Editor's note: These minutes have not been edited.
Area Summary Report by Keith Moore/University of Tennessee
and Harald Alvestrand/UNINETT
This is a short report on the status of the Applications Area as of
the conclusion of the San Jose IETF meeting in December 1997.
I. General Summary
The Applications Area currently consists of the following sixteen
working groups:
Access, Searching and Indexing of Directories (asid)
Chair: Tim Howes <howes@netscape.com>
Responsible AD: Harald Alvestrand
Application Confgiruation Access Protocol (acap)
Chair: Chris Newman <Chris.Newman@innosoft.com>
Responsible AD: Keith Moore
Calendaring and Scheduling (calsch)
Chairs: Anik Ganguily <anik@ontime.com>
Robert Moskowitz <rgm3@chrysler.com>
Responsible AD: Harald Alvestrand
Common Indexing Protocol (find)
Chairs: Patrik Faltstrom <paf@swip.net>
Roland Hedberg <Roland.Hedberg@umdac.umu.se>
Responsible AD: Harald Alvestrand
Detailed Revision/Update of Message Standards (drums)
Chair: Chris Newman <chris.newman@innosoft.com>
Responsible AD: Harald Alvestrand
Electronic Data Interchange-Internet Integration (ediint)
Chair: Rik Drummond <drummond@onramp.net>
Responsible AD: Harald Alvestrand
Extensions to FTP (ftpext)
Chair: Paul Hethmon <phethmon@utk.edu>
Responsible AD: Keith Moore
HyperText Transfer Protocol (http)
Chairs: Dave Raggett <dsr@w3.org>
Larry Masinter <masinter@parc.xerox.com>
Responsible AD: Keith Moore
This gruop is jointly sponsored with the Transport Area
Integrated Directory Services (ids)
Chairs: Linda Millington <Linda.Millington@cdc.com>
Sri Sataluri <sri@qsun.att.com>
Responsible AD: Harald Alvestrand
This group is jointly sponsored with the User Services Area.
Large Scale Multicast Applications (lsma)
Chairs: Michael Myjak <myjak@ist.ucf.edu>
Jon Crowcroft <jon.crowcroft@cs.ucl.ac.uk>
Responsible AD: Harald Alvestrand (???)
Area Advisors: Allison Mankin
Allyn Romanov
This group is jointly sponsored with the Transport Area. (???)
MIME - X.400 Gateway (mixer)
Chair: Urs Eppenberger <urs.eppenberger@switch.ch>
Responsible AD: Keith Moore
MIME Encapsulation of Aggregate HTML Documents (mhtml)
Chair: Einar Stefferud <stef@nma.com>
Responsible AD: Keith Moore
Mail and Directory Management (madman)
Chair: Steve Kille <S.Kille@isode.com>
Responsible AD: Harald Alvestrand
NNTP Extensions (nntpext)
Chair: Erik Fair <fair@apple.com>
Responsible AD: Keith Moore
Receipt Notifications for Internet Mail (receipt)
Chairs: Urs Eppenberger <urs.eppenberger@switch.ch>
Ned Freed <ned@innosoft.com>
Responsible AD: Keith Moore
Uniform Resource Names (urn)
Chairs: Leslie Daigle <leslie@bunyip.com>
John Curran <jcurran@bbn.com>
Responsible AD: Keith Moore
Of these, CALSCH and FTPEXT are new since the last IETF. In addition,
the Application Configuration Access Protocol (ACAP) BOF held at the
Montreal IETF has evolved into a working group with the same name, the
Large Scale Multicast Usage BOF held in Montreal has evolved into the
Large Scale Multicast Applications (LSMA) working group, the Network
News Transport Protocol BOF held in Montreal has evolved into the
Network News Transport Protocol Extensions (NNTPEXT) working group,
and the Uniform Resource Names (URN) BOF held in Montreal has evolved
into a working group by the same name.
No Applications Area working groups have concluded since the last
IETF. IDS, MIXER, MHTML, MADMAN, and RECEIPT are expected to conclude
soon.
The Applications Area sponsored five Birds of a Feather (BOF) sessions
at the last IETF: FAX, Internet Printing Protocol, Telnet 3270
Extended, Uniform Resource Locators, and Web Distributed Authoring and
Versioning. The reports for these follow the working group reports,
below.
II. Working Group Summaries
1. Access, Searching and Indexing of Directories (asid)
Michael Mealling presented a plan to create a registry of company
names that would return referrals to directory services for those
companies. The registry will initially be available via RWhois with
LDAP following.
New WHOIS++ drafts have been produced that address a few issues that
have surfaced since the documents went to proposed standard.
The two application/directory MIME type drafts (framework and vcard
profile) were discussed. All issues raised at the last meeting have
been addressed, but several new issues were raised at this meeting and
agreed to be discussed on the list. Note that several other groups are
depending on these documents progressing (e.g., calendar, vpim).
The LDAP API was discussed. The group wants to add this document to
its charter, and agreed to try to progress it on the standards track,
falling back to informational if heavy resistance is encountered.
LDAPv3 was discussed at length. The group identified outstanding
issues with the current drafts, and consensus was that the
specification was too complicated and represented a revolutionary
rather than evolutionary step from LDAPv2. After much discussion, the
group came to consensus that the current documents should undergo a
hard feature review and cut, based on known deficiencies with LDAPv2,
and that the extensibility mechanism should be used for other
features.
2. Application Configuration Access Protocol (acap)
The design team presented proposed solutions to all the open issues in
the current spec. Only two issues remained controversial. There was
guidance from the WG to separate the base protocol from initial
dataset schema definitions. There was also direction to search for
more prior work similar to the personal addressbook schema before
proceeding.
3. Calendaring and Scheduling (calsch)
The calendaring and Scheduling WG meeting was attended by 110 people
and covered the planned agenda. The new agreement about the "Standard
Content Type" was announced and was well received. Issues about the
standard content were discussed, several open issues remain for
discussion and resolution on the mailing list. The issues regarding
the peer-to-peer protocol were summarized and discussed, the open
isses will be discussed on the mailing list. Based on the meeting,
the WG decided to explore collaboration with the ASID WG regarding
syntax used to represent the content. The WG requested the help of
the ADs to document a standard server location method to be used by
all new distributed application protocols, this would include RFC 2052
and the work proposal made by the MMUSIC WG in their SIP (Keith do I
have the acronym right?). The editors of the peer-to-peer protocol
with help from Keith Moore, AD will explore the use of existing
protocols, incuding possible extensions/subsets currently under
discussion in other WGs.
4. Common Indexing Protocol (find)
By Patrik Faltstrom:
The FIND Working Group has decided that the current draft needs to be
split into an architecture document and a protocol document. They
discussed the open issues with CIPv3 including aggregation and the
meaning of the dataset identifier. They also discussed the draft on
the ldap schema as well as CIP and the WWW. The ldap schema will be
turned into a description of an index-object type. They updated their
charter to reflect the new definitions to the various parts of the
protocol.
Addendum by Roland Hedberg:
Actually the ldap-cip draft will be split into two documents: One
document will be a description of a index object while the other will
describe how a LDAPv2 client will interact with and view a index mesh.
5. Detailed Revision/Update of Message Standards (drums)
All three drums documents (ABNF, 822bis, smtp) were reviewed quickly.
Progress was made on a number of open issues. The abnf spec is nearing
completion and the other two documents are making progress.
6. Electronic Data Interchange-Internet Integration (ediint)
We presented two papers and the initial requirements for the
third. The first papers, focused on EDI over SMTP, were well received
with two predominate modifications: change applications/x-pkcs7-mime
to multipart/signed and move the examples from the second paper into
the first -- shortening the second paper. The feeling was that
additional requirements for the third paper on process-to-process EDI
require more detail before preceding. Also we should not be
recommending new protocols, only recommending how to use existing one.
7. Extensions to FTP (ftpext)
The i18n draft was generally thought to be ok. Minor suggestions on
reorganization. On the MLST draft, there were numerous minor
enhancements, changes suggested. Nothing especially contentious and
several point were relegated back to the mailing list for final
discussion and disposition. A couple of new topics were brought up:
ability to abort a MLST response (ie use data connection), ability to
specify what facts to return in MLST. Someone volunteered to write up
a draft on checkpoint/restart, SIZE, and MDTM. Security discussed
briefly and the group will be reviewing a couple of drafts in the
area.
8. HyperText Transfer Protocol (http)
The HTTP working group met twice. We heard about (good) performance
results from HTTP/1.1 pipelining, and discussed several issues that
have arisen as implementations of HTTP/1.1 is proceeding. Some of
these may result in either clarifications or revisions of the spec as
it moves to Draft Standard. There are a few major issues still left:
e.g., hit metering, content negotiation, and the working group will
keep going for at least until the next IETF.
There is confusion in the standards arena for security-related HTTP
protocols: SHTTP will be revised, there are proposals for new
authentication methods, and to revise the SSL Tunnelling
draft. Somehow these all need to get sorted out. There are many
groups that want to use HTTP or something like it, but with
extensions. Some of those seem like they're appropriate (web
authoring) and others are more questionable. We're awaiting a new PEP
draft, and perhaps that will help.
9. Integrated Directory Services (ids)
Progress has been made on coming to closure on a number of Drafts and
the IDS WG is within sight of completing the activities on its
Charter.
The Root Naming Context Draft is being progressed as Experimental and
the X.500 Catalogue as Informational. The DNS Aliases Draft is being
sent back to the list with a number of comments for the Group to
discuss before being progressed. A revised version of the Schema
Requirements Draft incorporating the comments received in the meeting
will be circulated by Friday 13th December and the group decided that
this Draft should be progressed as a Proposed Standard. Some
revisions will be made to the BCP on Directory Draft before
progressing as a BCP before Christmas. A revised Draft on the CCSO
Nameserver (Ph) Architecture is due this week then should be ready for
progression.
Presentations were given on the Naming Plan for an Internet Directory
Service and an approach to using Domains in LDAP. The conclusion was
that the authors would get together to produce a combined Draft and
that Hoyt Kesterson would obtain feedback from ISO about expanding the
registration authority to accommodate this scheme.
The WG discussed the open issues on the Finding stuff Draft which
included whether this work should be part of IDS or the DNS work. The
conclusion was that isuue will be taken to the SRV WG with a possible
solution being a new WG.
10. Large Scale Multicast Applications (lsma)
Michael Myjak introduced the BOF, and announced that it had now been
approved as a Working Group, Applications Area directors and Transport
Area Co-directors would be responsible for this groups output. Michael
reviewed the groups charter, and was followed by an outline review of
the two products the group intends to produce. Three presentations
followed, which identified a variety of related work in the field of
LSMA.
11. Mail and Directory Management (madman)
MADMAN did not meet in San Jose.
The MADMAN WG last met in Los Angeles in March 1996, where various
agreements were made. There are three related standards track MIB
documents being worked on by the WG. Editorial work on two of these
is complete and the final one (Mail) will be completed in the next
month. The WG expects to submit these documents as RFCs and then
disband.
12. MIME - X.400 Gateway (mixer)
The MIXER working group did not meet at the San Jose IETF. The group
is one year behind schedule.
The most recent version of the core specification has been distributed
on the mailing list, publication as I-D is expected end of January 97.
Five companion documents are ready and will be sent to the IESG together
with the core document for consideration as Proposed Standard.
13. MIME Encapsulation of Aggregate HTML Documents (mhtml)
Our Application Area Directors reported that Last Call has expired on
our MHTML Proposed Standard drafts, and that they should be approved
by the IESG in its next Teleconference. The WG then proceeded to work
on its last remaining deliverable, which is an Informational RFC Draft
to deal with implementation issuses that have been found to be
important, and which will arise as implementors proceed with their
implementations. Five vendors have indicated plans to implement, and
one of these vendors has implemented an MHTML Client. The WG Mailing
list will remain in operation in support of implmentation efforts and
to work on the informational draft. The WG in mominally deactivated,
except for staying alive in support of mplementation efforts. No
meetings are planned until the WG is reactivated to consider
advancment of the Proposed MHTML Standard to Draft status.
14. NNTP Extensions (nntpext)
The Working Group met for the first time in San Jose. The WG charter
and milestone list was reviewed by those present to focus effort.
Consensus was reached on the following items, subject to full working
group approval:
1. Stan Barber as document editor.
2. publication with suitable warnings about it being a survey of
commonly implemented extensions and not to be interpreted as a standard
of draft-barber-nntp-imp-05.txt as an informational RFC.
Some discussion of how to handle the existing extensions versus new
extensions with an extension negotiation mechanism in place. Some
issues were deferred to discussion on the mailing list; minutes to
follow later.
15. Receipt Notifications for Internet Mail (receipt)
[ note by Keith Moore ]
RECEIPT did not meet in San Jose. It is putting the final touches
on its internet-draft on Message Disposition Notifications and
expects to complete its work before the Memphis IETF.
16. Uniform Resource Names (urn)
The URN WG session included a discussion of the 4 current Internet
drafts that we have (in generic terms, URN Discovery System
Requirements & Framework, URN Syntax, NAPTR, and HTTP Conventions for
URN Resolution). A few issues were identified with each document, and
consensus was felt on most of them. Therefore, URN Syntax, NAPTR and
HTTP Conventions will go through one more quick editing phase before
being circulated on the mailing list and (barring further issues)
submitted for consideration as RFCs. The System Requirements document
is younger and will be restructured before calling for discussion on
particular items. The issue of "URN:" in the syntax was addressed:
there seemed to be general consensus in the room that it should be
part of the syntax; of the people in the room, several said they
_required_ it in order to get their work done, and no one said having
it would break their work. Volunteers were sought and obtained for
tackling the URN Namespace Requirements document, and also proposed
namespaces for grandfathering. To everyone's amazement, we
accomplished this with the minimum of discussion and were done in 1
1/2 hours. Hurrah!
III. BOF Summaries
1. FAX BOF
This was the second meeting of the Internet Fax BOF. The meeting as
"conducted like a WG" by co-chairs Dave Crocker and James Rafferty,
based on the draft WG charter. The group concentrated on the
terminology and data format type items from the draft charter at this
meeting. In review of service models, the group agreed that two
models: 1) Messaging and 2) Session-oriented, cover the broad
requirements for Internet Fax and that other potential service models
such as Real-time and Internet Aware Fax can be considered as subsets
under two main service models. The group agreed that the Messaging
model would take priority in the WG work. Three presentations were
made: 1) IFax from the WIDE consortium 2) VPIM (Voice Profile for
Internet Mail) and its use of fax, from VPIM editor Glen Parsons and
3) Concept for single Internet Fax messaging/session-oriented protocol
from Larry Masinter. The IFax and VPIM proposals both made use of
TIFF-F as a data format. The group then discussed data formats in
detail. Consensus was reached that a common file format was needed
which would meet most of the current needs, but there is also interest
in other file formats. TIFF Class F was proposed and accepted in
principle as the (first) common file format. The next step is to
prepare a standards track version of an "image-tiff; class=f" MIME
type, which draws from existing documents produced by other groups
(S.100, original TIFF-F spec from Joe Campbell). The group also
agreed that the ability of a file format to be used in a "streaming"
mode, per the Masinter proposal, is important. There was also
discussion on potential mandatory support for a text data type, but no
agreement was reached.
2. Internet Printing Protocol BOF
The meeting was attended by 80 people. Presentations were made of
current requirements, current technical draft and proposed WG
Charter. The charter called for the development of new MIME types, an
application level protocol using these MIME types, a mapping onto the
current HTTP, and a directory schema for IPP related information plus
a mapping to LDAP for that schema. The meeting was almost unanimous in
the agreement that a WG should be formed, but the IETF Area Directors
had some issues with the planned scope. Several attendants had an
issue with the use of HTTP as transport mechanism. A better motivation
for why RFC 1179 cannot be enhanced was also requested. The Area
Directors suggested that the group reconsiders the scope of the
project, before trying to get the approval as WG.
3. Telnet 3270 Extended BOF
The tn3270e working group BOF occurred and was attended by
approximately 20 members. Discussion of the charter and outstanding
issues resulted in consensus for establishing the working group and
addressing the next set of enhancements. A number of attendees
offered assistance in working on the various issues and new rfc's. A
general understanding was established regarding the process and
expectations of the IETF. The working group will most likely extend
beyond twenty four months and produce several rfc's. An updated
charter was prepared and submitted to the listsrv for review. Closure
of the charter is expected in the next 2-3 weeks.
4. Uniform Resource Locators BOF
The URL BOF met. We decided that we needed a new working group to
consider the procedures for vetting new URL schemes. The working group
might also consider the current raft of URL scheme proposals for their
compliance with those rules. While the current A-D's rules are
interesting, the goal is to create community consensus. It's hoped
that the generic syntax document would be submitted for Draft Standard
before the URL process working group really gets going.
There were volunteers for (co)chair (Rich Petke) and document editors
(Larry Masinter, Dan Zigmond).
5. Web Distributed Authoring and Versioning BOF
The BOF on WWW Distributed Authoring and Versioning was held on
Wednesday, December 11, 1996, from 09:00-11:30, with 79 people in
attendance over the duration of the meeting. Jim Whitehead presented
an overview and history of WWW Distributed Authoring and Versioning,
including information on how to become involved in the work of the DAV
group, which was followed by a detailed presentation on the current
DAV functionality requirements. There was a lively discussion on the
requirements, including such issues as lock administration, use of
public key technology, access control models, the degree of
intelligence of "move" functionality, and recommendations of related
protocols and systems to investigate (such as AFS, IMAP, and ACAP for
their models of access control). These issues will be discussed
further on our mailing list. At the end of the session, the attendees
were polled on whether they thought it would be worthwhile for the
IETF to work on WWW Distributed Authoring and Versioning.
Overwhelmingly, the attendees thought the IETF should pursue work in
this area (the vast majority were in favor, with 2-3 opposed, and a
handful of abstentions).
IV. Efforts Which May Result in Future Working Groups
1. HTTP Lite
Several working groups and BOFs have been considering use of HTTP as
the basis for other application-layer protocols. There was
considerable interest in development of a small subset of HTTP,
tentatively called "HTTP lite", for such purposes.
2. rwhois++
There is interest in developing an extended version of the rwhois
protocol, called rwhois++. This would be loosely based on the rwhois
protocol currently used in the Operations Area, but is more of a
general directory services protocol. If this effort were accepted by
the APPS ADs, it would presumbly take place within the ASID working
group.
3. Schema format registry
Several individuals expressed interest in a standard format and
registry for database schema. The schema format would allow for a
standard representation of data models (NOT for the data itself) which
could be used across different resolver and/or directory applications.
The registry would allow public access to those data models. This is
intended to aid evolution of directory services by providing a
well-known place to find existing schema.