Re: [renum] Working Group Last Call: draft-ietf-6renum-static-problem-01.txt

Hi Ran,

Thanks for the comments. We are in WG Last Call until Sept. 28 so
we will roll all the comments up at that time.

See my small comments in line below.

Regards
   Brian

On 19/09/2012 13:43, RJ Atkinson wrote:
> Hi,
> 
> 	While I am not a subscriber to this WG list,
> I have a few comments on draft-ietf-6renum-static-problem-01.
> I have tried to organise the comments by Section.
> 
> 
> Section 2.2, 1st Paragraph:
> 	For about a decade now, most network-capable printers 
> 	have silently and automatically enabled SLP without 
> 	requiring any user interaction.  Such printers often
> 	use a factory-default algorithm to generate a printer
> 	name, unless a user has explicitly configured a 
> 	printer name.
> 	
> 	Hosts that are SLP-aware, automatically discover such 
> 	printers and can use them, without requiring the printer 
> 	to have a preconfigured fixed IP address.  Many deployed 
> 	hosts have supported SLP for the past decade.
> 
> 	So, the statement that "It is also unusual to enable 
> 	the Service Location Protocol for the same purpose..."
> 	is not correct in the deployed world.  It was correct
> 	in the 1990s.  It has not been correct for about a
> 	decade, however.

Opinions on the extent of deployment seem to differ, judging by
some recent comments on the homenet WG list. I recently had occasion
to install an HP printer and it was necessary to install some proprietary
discovery software on the (Windows) hosts as well; so at least in that case
there was no SLP by default. But we can certainly remove "unusual"
and say something like "It is not universal to enable...".

> 
> Section 2.2, Last Paragraph:
> 	The DNS Service Discovery (DNS-SD) technology
> 	described in draft-cheshire-dnsext-dns-sd-11.txt
> 	is directly relevant and should be both mentioned
> 	and referenced in this paragraph.  mDNS is great,
> 	but for the problems of Section 2.2, mDNS really
> 	works much more effectively in combination with
> 	DNS-SD.  The RFC Editor web site indicates that
> 	both DNS-SD and mDNS are IN-QUEUE for publication
> 	on the IETF standards track.
> 
> 	Also, since around 2005/2006, many printers have
> 	shipped with mDNS/DNS-SD silently and automatically
> 	enabled (in addition to SLP).  Further, many Windows
> 	PCs, all Apple Macs, and many UNIX systems have
> 	silently enabled mDNS/DNS-SD support for printing.
> 	(NOTE: I'm told that when one installs "iTunes" onto 
> 	MS Windows, the installer silently installs mDNS/DNS-SD 
> 	support also, since mDNS/DNS-SD is how iTunes finds 
>         "shared libraries" of digital content.)  So the last 
>         paragraph also needs to be edited to clarify that the 
> 	mDNS/DNS-SD combination is now widely supported and 
> 	widely deployed for a range of services, including 
> 	location of network printers.

Again, this has come up recently in homenet, and whether mDNS
or "genuine" DNS is the way to go is a hot topic. The issue with
mDNS is that it only works on a bridged network, as I understand
it, which is an issue in many enterprise networks (our scope
for 6renum).

> 
> Otherwise the I-D seems fine to me.
> 
> Yours,
> 
> Ran
> 
> 
> 
> 
> _______________________________________________
> renum mailing list
> renum@ietf.org
> https://www.ietf.org/mailman/listinfo/renum
> 