Applications Area o Harald Alvestrand: o John Klensin: Area Summary Reported by John Klensin, MCI, and Harald Alvestrand, Uninett This is a short report on the status of the Applications Area as of the conclusion of the Dallas IETF meeting, December 1995. The Applications Area currently contains the following working groups: Access/Synchronization of the Internet Directories (asid) o Chair(s): Tim Howes o Responsible AD: Harald Alvestrand Detailed Revision/Update of Message Standards (drums) o Chair(s): Keith Moore o Responsible AD: Harald Alvestrand Electronic mail read receipts (receipt) o Chair: Urs Eppenberger o Responsible AD: Harald Alvestrand Find o Chair: Patrik Falstrom o Responsible AD: Harald Alvestrand Hypertext Markup Language (HTML) o Chair(s): Eric Sink o Responsible AD: John Klensin HyperText Transfer Protocol (http) o Chair(s): Larry Masinter Dave Raggett o Responsible AD: John Klensin This group is jointly supervised with the Transport Area. Integrated Directory Services (ids) o Chair(s): Linda Millington Sri Sataluri o Responsible AD: Harald Alvestrand This group is jointly supervised with the User Services Area. Mail Extensions (mailext) o Chair(s): C. Allan Cargille o Responsible AD: Harald Alvestrand Mail Management (madman) o Chair(s): Steve Kille o Responsible AD: John Klensin MIME Content-Type for SGML Documents (mimesgml) o Chair(s): Ed Levinson Paul Grosso o Responsible AD: John Klensin MIME-X.400 Gateway (mixer) o Chair: Urs Eppenberger o Responsible AD: Harald Alvestrand Of these, MADMAN is new to the Applications area, having been reactivated to review the existing Proposed Standard, developed in the Network Management Area, and create a new version which, depending on the changes incorporated, will be processed as a Draft Standard or cycled at Proposed. The Notifications and Acknowledgments Requirements (Notary) Working Group completed its work since the last IETF, and has been concluded. Its four resulting documents have been approved by the IESG and are in the RFC Editor's queue for publication. The Applications Area sponsored four BOF sessions, two on resource identifiers, one on voluntary labeling of content, and one on MIME- embedding of HTML files. We expect that one of these uniform resource characteristics will rapidly evolve into a working group. Charter development for the other three will continue on their mailing lists. The four BOFs are reported on below. As noted in the previous Area Report, the Applications Area continues work to evolve a workable mechanism for MIME content type registration. This mechanism, on the one hand, maintains reasonable balances between speed and ease of registration, while also maintaining quality of review and the specifications. In addition, it avoids difficulties with claims of proprietary rights in names. A new mechanism, with multiple registration trees, is being developed and should be circulated for community review and approval before the Los Angeles IETF. The biggest challenge facing the Area at this point is that previously- separable work items are becoming significantly intertwined, both within the area and between different areas. In Dallas, we had at least six working groups and BOFs working on, or making strong assumptions about, the format and use of resource identifiers. At least part of the strategy deadlock in the MIMESGML Working Group (discussed briefly below and at more length in the minutes) appears to be due to disagreements about the relative importance of Web and email-based document distribution and how format should be optimized for each. Across areas, it is becoming clear that the MIME Object Security Services (MOSS) work, developed in the Security Area, will be critical for issuing and verifying mail delivery receipts; it is less clear that MOSS, in its present form, is working on that task. Similarly, informal review of IPv6 data formats in a few applications working groups has led to the tentative conclusion that the presentation format specified for IPv6 addresses is not going to be acceptable for several Internet traditional applications. In addition, the application presentation format may have to be altered. We need to increase coordination among all of these efforts, but the coordination effort itself will have to call upon scarce resources and talent. Reports on specific working groups and BOFs Access/Synchronization of the Internet Directories (asid) ASID Summary The WHOIS++ query protocol documents have been submitted for proposed standard status. The LDAP documents will be submitted for full standard as soon as we confirm multiple independent implementations of one feature (kerberos bind). The application/directory MIME type draft will be split into two: one draft that defines the general framework, without mentioning specific attributes; another that talks about the person attributes which will be aligned with the white pages schema documents. Also, a third document will be written that includes the necessary attributes to do directory synchronization. The LDAP URL format draft will be checked against Harald's URI checklist and submitted for proposed standard. The labeled URI draft will be submitted for proposed standard. The simple caching in LDAP/X.500 will be changed to be an operational attribute definition and resubmitted. The leaf/non-leaf draft will be withdrawn since its function is subsumed by an existing operational attribute. The revised (for ipv6) string encoding of presentation address draft will be sent to the asid list for last call and submitted as proposed standard. The PGP attributes in LDAP/X.500 draft will be revised to include both suggested schemes (storage in LDAP/X.500 and storage of URL pointers to keys) and resubmitted. The SUM500 draft will be revised to reference the application/directory MIME draft and resubmitted as experimental. Renewed efforts will be made to begin progress on an LDAP draft for the next IETF that covers all three pending extensions. Detailed Revision/Update of Message Standards (drums) The DRUMS Working Group held two meetings at the Dallas IETF. The first meeting began with a review by the Chair of the Group's mission and scope, which was followed by a discussion of decision- making criteria, differences in severity of "brokenness", and possible choices for fixing such "brokenness". The discussion on scope evolved into a discussion of various documents which the Group might want to produce. The remaining time of the first session was devoted to discussion of the revised SMTP draft. The second meeting consisted of brief discussions of several open issues relating to known problems with both the SMTP and 822 protocols. Electronic mail read receipts (RECEIPT) The meeting was attended by 43 experts. An I-D was distributed on schedule last October. During the very constructive meeting we discussed issues brought up both on the mailing list and at the beginning of the meeting. All issues were disposed of with two exceptions. The Group is facing a number of substantive pitfalls including: o Avoiding e-mail loops, especially where distribution lists are involved; and, o distribution list member disclosure and other privacy issues. Since read receipts have been controversial almost from the dawn of the ARPANET, the Group has also been forced to spend time revisiting the old (but critical) issues of whether Message Disposition Notifications (MDN) are useful and whether automatically generated MDNs should be encouraged (or even permitted). A need for signed MDNs has been expressed, and has been handed off as a MOSS issue. Roger Fajman will create an updated I-D by the end of January. FIND The FIND Working Group held its first meeting The meeting was chaired by Patrik Faltstrom. Patrik reviewed the charter, which states as its goal to define one, and only one, common indexing protocol which all directory services can use when passing indexing information. Patrik presented the work he and Chris Weider have done in support of WHOIS++. Tim Howes gave a similar presentation on centipede and LDAP. The Group listed issues to work on within the common indexing protocol problem space. Hypertext Markup Language (HTML) The HTML Working Group discussed three remaining technical issues on the tables draft. After resolution of these issues, this draft will be revised and submitted for Last Call. The authors of the i18n draft were not present, so it was not discussed. The stylesheet draft was discussed only briefly, as most attendees indicated that they were not prepared for a discussion of its content. The CDA draft and its likely successor, an I-D describing a proposed tag called INSERT, designed in a non-Working Group meeting called by the W3C, were discussed, but no final conclusions were reached. A volunteer is needed to interface with HTTP Working Group on the issue of content negotiation and the richness of information conveyed by the text/html media type. After closure of the main portion of the meeting, Alex Hoppman presented a proposal for a "Template System", which turned out to be only peripherally related to HTML. HyperText Transfer Protocol (HTTP) The HTTP Working Group met twice at the Dallas IETF. A revised set of milestones for the charter will be submitted immediately. o The HTTP/1.0 draft had been rejected as a 'best current practice.' The HTTP/1.0 draft will be revised to become an 'informational RFC' which describes common current practice; and, o The HTTP/1.1 draft will be reviewed independently by separate sub- groups. The sub-groups are chartered to review the HTTP/1.1 draft for text related to their issue, and propose changes to the HTTP/1.1 draft that consist of either wording changes, or movement of major chunks of HTTP/1.1 to separate documents, as appropriate. The issues are: o Persistent connections (this contains all of the 1.1 proposals for keep- alive, and maintaining connections to avoid TCP startup costs); o cache-control and proxy behavior; o content negotiation; o authentication; o state management; o range retrievals; o extension mechanisms; and, o other new methods and header features. The volunteers for these activities, along with details of the subgroup tasks, are elaborated in the Minutes. Subgroups should conclude their work by January 1996, in time to publish their conclusions (or lack thereof) to the rest of the Working Group so that new Internet draft(s) for HTTP/1.1 will be ready in March 1996 and ready for Proposed RFC status by June 1996. Any proposed HTTP/1.1 features not in HTTP/1.0 for which there is no consensus will revert to HTTP/1.0 status in 1.1 and be considered for inclusion in HTTP/1.2. Integrated Directory Services (ids) Barbara Jennings' paper on building a directory service in the US will be progressed as an informational RFC. A new ID with a few minor changes and corrections will be submitted shortly. The IDS Group decided that the directory paper by Peter Jurg and Erik Huizer should be progressed as a BCP RFC. Harald will take up this matter with the IESG. December 31, 1996 will be the deadline for providing new product descriptions and corrections for the X.500 Catalog (RFC1632). An ID will be submitted by January 15, 1996. X.500 catalog updates can be submitted on-line via the URL http://www.internic.net/products/x500catalog. The WHOIS++ on-line catalog (http://www.bunyip.com/products) will also be updated by December 31, 1995 and an ID produced by January 15, 1996. The first draft of the CCSO (Ph) nameserver architecture paper has been submitted and volunteers found to write a preferred practices for PH directory service and support with an initial draft due on March 1, 1996. The X.500 Root Naming context draft will be discussed on the net until the end of the year and then put out as an experimental RFC. In relation to the Internet White Pages Service (IWPS), it was decided to discuss the User Requirements Document on the list, beginning with the table of contents. The schema requirements document will be further refined on the list and examples were solicited. The privacy requirements paper will be circulated for comments and the Group decided to include the quality of directory services issues in the IWPS user requirements. The IDS charter will be revised in light of discussions and published to the list. A revision was suggested to the charter to make the directory experience of the Group more visible to the rest of the IETF. The X.500 Schema Task Force will be reconstituted and the list will be approached so that volunteers may accomplish this task. Mail Extensions (mailext) Mailext, which did not meet at this IETF, currently has three outstanding documents: 1. An Informational RFC by Jacob Palme summarizing the names and usage of email and News headers. This document awaits review by a subject matter expert. Once that is complete, a Last Call will be issued. 2. A standards-track RFC by Jacob Palme defining/clarifying several new email headers is currently outstanding. An update of the ID is expected from the author; its publication will be followed by a Working Group Last Call. 3. The final outstanding document is a standards-track RFC by Ned Freed on the URL external body part. An experiment has been agreed to whereby a variation on the format that should raise many of the same issues will be used for the IETF I-D document announcements. If the experiment is successful, a Last Call will be issued on the document for Proposed Standard status. If the experiment identifies problems, the document will be returned to the Working Group. MIME Content-Type for SGML Documents (mimesgml) The four drafts (sgml-types, multipart/related, access-cid, and encap) have been submitted for publication as RFCs and are in the RFC Editor's queue. There was a discussion of the differences between the encap and exch proposals. The Area Director made it clear that to progress to standard track we will need interoperable EMAIL implementations; given the Working Group's charter, Web implementations will not count. If we wanted to change the charter to add "Web" or eliminate "email" the work could very well be move to the HTTP Working Group, thus introducing further delay. The Working Group is still deadlocked between the two major strategies. MIXER The MIXER Working Group was attended by 32 experts, four of them are implementors of the specification. No further technical changes were suggested to the core document. The body part mapping document (RFC1494bis) received a serious technical review. The three body type specific documents are acceptable. A fourth handling postscript will follow soon. Participation of EMA members was especially helpful since an EMA profile would need to be changed to allow for optimal mapping of the File Transfer Body Part. The authors will produce new documents in January 1996. A Group Last Call of two weeks will be followed by the handing over of the documents to IESG for consideration as Proposed Standard. Text/HTML in Email (mhtml-bof) This BOF, organized by Jacob Palme, discussed some of the issues and then began to identify the Group's goals and the requirements for the solution space. A set of milestones was proposed, indicating completion of the work in a year. Volunteers have been identified as co-chairs and editors; the next step is for a coherent charter to reach the area directors. Uniform Resource Characteristics (urc-bof) Attendance was about 50 people. About 75% of those attending indicated that they were there explicitly because of the URC issue, the remaining folks were in attendance because they had an empty time slot and were interested in finding out more about URCs. About 30% of those in attendance were subscribed to the mailing list. The proposed charter was reviewed and a number of good comments were received. The number of adjustments proposed, a few major, force another edit of the charter to be posted to the mailing list for review before submission to the area directors. The Group saw a nice focus on the URC issue and suggested a number of items that should be added to the charter as areas that will not be addressed. The milestones were also discussed (content only, not dates) and these will be edited and posted to the list again before being incorporated into the draft charter. Ron Daniel presented his work to date on URCs as a possible starting point for the Working Group. Keith Moore also presented his thoughts on the subject. Uniform Resource Names (urn-bof) The URN BOF started with a brief recap of the history of the URI Working Group from Larry Masinter. Keith Moore gave a presentation on the Univeristy of Tennessee, Knoxville (UTK) meeting in which most of the major implementors of URN schemes met. The proposal which arose from this group (later modified via discussion on the mailing list) was accepted as the basis for continuing work. The proposal is: URN: : (where URN: indicates the space as being a URN space; NID is the namespace identifier; and NSS is the Name Specific String). The NID governs the syntax of the NSS. The Naming Authority is embedded in the NSS. There are still some areas of contention (most notably the use of URN at the beginning), but this allows implementors to focus on what is agreed on rather than what is not. The Group decided that further informal work must be done before a charter could be written and submitted for this activity. Voluntary Access Control (VAC-BOF) Ted Hardie presented a draft document with requirements and scenarios for a voluntary access control system. Jim Miller and Paul Resnick, co- chairs of the PICS work sponsored by the Web Consortium (W3C), discussed the PICS proposals and the political climate surrounding those efforts. They also announced that the W3C would not turn the PICS work over to the IETF until the effort was completed sometime in March. The group consensus was that PICS addressed political requirements rather than engineering requirements, and several technical concerns were expressed. Without change control, the Working Group did not feel it made sense to do anything with PICS at the current time. However, there was consensus that there would be value in the IETF working in this area, especially in terms of defining requirements. Further discussion on the mailing list will occur to see whether a more focused charter that clearly states the boundaries of the work can be formulated.