Re: draft name changes

IMO, the internet-drafts upload tool should be capturing this information.

If you put in a name that ends in -00 and it's a draft-ietf name, it
should practically require the name of the I-D that is being replaced.
It's fairly probable that a WG document will have starte as a personal doc.

It's not just -00's that can replace other docs. I've had several docs
where a couple different drafts by different authors were merged into
one of the document streams.

	Tony Hansen
	tony <at> att.com

Julian Reschke wrote:
> Henrik Levkowetz wrote:
>> Right, good point and good idea.  One problem is that there's a lot of
>> missing information about which document replaces which.  I'm going to
>> try to synthethisise that using Lars Eggert's idea -- should have put
>> that in place some time ago, but no time...
>>
>> Anyway.  I've added this now, and the replaced-by links should appear if
>> the replaced-by information is available.
> 
> Would it be useful if the xml2rfc format captured that information somehow?
> 
> BR, Julian

Re: draft name changes

Tony,

> IMO, the internet-drafts upload tool should be capturing this information.
>   

Yes.

> If you put in a name that ends in -00 and it's a draft-ietf name, it
> should practically require the name of the I-D that is being replaced.
> It's fairly probable that a WG document will have starte as a personal doc.
>   

Right.

Jari

Re: draft name changes

>> IMO, the internet-drafts upload tool should be capturing this 
>> information.
>>
>
> Yes.
>
>> If you put in a name that ends in -00 and it's a draft-ietf name, it
>> should practically require the name of the I-D that is being replaced.
>> It's fairly probable that a WG document will have starte as a personal 
>> doc.
>>
>
> Right.

I'm bridling because of "practically require", and I'll grant that 
"practically require" could reasonably be:

   "Enter the name of the individual draft(s) being replaced:
      (or "none" if this working group draft has no predecessor drafts)"

But - individual draft NAMES -> working group draft NAMES is certainly most 
common practice, but please DON'T require a name of the I-D that's being 
replaced, because our WG BCPs do not require WG documents to start as a 
personal doc, and I have seen WG documents that sprang fully formed from 
thin air.

Another case where ALMOST everyone does something, but our WG BCPs don't 
require it - a lot of people are surprised that WGLC is not required by WG 
BCPs, either...

Re: draft name changes

On 2008-09-01 14:24 Spencer Dawkins said the following:

> I'm bridling because of "practically require", and I'll grant that 
> "practically require" could reasonably be:
> 
>    "Enter the name of the individual draft(s) being replaced:
>       (or "none" if this working group draft has no predecessor drafts)"

...

> My views of how badly most of our process BCPs need to be updated is no 
> secret, but evolving our processes via tool design is not the way we should 
> be going.

I agree.  If there are obvious ways for tools to make it easy to 'do the
right thing', that's fine, but they shouldn't pre-empt the decision process
and institute policy which hasn't been agreed and decided upon (by whichever
process we have).

	Henrik

_______________________________________________
Tools-discuss mailing list
Tools-discuss <at> ietf.org
https://www.ietf.org/mailman/listinfo/tools-discuss

Re: draft name changes

In the same line of thinking (making it easy to do the right thing
without enforcing any new processes), it seems to me that it would make
sense to allow a WG draft to have *more than one* precursor individual
drafts. 

Indeed, there are cases where a WG draft results from the "merger" of
individual drafts aiming at solving the same issue, or proposing very
similar solutions.  Forcing to state zero or one precursor would force
author to settle for zero (loss of information) or to arbitrarily choose
one of the precursors (the latter being possibly difficult if the merger
resulted from a non obvious consensus).

-Thomas

Henrik Levkowetz :
> 
> On 2008-09-01 14:24 Spencer Dawkins said the following:
> 
> > I'm bridling because of "practically require", and I'll grant that 
> > "practically require" could reasonably be:
> > 
> >    "Enter the name of the individual draft(s) being replaced:
> >       (or "none" if this working group draft has no predecessor drafts)"
> 
> ...
> 
> > My views of how badly most of our process BCPs need to be updated is no 
> > secret, but evolving our processes via tool design is not the way we should 
> > be going.
> 

Re: draft name changes

Henrik Levkowetz :
> Hi Julian,
> 
> On 2008-08-31 22:02 Julian Reschke said the following:
> >
> > Would it be useful if the xml2rfc format captured that information somehow?

I would support this.

> Mmm.  I'm not sure.
> 
> The intention has been that when we have Bill's rewrite of the draft
> submission tool deployed, we should consider adding fields in the form
> to capture information like this.  
> 
> Capturing it in the xml2rfc format would mostly be appropriate if xml2rfc
> also had an option to render it in a standard way in the draft itself.
> For drafts which are submitted with an accompanying xml file it would be
> possible to pre-populate the replaced-by field in the submission meta-data
> verification page based on a replaced-by indication in the xml file, but
> it would never be a complete solution which could do away with the form
> field.  So once the metadata verification page of the submission tool has
> a field for 'replaced-by', having it also in the .xml file would make
> things a bit easier for xml2rfc users.  Before then I think it would not
> make a big difference, I think.

If, at some point, it makes it easier for xml2rfc users, then wouldn't
it be nice to work on a such new field today ?

Also, independently of the submission tool, adding this type of

Re: draft name changes

Henrik Levkowetz :
> So once the metadata verification page of the submission tool has
> a field for 'replaced-by'.

By the way the submission tool field has to be named in terms of "draft
being replaced: " rather than "replaced by", or that will certainly
cause some confusion (usability nitpicking).

Cheers,

-Thomas

Re: draft name changes

On 2008-09-01 16:18 Thomas Morin said the following:
> Henrik Levkowetz :
>> So once the metadata verification page of the submission tool has
>> a field for 'replaced-by'.
> 
> By the way the submission tool field has to be named in terms of "draft
> being replaced: " rather than "replaced by", or that will certainly
> cause some confusion (usability nitpicking).

Quite so.

	Henrik

Re: draft name changes

I did not mean to imply that you have to give an individual draft name 
as the predecessor of the wg-00 draft. But I do want that question to be 
asked upon submission time.

Jari (who has sent in countless mails to the secretary to add a history 
link between drafts)

Re: draft name changes

Spencer Dawkins wrote:
> 
> I'm bridling because of "practically require", and I'll grant that 
> "practically require" could reasonably be:
> 
>    "Enter the name of the individual draft(s) being replaced:
>       (or "none" if this working group draft has no predecessor drafts)"

Spencer, this is exactly the sort of thing I was thinking of. It
requires you to think about the question and provide an answer one way
or another.

> But - individual draft NAMES -> working group draft NAMES is certainly most 
> common practice, but please DON'T require a name of the I-D that's being 
> replaced, because our WG BCPs do not require WG documents to start as a 
> personal doc, and I have seen WG documents that sprang fully formed from 
> thin air.

Exactly why it can't be "require". There's *always* an exception and
whatever we put in to place *must* allow for such exceptions.

	Tony Hansen
	tony <at> att.com