Re: [Tools-discuss] Comments on draft-ietf-tools-draft-submission-08.txt

On Tue, 2005/05/10 (MDT), <matthew@elvey.com> wrote:

> Does the draft take into account that it is probably much easier for the  
> Secretariat to convert a document from one format to another (especially  
> given appropriate instructions or better yet standard procedure  
> regarding use of existing tools) than it is to compare two documents to  
> confirm that they are not substantially different?  Or am I mistaken?

Secretariat is not and will not be manipulating submitted drafts, including
format conversion. While it may be easy to convert from one format into
another (and the draft requires such conversion from XML sources), the
danger of autoconversion is that the result does not match the author
expectation.

For example, whenever a new version of xml2rfc is out, there
is often at least one complaint that the new plain text output differs
 from the old plain text output, and the difference breaks something
important in the author's draft.

I believe it is better to have in imperfect auto-comparison (and risk
a few rogue drafts to be published) than an imperfect conversion
(and risk many corrupted drafts to be published). The current Toolset
requirements reflect that belief and give the submitter a choice:
let the Toolset to auto-convert from submitted XML to plain text or
let the Toolset auto-compare submitted XML and submitted plain text.
Personally, I think I will use the latter mode more often.

> Should one of the validation steps be approval (a good score from) an  
> anti-spam system such as SpamAssassin?  I guess this can be a feature  
> added later if it turns out it's needed.  I'd bet it will be.

I do not see why an email draft submission that SpamAssassin marks as
spam but that also passes all IETF-specific checks should be rejected.
For example, I doubt spam will come with IETF-compliant IPR disclosures
any time soon.

However, I agree that adding spam or other filters to the email interface
would be possible if such filters are needed.

Thank you,

Alex.

_______________________________________________
Tools-discuss mailing list
Tools-discuss@ietf.org
https://www1.ietf.org/mailman/listinfo/tools-discuss