ietf
[Top] [All Lists]

Re: Is the datatracker authoritative (was: WG Review: Dynamic Host Configuration (dhc))

2013-08-31 08:06:34
On Aug 31, 2013, at 12:10 AM, S Moonesamy <sm+ietf(_at_)elandsys(_dot_)com> 
wrote:
announcement really ought to just point to the datatracker, since what's 
there is normative.

This is an individual opinion.  Please assume that the entire IETF disagrees 
with it.  The datatracker is not the authoritative version of a charter.  It 
is the message archived in the relevant mailing list which is considered as 
authoritative.  The rationale is related to note of record.

I have to admit that I did not expect my assertion here to be controversial.   
The assertion is certainly true about drafts, which we do not mail to mailing 
lists.   The charter that was mailed to the IETF mailing list is in fact the 
version that's in the datatracker, so you could argue that it's fine for the 
mailing list version to be normative.

However, you mentioned that there's no way to track versions of the charter in 
the datatracker, and this is actually not true—you can indeed track versions in 
the datatracker.   This is particularly useful since during the review process, 
changes are made.   So you have a version that goes out on the wg mailing list. 
  And then ultimately that gets reviewed by the IESG, and very typically the 
IESG wants changes.

In the DHC charter, there were three changes made during the IESG review 
process—two work items were separated out as their own bullet items that had 
been included in larger bullet items, and two bullet items were reordered and 
reworded to change the emphasis and restrict the scope of one of them.   Two of 
these changes were made to satisfy requests by IESG members.   The third change 
was made because one of the working group chairs noticed that one of the first 
two changes actually unintentionally excluded some work the chair anticipated 
that the working group should do.

You can actually see all of these changes by looking in the datatracker, but 
you would have to review the IESG mailing list and the dhc-chairs mailing list 
to find out about them otherwise.

It's worth pointing out that the datatracker is a _lot_ more capable than it 
used to be.   When I first started using it, I pretty much never used it, 
because it was almost always easier to get what I needed on the tools site.   I 
still find the tools site's UI easier to work with, but the datatracker has a 
lot of very useful functionality for tracking work as it moves through the IETF 
process.   I feel very lucky that the datatracker update happened just before I 
became an AD, because I don't know how I would have done my job without the new 
functionality.

So my statement was one of practicality, not one of process.   On a practical 
level, it's easiest to have one place to go to look for things, and not have to 
look all over the place to get a sense of what has gone on.   That's the 
datatracker.   If the IETF in fact feels that it should not be normative, I 
guess that's okay, but I think it's a lot more work for everyone.