Re: [Tools-discuss] Replaced-by inconsistencies

[Henrik Levkowetz <[e-mail]>]

Hi Adam,

The explanation is that the generated html files aren't always up-to-date,
and the reason for that in turn is a bit more complex:  

   1) The load (number of accesses) on the html converted drafts and
   RFCs is sufficiently high that the 4 servers currently available
   wouldn't be able to handle it if the html version was dynamically
   generated in order to be fully up-to-date,

   2) I try to keep the update frequency low, as I don't want search
   engines to believe that they have to check the whole 65,000+ corpus
   every day in order to catch changes...

This means that 

   a) I have a very long re-generation period of ~3 months -- unless
   something about the document state changes, a document is only re-
   generated if it is older than 3 months.  That re-generation is done
   mainly to catch updates to the conversion tool, which happens now
   and then (see http://tools.ietf.org/tools/rfcmarkup/changelog).

   b) In order to catch status changes in documents, an explicit re-
   generation has to be triggered when something changes which should
   be reflected in the html document.  Re-generation is now triggered
   when a new version of a draft appears, when a draft appears in the
   datatracker, when a document is published as an RFC, and in some
   other cases.

However, regeneration when new replaced-by information appears has not
up to now been triggered.  Which I'm now about to fix :-)


Best,

	Henrik
   
   

On 2009-07-18 12:50 Adam Roach said the following:
> It looks to me like the replaced-by information in drafts is being 
> applied inconsistently.
> 
> For example, if you look at:
> 
>   http://tools.ietf.org/html/draft-holmberg-sipping-199-04
> 
> The "Versions" line reads "00 01 02 03 04 draft-ietf-sip-199", which is 
> correct -- that is, because draft-ietf-sip-199 replaced 
> draft-holmberg-sipping-199, it appears after all the versions.
> 
> If you click on the draft-ietf-sip-199 link, you get taken here:
> 
>   http://tools.ietf.org/html/draft-ietf-sip-199-08
> 
> The "Versions" line links back to the draft that it replaced. Great!
> 
> Now, lets look at:
> 
>   
> http://tools.ietf.org/html/draft-ietf-sipcore-presence-scaling-requirements-01
> 
> It correctly links back to 
> draft-ietf-sipping-presence-scaling-requirements, which it replaced. So 
> far, so good. However, if you go to 
> draft-ietf-sipping-presence-scaling-requirements:
> 
>   
> http://tools.ietf.org/html/draft-ietf-sipping-presence-scaling-requirements-03
> 
> There's no link forward to the sipcore draft! The "Versions" line just 
> goes up to 03, and stops.
> 
> Certainly the tools should understand that the "replaces"/"replaced by" 
> relationship is reciprocal. Is this database corruption, or incomplete 
> data entry? It doesn't appear to be an isolated incident (cf., 
> http://tools.ietf.org/html/draft-ietf-sipcore-keep-00) -- I would 
> propose that an audit of the "replaces"/"replaced by" relationship data 
> is in order...
> 
> /a
> 
> _______________________________________________
> Tools-discuss mailing list
> [e-mail]
> https://www.ietf.org/mailman/listinfo/tools-discuss
>