Minutes of the Steering Committee Meeting Eynsham Hall, Oxford, 17 May 1993 C. M. Sperberg-McQueen Draft May 24, 1993 (14:45:46) 1 APOLOGIES FOR ABSENCE Present were: * Susan Armstrong-Warwick (SA) * Lou Burnard (LB) * Susan Hockey (SH) * Nancy Ide (NI) * C. M. Sperberg-McQueen (MSM) * Donald Walker (DW) * Antonio Zampolli (AZ) 2 MINUTES OF THE LAST MEETING (SC M29) The minutes (TEI SC M29) were approved as distributed and amended. 3 CURRENT STATE OF BUDGET MSM reported that after current costs and travel for the technical review and advisory board meeting, we expect to have ca. $35,000. The available funds from EEC should suffice to cover the outstanding and remaining expenses through the end of June; there will then be 50 000 ECU available on the new contract for the coming year. 4 REVIEW OF OUTSTANDING ITEMS FROM TECHNICAL REVIEW MEETING SH suggested sending out a note to those who had volunteered to work on a chapter, reminding them of who had undertaken which chapter and section, by what date. As usual at these, the meeting has generated as much work as it has performed. Document TEI P2 as it will be sent to the Advisory Board should con- tain all chapters classed at the technical review meeting as PUBLISH or PROVISIONAL. It was not clear whether to include the chapters classed as CONTINUE. We could 1. pass them over in silence 2. include them as separate documents in a separate packet 3. include a note mentioning them and saying they are not ready for inclusion in the draft, but that members interested in the topic may request the current versions It was agreed to include only the note. As regards the PROVISIONAL chapters, we should send the AB the drafts in the form we have them. For those chapters without drafts, a stub will be included with a note, as in the Technical Review version, saying that a draft is being prepared and will be included in the formal publi- cation in the fall. The committee then discussed what to send the EEC. AZ said that we must send the Guuidelines, and an evaluation of the project describing what has been and what remains to be accomplished. DW suggested sending only the chapters classed PUBLISH, rather than all the PUBLISH and PROVISIONAL chapters. After discussion, it was agreed to send to the EEC the same material as goes to the Advisory Board. NI asked whether the document to be published in the fall was to be called TEI P2 or TEI P3. It was agreed that the version in the fall must be viewed as the fulfillment of the TEI's promise to create Guide- lines. The document number TEI P2 applies to the current draft, and to the version to be mailed to the advisory board. The version in the fall will be called TEI P3. 5 FUTURE OF THE TEI AZ requested that the acknowledgements in the front matter of TEI P2 and P3 include * Istituto di Linguistica Computazionale * Bellcore * generic reference to the generosity of the employers/host institu- tions The committee then discussed at some length the future work of the TEI and the organization necessary to accomplish it.(1) The discussion started from the idea that any project proposal should discuss (not necessarily in this order): 1. the activities to be performed 2. procedures for performing them 3. the bodies or organs needed to carry out the procedures 4. the funding requirements of those organs 5. the goals of the project These topics were to be discussed in the order given. Goals were put last because it was felt easier to infer a coherent set of goals from the agreements on the other topics than vice versa. It was agreed to retain the distinction between a transitional period of about a year and the long-term project thereafter, but to focus the discussion on the transitional period rather than on the long-term future. Activities agreed upon were: * complete TEI P3 as planned [this assumes the summer of 1993 counts as part of the transitional period] * prepare written tutorials and casebooks * publish and distribute same * modify and extend (i.e. maintain) P3 * outreach: workshops, tutorials * evaluation of the Guidelines * liaison with other projects * document principles of TEI encoding These activities were grouped systematically and the list extended thus: * Technical work - principles document - complete P3 [existing resources] - maintain the Guidelines -- new work -- revision * Work with users - tutorial texts, - case books, - tutorial sessions -- basic/advanced; -- academic/industrial - "workshops" -- exchanges of information among individuals or projects -- user results and problems * Evaluation - TEI-commissioned review group or committee (internal or exter- nal) - external evaluation by funding agencies - ongoing (spontaneous or elicited) feedback from users (project/ individual) * Public relations - conference presentations - brochures - newsletter - annual TEI conference - journal - TEI user group Target decision makers, vendors, consultants. * Liaison - other standards projects / committeees - projects for data collection - network use - electronic publishers - user groups * Management and budget * Grant proposals After discussion, NI agreed that the item "document the principles of TEI encoding" should be subsumed under the rubric "maintain the Guide- lines". Expounding the item, she explained that in her view the techni- cal meeting just ended had shown that we had no well thought out, clear theoretical statement of the principles underlying the Guidelines. She felt that a number of design principles which we could not have stated five years ago are now emerging, which should be specified, including the ideas connected with synonymy and with recoverability of the source; what was needed was a clear theoretical statement, a revised list of design principles, and an effort to consolidate the experience of the last five years, generalizing the underlying principles of design and implementation. This, she felt, would be a good new work item for a future work group. AZ asked what sorts of topics were being imagined under the rubric "new work items". The following list (not exclusive) was generated: * encoding-principles documentation * chapters now classed either CONTINUE or PROVISIONAL (VE verse, DI dictionaries, LM letters and memoranda, UB user-defined bases, CB combined bases, PH physical description, ND names and dates, GF graphics and figures, FT formulas and tables, MD modifying DTDs) * computational lexica * IR * multimedia * literary analysis * speech transcription * scientific and technical documentation * newspapers * encyclopedias and reference works, annotated (enumerative) bibliog- raphy * prosody * discourse analysis, content analysis, rhetorical structure, speech acts, stylometry, thematics * non-western languages and literatures The discussion then turned to the procedures for performing these activities. In completing TEI P3, the editors and volunteers will revise the chapters to be included, the editors will print copies and disseminate to the advisory board (AB), the AB will endorse; volunteers will be solicited to read proof by mail, and in the fall the document will be published. Ongoing maintenance was discussed primarily from the point of view of handling new work items which receive their development by TEI-funded work groups; there was no explicit discussion of continuing work items or of periodic revision of chapters, and very little discussion of the interface between the TEI and externally funded work groups. One specific model was proposed for the handling of new work items, in which responsibility for the content of the guidelines would be shared in a multi-stage process among the steering committee, work groups, editors, technical review committee, and advisory board; varia- tions were discussed in which the editors had a more clearly defined, or a less clearly defined, role, and in which the advisory board had or did not have a role. The model included the following stages (with unre- solved questions as noted): * a new work item is proposed, approved, and assigned to an existing or new work group (left undecided: who has standing to propose a work item, who approves them, who assigns work items to work groups, and who commissions work groups) * the work group drafts a chapter for the Guidelines; when it is fin- ished, they transmit it to the next step (varying opinions over whether work groups should transmit drafts to the editors or direct to the technical review committee) * the editors review and revise the draft; when both the editors and the work group are happy with the draft (i.e. willing to sign off on it), it is published as a draft chapter, for public comment, and transmitted to the technical review committee for action (some expressed the opinion that this step should be omitted) * the technical review committee reviews the draft and takes one of the following actions: - returns the draft to the work group for further work - approves it for publication, possibly with suggestions for edi- torial changes - amends it and approves it as amended (left undecided: whether the work group must approve any amendments to the chapter) * any editorial changes suggested by the technical review committee or resulting from technical changes are made and the resulting chapter is transmitted to the advisory board for final approval * the advisory board reviews the draft and approves its inclusion in the Guidelines (left undecided was the proposition that this stage should be omitted and that approval by the technical review commit- tee should suffice for inclusion in the Guidelines, or that the steering committee, rather than the advisory board, should make the final approval) Discussion of the model revealed varying opinions as to the proper role of the editors, especially on whether they should review draft chapters before they are submitted to the technical review committee, and as to the appropriateness of a final review by an advisory board in which the interests of various consituencies would be represented. There were also varying opinions on how the intellectual and physical integration of a draft chapter with rest of the Guidelines ought best to be per- formed. AZ asked how the model described differed from the TEI's current practice; the main differences, it was suggested, were the formal con- stitution of a technical review committee, the clearer definition of its role, and the clarification of the relationship between the editors and the work groups. No technical review committee now exists, the first technical review meeting having been a joint meeting of working commit- tees TR and AI, with invited guests, and the latter two being ad hoc sessions convened by the steering committee without a fixed membership. And the current arrangements, because they do not define a clear rela- tionship between work groups and editors, effectively give the editors carte blanche to change work group drafts. The editors have attempted to avoid wholesale technical changes in work group drafts, for obvious reasons, but wholesale editorial changes have been unavoidable given the state of the materials produced by some work groups, and in the past, under pressure of deadlines, the editors have not consistently requested approval from the work groups for changes made in the work group drafts. AZ observed that in his opinion, the editors have done far more in the past than indicated by the model for the future or by the answers to his query: the editors have not merely revised work group drafts but have drafted many of the existing chapters from scratch, and have thus exercised far more intellectual direction over the project than is visi- ble in the proposed model. LB remarked that just as in the present, the editors would have to continue to ensure that the process worked, and help out wherever it failed to work: their function is in effect to keep the machinery oiled and functioning, whatever that might entail. A discussion ensued on the proposition that the editors should revise chapters only after they have been acted on by the technical review com- mittee; some proposals appeared to involve inserting a separate editori- al approval phase after action by the technical review committee. It was agreed that full details of the technical review process need not be agreed upon at this time, since the immediate purpose of the exercise was fulfilled by observing that the technical work would almost certainly involve the following organs of the TEI: * the steering committee (as now constituted or as reorganized) * a technical review committee (to be constituted) * editors * work groups * an advisory board (as now constituted or, as seemed to be the con- sensus, as slightly revamped) At this point the committee adjourned for lunch. After lunch, the committee proceeded to consider procedures for the remaining activities: work with users, evaluation, liaison with other bodies, management and budget, and proposal writing. It was agreed that work with users would involve a cadre of technically skilled people willing to serve as * TEI uncles to ongoing projects * teachers of workshops and tutorials * drafters of tutorials * drafters of chapters for the case book In addition, an administrator of some sort would be needed for handling arrangements for tutorials, etc. -- it was not clear whether this should be a simple low-level bookkeeper, or a higher-level chief financial officer for the project. A low-level assistant to handle public rela- tions and edit a newsletter was also felt desirable. Written tutorials might be drafted by work groups, editors, volun- teers, members of the steering committee, or a hired technical writer. Audiences to be addressed (eventually) include * generic * historians * terminologists * literary scholars * creators of historical (and philological?) editions * creators (and users?) of language corpora * creators of electronic dictionaries (creators of modern commercial dictionaries, of ethnographic dictionaries, and historical diction- aries, and encoders of older dictionaries) Casebooks should include chapters written by affiliated projects or by their uncles. Courses or live tutorials will require both administrative staff to handle the arrangements and teachers (from the pool of technical experts already mentioned). They could usefully be integrated with the annual conferences of participating organizations, and might be free-standing events organized by other organizations at their own initiative. Evaluation of the guidelines will involve: * eliciting spontaneous feedback from users (TEI-L should have an activist list editor to make it a more useful forum for user feed- back; we need ways of tracking users of various types, and we need to organize forums for the presentation of user experiences with the TEI) * systematic review of existing chapters, to be commissioned external- ly or to be performed by the technical review committee * channeling of other results from the editors to the technical review committee In the field of public relations, it was agreed that presentations at conferences should be expected of the editors, members of the steering committee, and members of the technical review committee; a brochure for display at conferences might be drafted by a public relations assistant (LB pointed out that a brochure-length text does already exist and is regularly distributed to inquirers); a newsletter would require a news- letter editor; organizing a TEI conference would require administrative assistance. Liaison with other bodies was, it was felt, a task for the steering committee and editors, which could not be delegated to others. The edi- tors stressed the importance, in their view, of having a successor to the program of work with the old affiliated projects. Budget management would, it was agreed, best be handled by hiring someone specifically to keep the TEI's books. The priorities for the next phase of work, it was thought, should be these: * complete P3 (target date: 1 October) * tutorial texts, brochure, newsletter * simple casebook * tutorial sessions * planning * keep new work items going (SA felt these should be low priority) * start liason with external groups * work with users (answering questions) * affiliated projects * specify evaluation procedures Costs for this phase were estimated thus: * editorial salaries: $84,000 (including assistance) plus 40,000 ECU * steering committee meetings: $24,000 * technical review meeting: $20,000 * printing P3 $5,000 (total print run will cost more, but the rest should be recouped from sales) * total: $133,000 plus 40,000 ECU AZ proposed that someone needs to draft a program of work. SA point- ed out that we also need someone to put the various documents into a proposal or pre-proposal description. ACTION: SH to draft tentative program of work and circulate to SC for com- ment Due: asap Document number: SC W7 MSM proposed to locate as many as feasible of the administrative functions at CETH, in the hopes of simplifying the administrative organ- ization of the project. There was discussion, but no decision. AZ proposed that we authorize the American contingent to seek funding from NSF, provided it is agreed that the proposal will not go to Europe- an authorities. No objection was heard. At this point, AZ departed. It was agreed that the details of the proposal should be left to NI, SH, and DW. 6 PLANS FOR ACH/ALLC 93 We have two hours on the agenda at ACH/ALLC 93. Since most people mostly want to see what it looks like, we can do worse than show them examples. LB suggested that we should think very seriously about how to get more intellectual content into the session. DW said the focus should be on providing an update, reporting on the Eynsham Hall meeting, describe the publication schedule, and give examples. It was agreed to ask Allen Renear to give a talk about the TEI during the session, focusing on the Eynsham Hall meeting, issues that came up there with particular relevance to the humanities (e.g. certainty, responsibility, and segmentation-alignment and its relevance to hyper- text and humanities), and the implications of the TEI for humanities research projects. The session will open with an overview by NI on the current status of the project and the publication schedule. Allen Renear will then speak. LB and MSM will then give some examples and an overview of the current table of contents. The meeting will conclude with open questions. NI volunteered to work on the text of TEI J8 to produce a brochure for distribution at ACH/ALLC and ACL. LB volunteered to see that she got the text of J8. SH volunteered that CETH would do the DTP work, prettify it, and produce copies of the brochure. NI suggested that it should also be sent out for distribution at ACM/SIGIR. 7 PLANS FOR ADVISORY BOARD MEETING MSM reported that the AB meeting would be at the Carleton Hotel in Oak Park, Illinois. People would be arriving Saturday or Sunday, and Sunday afternoon we have promised them an informal question-and-answer session. SA suggested that a social gathering Sunday night would be very useful, since the members of the AB are largely higher-level repre- sentatives of their organizations, who will be involved in public rela- tions and dissemination of the Guidelines, rather technical people who have been deeply involved in their development. It was suggested that the Monday should be devoted to technical issues: a review of the design principles approved by the advisory board at its first meeting, showing how they have played out in prac- tice, and a summary of the contents (focusing on how the members of the AB can sell various chapters to their constituencies). We will hope on Monday to get through the formal vote for endorsement by the participat- ing organizations. On Tuesday, we will focus on the future of the TEI, asking questions like what they can do to help with the dissemination of P3, how the TEI central organization can help, whether they will help organize a work- shop at their annual conference, whether they (as AB representative) or the executives of their organization will write a letter in support of our next funding proposals. 8 DATES OF FUTURE MEETINGS No meeting can take place before or after the AB meeting. No meeting is scheduled for Georgetown, but in case of crisis an emergency meeting can be convened some evening. ------------------------- (1) The following account of this discussion is based on notes taken by LB, to whom thanks. -MSM Draft May 24, 1993 (14:45:46)