Very good ideas that don't become standards because during the ID
>> development don't get implemented (even if they will become
>> implemented once they become RFCs), lose.
> 
> Can you point me to a protocol extension specification that was a
> valuable publication, specifically as RFC, but does not have any
> implementations? I see no harm for such drafts to wait for publication
> until some implementation is done.
> 
>> I'm sure there have been already many IETF documents that have been in
>> both situations. Market decides at the end. But that's the thing: at
>> the end, most of the time only when there is an RFC.
>>
>> A couple of examples I can remember, IPv6 site local (it was an RFC)
>> and draft-ietf-ipv6-dns-discovery, different process stages, both got
>> implemented by many OS vendors and also Linux. No longer used.
> 
> No longer used? Consider proposing them to be deprecated! Pruning
> specifications that turned out to be "not the best idea" is also an
> important part of protocol maintenance.
> 
>> I'm sorry but it is discrimination.
> 
> I really don't see it, for more information on what discrimination means
> to people please take a look at https://en.wikipedia.org/wiki/Discrimination
> 
> I suspect you are using such strong words to make an appeal to emotion.
> 
> Kind regards,
> 
> Job
> 
> --------------------------------------------------------------------
> IETF IPv6 working group mailing list
> [e-mail]
> Administrative Requests: https://www.ietf.org/mailman/listinfo/ipv6
> --------------------------------------------------------------------
>