Minutes <title>Of the Steering Committee Meeting, <title>Oxford, 6-7 July 1990 <author>C. M. Sperberg-McQueen <docnum>TEI &docfile. <date>&docdate. <attend> Present: Lou Burnard (LB), Susan Hockey (SH) in part, Nancy Ide (NI), C. M. Sperberg-McQueen (MSM), Donald Walker (DW), Antonio Zampolli (AZ). Absent: Robert Amsler (RA), Susan Hockey (SH) for the most part. </attend> </titlep> <!> <toc> </frontm> <!> <body> <h1>1. Agenda The agenda proposed by the chair was adopted. This included: <ol compact=2> <li>Report (i.e. TEI P1 versions 0 and 1) <ol compact=2> <li>review contents of June distribution <li>revisions for July release (examples, DTDs, character sets) <li>distribution list for July <li>executive summary <li>comments log for responses </ol> <li>Publicity <ol compact=2> <li>news releases <li>conference papers and talks <li>journal articles? <li>tutorials and briefings <li>working paper releases </ol> <li>Usability Issues <ol compact=2> <li>user's manuals, tutorial guides, implementors' notes <li>example applications from raw data through softwar tools to SGMLized files with illustrations of processing capabilities <li>get software developers involved </ol> <li>Tutorials <ol compact=2> <li>feedback from Siegen and Pittsburgh <li>fall tutorials in North America and Europe <li>future tutorials? </ol> <li>Scope of Final Report <ol compact=2> <li>conservative approach: make first draft coherent <li>expansive approach: extend to new areas </ol> <li>Committees <ol compact=2> <li>restructure in relatino to scope of final report <li>decentralize TR and AI <li>committee heads (SJ, DTB) <li>subcommittee heads (topics and names) <li>schedules for deliverables for committees and subcommittees <li>encourage more communication <li>how establish productive working groups </ol> <li>Administrative Support <ol compact=2> <li>full-time commitment from MSM <li>more time from LB <li>additional staff members needed? </ol> <li>Software Support <ol compact=2> <li>relate report contents to existing software <li>acquire and experiment with existing software <li>motive development of new software <li>incorporate software recommendations and specifications into final report <li>establish formal relations with software suppliers </ol> <li>Affiliated Projects <ol compact=2> <li>establish effective communication channels <li>facilitate use of software <li>incorporate results into interim and final reports <li>coopt results of their work as TEI working papers </ol> <li>Financial Issues <ol compact=2> <li>Pisa payments <li>redirection of funds toward subcommittees, clerical support, affiliated projects <li>industry solicitation </ol> <li>Schedule <ol compact=2> <li>Sept. 1990 Committee and subcommittee plans <li>Sept./Oct. 1990 workshops <li>Nov. 1990 comments due <li>Jan. 1991 response to comments due <li>Mar. 1991 feedback and second round of comments due <li>Jun./Jul. 1991 version 2 of report <li>Dec. 1991 final comments due <li>Mar. 1992 draft of final report <li>Jun. 1992 distribution/publication of final report </ol> <li>Publication and Distribution of Final Report <ol compact=2> <li>commercial publisher (Kluwer, Oxford University Press?) <li>electronic distribution </ol> <li>Rockefeller Grant and Conference <li>Future Steering Committee Meetings <ol compact=2> <li>Sept. 1990 with committee and subcommittee heads? <li>Dec. 1990 <li>Mar. 1991 (ACH/ALLC Tempe) <li>Jun. 1991 (ACL Berkeley) </ol> </ol> Some of these topics were not addressed. <!> <h1>2. Versions 0 and 1 of Report <p> Various corrections to version 0 were passed to the editors and noted. The editors suggested that the top priorities in preparing version 1 for release July 15 must be preparation of the DTDs and correction of errors in the prose, followed by addition of more examples and refinements to the DTDs, followed by the alphabetical list of tags (reference section), which might be provided with only minimal information, or with full information as specified in working paper EDW5. They hoped to get through at least the first two, but were not hopeful as regards the alphabetical list of tags. This order of priorities was accepted. <p> DW suggested that at some point chapters 2 and 8 (SGML tutorial and notes on extensions) should be more tightly coupled, if only with cross references. He also reported Terry Langendoen's concern that the <tag>S</tag> tag for sentences should have analogs for words and characters; MSM explained that <tag>S</tag> stands for <q>segment</q> not <q>sentence</q> and agreed to try to make that more apparent throughout. <p> The preface needs to contain the editors' addresses, in case the user feedback form is missing from a copy. <p> If an interim version 1.1 is distributed before summer 1991, it was agreed that recipients of version 1 may receive just the sections which have changed, and not necessarily the entire text. <p> The executive summary is yet to be expanded and revised. <action> <who>editors <act>to revise draft executive summary and distribute <duedate>asap </action> <p> Version 1 should go (in its entirety) to working committee members, affiliated projects, the advisory board, those who requested copies in Siegen, Pittsburgh, or Chicago, those who requested information some time ago, and all of the original volunteers for the working committees. DW will prepare cover letters for these groups. <action> <who>DW <act>to draft cover letters for recipients of version 0 (participants in the project), volunteers not chosen for the working groups, those who have requested information or copies, and those who are getting the document unannounced and unrequested <duedate>asap (15 July) </action> <!> <h1>3. Publicity <p> NI has sent out a press release describing the funding of cycle 2. LB volunteered to draft a release describing the public availability of version 1. <action> <who>LB <act>to draft <q>TEI version 1 released</q> release <duedate>15 July 1990 </action> <p> The TEI leaflet needs to be brought up to date. DW volunteered to undertake this together with SH. <action> <who>DW, SH <act>to redraft TEI leaflet / brochure <duedate>unspecified </action> <p> All agreed that journal articles describing the TEI were needed to keep the relevant communities informed. An article describing the TEI design goals in the light of actual developments seemed appropriate at this point in the project. Some working papers should also be published in journals; the editors and steering committee should encourage their authors to rework them for publication. Further discussion is needed to choose specific papers. <p> The editors are to contact committee heads and committees with a list of documents created so far, inquiring which if any should remain internal to the project or retired entirely, and which should be available to the public. <action> <who>editors <act>to contact committees and heads about public release of working papers and other documents <duedate>asap </action> It was decided not to create a separate series of technical reports or preprints; publicly available working papers will be distributed with their internal TEI numbers and without any series designation. <!> <h1>4. Usability Issues and Workshops <p> It was agreed that tutorials on the current draft (version 1) should be held in September or October, one in North America and one in Europe. The sites should have clusters of Macintosh and PC computers; Oxford and UIC are obvious candidates. The committee discussed the possibility of restricting the workshops or of making them public; the consensus was that the workshops are needed particularly to ensure that affiliated projects and members of the work groups for the second cycle have a chance to work intensively with the proposals, and that therefore this internal public should be the primary target audience. Those outside the project may come if there is room. <p> If thirty people attend each workshop, with TEI funds, the travel cost could rise as high as $24,000 per workshop; the likelihood is of a much smaller attendance and cost. <!> <h1>5. Final Report <p> The committee discussed what should happen in the second cycle; should the work be restricted largely to completing and fixing problems in the current draft or should an expansive approach be taken and a large program of extensions be formulated? MSM expressed the view that for the areas covered, the current draft is usable as is, and that a technical person knowledgeable about SGML might adapt the scheme for a simple project in a matter of days or weeks. Adaptation for text-enrichment projects (e.g. tagged corpora) would take somewhat longer, as no guidance is now provided for word or phrase analysis. <p> The three goals of the second cycle were summarized thus, in the consensus of the committee: <ol> <li>complete the work left unfinished during the first cycle (mostly marked explicitly as such in version 1) and make the draft more coherent <li>correct errors in the work done during the first cycle <li>extend the work to new areas </ol> The affiliated projects are important to all three goals and crucial to the second, and so relations with them must be carefully cultivated. <!> <h1>6. Committees <p> The committee agreed that commmitee work should be decentralized during the second cycle. The ML and TD committees should continue to exist, as they have clear and coherent programs to pursue, but the TR and AI committees must effectively divide into smaller work groups to cover the remaining work. <!> <h2>6.1. Metalanguage <p> The ML committee has outlined a number of areas for further work: translation into and out of the TEI interchange scheme, an SGML bibliography, issues of data capture, specification of packing and unpacking software analogous to SDIF, and further investigation of TEI validation issues. <p> LB suggested Franc,,ois Chahuneau of Advanced Information Systems in Paris be added to the committee; AZ also has a nominee, whose name he will transmit to the committee after he returns to Pisa. <action> <who>DW <act>to propose Chahuneau and AZ's nominee to DTB for committee service <duedate>asap </action> <p> Relations with software houses must be set up; this will not necessarily mean adding people to the committee. <!> <h2>6.2. Text Documentation <p> This committee must refine and correct its section of the guidelines. No changes in membership were proposed. It is expected that the committee will meet only once during this cycle. <!> <h2>6.3. Text Representation <p> Areas flagged in the current draft as needing further work include: <ul> <li>character sets, including Asian scripts <li>physical description of source documents <li>text criticism and historical-critical editions <li>hypertext links and cross references <li>structural descriptions of further text types <li>formulas, tables, figures; more generally multimedia documents <li>canonical reference systems <li>language corpora </ul> <p> Membership changes were discussed. Libermann and Thaller will leave the committee; Cencioni, Cover, DeRose, Hockey, Huitfeldt, Mylonas, and Renear should stay. Status of Chesnutt, Ott, and Johansson is unclear; the latter two appear tempted to step down now. Some possible nominees for the character set subgroup were discussed, including a French representative from Grenoble, Harry Gaylord of Groningen, and Ed Hart from the Johns Hopkins Applied Physics Laboratory. Makato Nagao has also expressed interest in having Japanese participation; this has technical as well as political ramifications, as some approaches to character-set problems are supported along national lines, and it may be necessary to include Chinese participants as well or risk premature choice among the various suggestions now being discussed. <p> To replace Thaller and ensure treatment of historical text types, LB suggested, a nascent committee of the Association for History and Computing on the requirements of historical computing should perhaps be consulted. This group will meet in Montpelier in September, at the AHC conference. <!> <h2>6.4. Organization of Committees in General <p> The committee discussed for some time the proper organization of the coming cycle. Consensus was reached that most of the work should be done in small work groups (corresponding to the subcommittees of the first cycle). It was not clear whether the committees of the first cycle should continue to have any prominent role; TD and ML will continue to exist, but there is some possibility that TR and AI will exist only as coordinating bodies to which work groups will report. During the discussion, the consensus appeared to be that they will have no central role, but no formal decision was made to that effect. <p> The work groups should have few members (two or three should be the usual size), short lives (as little as three or four months), and focused charges and work plans. Many should have only one meeting; others might have several, if they could make a case for it and earlier meetings had been productive. <p> It was agreed that the steering committee must decide on a mechanism for deciding on the topics of work to be assigned to the work groups and monitoring their progress. Two proposals were made: that the steering committee should appoint an ad hoc committee (or more than one) to specify the areas needing work, and that the steering committee itself should decide on the areas needing work, possibly forming work groups to prepare advisory papers on the issue. No decision was made. <p> AZ asked the editors to prepare a paper listing the areas they thought needed work groups. They agreed to do this. <action> <who>editors <act>to draft list of work groups needed at present <duedate>asap </action>     <!> <h1>7. Affiliated Projects <p> NI urged that a clear understanding be achieved with the affiliated projects as to what exactly they and we should be providing each other, specifying deadlines for the completion of tagging of a certain amount of text and for reports. Reports should summarize problems encountered using the TEI scheme to express required information and methods used for data capture or tag generation. If possible, there should also be a money payment to the projects. <p> It was agreed that each affiliated project and the TEI should name a contact person to which the other could always send documents or questions; the possibility that there should be a single TEI contact person for all affiliated projects was discussed but not accepted. It was also agreed that a written plan with specifications of data and reports to be submitted, with deadlines, should be negotiated with each project's representative. <p> AZ described some sample types of interaction with the projects: initial technical meeting, consulting and advice during the implementation phase, second technical meeting after the implementation phase for debriefing, reception and routing of output from the project, and reaction to the results. All agreed that each project should have a single TEI contact person to answer questions and give advice (the TEI <term>uncle</term>). MSM urged that the projects not be restricted to contact with the uncle; it should be part of the responsibility of any work group head or member to assist any affiliated project as far as possible, and the projects should have the right to call anyone in the project directly, not just through their uncle. The uncle might suggest, also, that the technical meetings be handled by some other TEI participant rather than the uncle. This would be particularly true where the uncle was a participant in both the TEI and the affiliated project (true of several projects). AZ suggested that the projects might be grouped according to the type of work they are doing, and each group be assigned a common uncle. This will affect and be affected by the work-group structure. <p> NI was asked to draft a sample written plan specifying details of cooperation, to serve as the basis for talks with each project. This written plan should provide further detail within the framework established by the basic agreement with the affiliated projects, document TEI SCP1. <action> <who>NI <act>to draft detailed proposal for cooperation between TEI and affiliated projects <docref>TEI SCA7 <duedate>10 July [received -Ed.] </action> <p> New affiliated projects were formally agreed to: the Nietzsche project of Malcolm Brown at Stanford University, and the Network of European Corpora (contact person: AZ). <!> <h1>8. Final Report to EEC <p> AZ explained that the final report to the EEC should have three parts: an introduction describing the project and the results, the guidelines themselves (version 1), and the financial summary. A substantial amount of money remains unclaimed in the Pisa account; if it is not claimed soon, it will be lost. Methods of dealing with this situation were discussed at some length. [NI left the meeting during this discussion.] <!> <h1>9. Administrative Support <p> MSM described the inconclusive state of his negotiations with UIC management to free more of his time; various scenarios were discussed and their costs were estimated. The budget submitted to NEH with the proposal of September 1989 calls for 50% time for MSM, a continuation of the current level of undergraduate clerical help, and a graduate assistant at 50% time. If 100% of MSM's time is purchased instead, it will raise the relevant salary costs about $50,000 including overhead for the two years. If the line for a graduate assistant is changed into a 50%-time administrative assistant or secretary, it will raise the cost by perhaps as much as $25,000. If two graduate assistants are hired instead of one, the cost will rise about $25,000. The preferred solution is to release the rest of MSM's time; either of the other two solutions may be chosen otherwise, depending upon whom it seems possible to hire. <p> MSM was urged to list in some detail the responsibilities of the editors and those which may and may not be delegated to others. <action> <who>MSM <act>to draft summary of editorial responsibilities for cycle 2 <duedate>unspecified </action> <!> <h1>10. Schedule for Cycle II and Official Responses to Comments <p> The schedule included in the agenda was approved as shown above. <p> The committee agreed that public comments and queries should receive official TEI responses. These should be drafted by the competent working committee or by an editor; a working committee response is official if approved as such by either editor. If either editor sanctions a response as official without first consulting the other, he is to notify the other within one day. The editors are expected to inform the steering committee regularly of such officially sanctioned responses and may consult with them as appropriate. <p> All responses are to be logged in the comment-log database. <!> <h1>11. Software <p> The editors summarized their contacts with software vendors. SEMA group (through Yard Software) has generously made Mark-It available to anyone associated with the TEI, and both editors have copies. LB also has SEMA's Write-It. Software Exoterica has provided evaluation copies of CheckMark (a validating editor for the Macintosh) and XTRAN (a translator for other platforms; editors have the MS-DOS version). Datalogics have promised copies of their Edit Station software but have since fallen silent; MSM should nudge them. SoftQuad has resisted donations on the grounds that they cannot afford them, but their UK distributor seems to be taking a more liberal approach. LB also has a copy of Daphne from the Deutsches Forschungsnetz, but this has a hard-wired DTD. The Turing Institute in Glasgow has promised source code for TOLK, their SGML parser; this supports CONCUR <q>but not in the the way you'd expect.</q> TOLK has a strong Hypercard and database interface. LB and MSM have both been pursuing Michael Cowlishaw's LEXX editor originally developed for OUP and the OED project, in DOS and VM versions, respectively, so far without luck. <!> <h1>12. Unfinished Business <p> The meeting adjourned sine die without acting on the following items of the agenda: <ul> <li>future structure of TR and AI committees and subgroups <li>future role of TR and AI committee heads <li>topics of working groups and names of heads <li>clarification of EEC payments through Pisa <li>publication and distribution of final report <li>Rockefeller grant and conference </ul> <!> </body> <!> </gdoc>

.sr docfile = &sysfnam. ;.sr docversion = 'Draft';.im teigmlp1 .* Document proper begins. Minutes <title>Of the Steering Committee Meeting, <title>Oxford, 6-7 July 1990 <author>C. M. Sperberg-McQueen <docnum>TEI &docfile. <date>&docdate. <attend> Present: Lou Burnard (LB), Susan Hockey (SH) in part, Nancy Ide (NI), C. M. Sperberg-McQueen (MSM), Donald Walker (DW), Antonio Zampolli (AZ). Absent: Robert Amsler (RA), Susan Hockey (SH) for the most part. </attend> </titlep> <!> <toc> </frontm> <!> <body> <h1>1. Agenda The agenda proposed by the chair was adopted. This included: <ol compact=2> <li>Report (i.e. TEI P1 versions 0 and 1) <ol compact=2> <li>review contents of June distribution <li>revisions for July release (examples, DTDs, character sets) <li>distribution list for July <li>executive summary <li>comments log for responses </ol> <li>Publicity <ol compact=2> <li>news releases <li>conference papers and talks <li>journal articles? <li>tutorials and briefings <li>working paper releases </ol> <li>Usability Issues <ol compact=2> <li>user's manuals, tutorial guides, implementors' notes <li>example applications from raw data through softwar tools to SGMLized files with illustrations of processing capabilities <li>get software developers involved </ol> <li>Tutorials <ol compact=2> <li>feedback from Siegen and Pittsburgh <li>fall tutorials in North America and Europe <li>future tutorials? </ol> <li>Scope of Final Report <ol compact=2> <li>conservative approach: make first draft coherent <li>expansive approach: extend to new areas </ol> <li>Committees <ol compact=2> <li>restructure in relatino to scope of final report <li>decentralize TR and AI <li>committee heads (SJ, DTB) <li>subcommittee heads (topics and names) <li>schedules for deliverables for committees and subcommittees <li>encourage more communication <li>how establish productive working groups </ol> <li>Administrative Support <ol compact=2> <li>full-time commitment from MSM <li>more time from LB <li>additional staff members needed? </ol> <li>Software Support <ol compact=2> <li>relate report contents to existing software <li>acquire and experiment with existing software <li>motive development of new software <li>incorporate software recommendations and specifications into final report <li>establish formal relations with software suppliers </ol> <li>Affiliated Projects <ol compact=2> <li>establish effective communication channels <li>facilitate use of software <li>incorporate results into interim and final reports <li>coopt results of their work as TEI working papers </ol> <li>Financial Issues <ol compact=2> <li>Pisa payments <li>redirection of funds toward subcommittees, clerical support, affiliated projects <li>industry solicitation </ol> <li>Schedule <ol compact=2> <li>Sept. 1990 Committee and subcommittee plans <li>Sept./Oct. 1990 workshops <li>Nov. 1990 comments due <li>Jan. 1991 response to comments due <li>Mar. 1991 feedback and second round of comments due <li>Jun./Jul. 1991 version 2 of report <li>Dec. 1991 final comments due <li>Mar. 1992 draft of final report <li>Jun. 1992 distribution/publication of final report </ol> <li>Publication and Distribution of Final Report <ol compact=2> <li>commercial publisher (Kluwer, Oxford University Press?) <li>electronic distribution </ol> <li>Rockefeller Grant and Conference <li>Future Steering Committee Meetings <ol compact=2> <li>Sept. 1990 with committee and subcommittee heads? <li>Dec. 1990 <li>Mar. 1991 (ACH/ALLC Tempe) <li>Jun. 1991 (ACL Berkeley) </ol> </ol> Some of these topics were not addressed. <!> <h1>2. Versions 0 and 1 of Report <p> Various corrections to version 0 were passed to the editors and noted. The editors suggested that the top priorities in preparing version 1 for release July 15 must be preparation of the DTDs and correction of errors in the prose, followed by addition of more examples and refinements to the DTDs, followed by the alphabetical list of tags (reference section), which might be provided with only minimal information, or with full information as specified in working paper EDW5. They hoped to get through at least the first two, but were not hopeful as regards the alphabetical list of tags. This order of priorities was accepted. <p> DW suggested that at some point chapters 2 and 8 (SGML tutorial and notes on extensions) should be more tightly coupled, if only with cross references. He also reported Terry Langendoen's concern that the <tag>S</tag> tag for sentences should have analogs for words and characters; MSM explained that <tag>S</tag> stands for <q>segment</q> not <q>sentence</q> and agreed to try to make that more apparent throughout. <p> The preface needs to contain the editors' addresses, in case the user feedback form is missing from a copy. <p> If an interim version 1.1 is distributed before summer 1991, it was agreed that recipients of version 1 may receive just the sections which have changed, and not necessarily the entire text. <p> The executive summary is yet to be expanded and revised. <action> <who>editors <act>to revise draft executive summary and distribute <duedate>asap </action> <p> Version 1 should go (in its entirety) to working committee members, affiliated projects, the advisory board, those who requested copies in Siegen, Pittsburgh, or Chicago, those who requested information some time ago, and all of the original volunteers for the working committees. DW will prepare cover letters for these groups. <action> <who>DW <act>to draft cover letters for recipients of version 0 (participants in the project), volunteers not chosen for the working groups, those who have requested information or copies, and those who are getting the document unannounced and unrequested <duedate>asap (15 July) </action> <!> <h1>3. Publicity <p> NI has sent out a press release describing the funding of cycle 2. LB volunteered to draft a release describing the public availability of version 1. <action> <who>LB <act>to draft <q>TEI version 1 released</q> release <duedate>15 July 1990 </action> <p> The TEI leaflet needs to be brought up to date. DW volunteered to undertake this together with SH. <action> <who>DW, SH <act>to redraft TEI leaflet / brochure <duedate>unspecified </action> <p> All agreed that journal articles describing the TEI were needed to keep the relevant communities informed. An article describing the TEI design goals in the light of actual developments seemed appropriate at this point in the project. Some working papers should also be published in journals; the editors and steering committee should encourage their authors to rework them for publication. Further discussion is needed to choose specific papers. <p> The editors are to contact committee heads and committees with a list of documents created so far, inquiring which if any should remain internal to the project or retired entirely, and which should be available to the public. <action> <who>editors <act>to contact committees and heads about public release of working papers and other documents <duedate>asap </action> It was decided not to create a separate series of technical reports or preprints; publicly available working papers will be distributed with their internal TEI numbers and without any series designation. <!> <h1>4. Usability Issues and Workshops <p> It was agreed that tutorials on the current draft (version 1) should be held in September or October, one in North America and one in Europe. The sites should have clusters of Macintosh and PC computers; Oxford and UIC are obvious candidates. The committee discussed the possibility of restricting the workshops or of making them public; the consensus was that the workshops are needed particularly to ensure that affiliated projects and members of the work groups for the second cycle have a chance to work intensively with the proposals, and that therefore this internal public should be the primary target audience. Those outside the project may come if there is room. <p> If thirty people attend each workshop, with TEI funds, the travel cost could rise as high as $24,000 per workshop; the likelihood is of a much smaller attendance and cost. <!> <h1>5. Final Report <p> The committee discussed what should happen in the second cycle; should the work be restricted largely to completing and fixing problems in the current draft or should an expansive approach be taken and a large program of extensions be formulated? MSM expressed the view that for the areas covered, the current draft is usable as is, and that a technical person knowledgeable about SGML might adapt the scheme for a simple project in a matter of days or weeks. Adaptation for text-enrichment projects (e.g. tagged corpora) would take somewhat longer, as no guidance is now provided for word or phrase analysis. <p> The three goals of the second cycle were summarized thus, in the consensus of the committee: <ol> <li>complete the work left unfinished during the first cycle (mostly marked explicitly as such in version 1) and make the draft more coherent <li>correct errors in the work done during the first cycle <li>extend the work to new areas </ol> The affiliated projects are important to all three goals and crucial to the second, and so relations with them must be carefully cultivated. <!> <h1>6. Committees <p> The committee agreed that commmitee work should be decentralized during the second cycle. The ML and TD committees should continue to exist, as they have clear and coherent programs to pursue, but the TR and AI committees must effectively divide into smaller work groups to cover the remaining work. <!> <h2>6.1. Metalanguage <p> The ML committee has outlined a number of areas for further work: translation into and out of the TEI interchange scheme, an SGML bibliography, issues of data capture, specification of packing and unpacking software analogous to SDIF, and further investigation of TEI validation issues. <p> LB suggested Franc,,ois Chahuneau of Advanced Information Systems in Paris be added to the committee; AZ also has a nominee, whose name he will transmit to the committee after he returns to Pisa. <action> <who>DW <act>to propose Chahuneau and AZ's nominee to DTB for committee service <duedate>asap </action> <p> Relations with software houses must be set up; this will not necessarily mean adding people to the committee. <!> <h2>6.2. Text Documentation <p> This committee must refine and correct its section of the guidelines. No changes in membership were proposed. It is expected that the committee will meet only once during this cycle. <!> <h2>6.3. Text Representation <p> Areas flagged in the current draft as needing further work include: <ul> <li>character sets, including Asian scripts <li>physical description of source documents <li>text criticism and historical-critical editions <li>hypertext links and cross references <li>structural descriptions of further text types <li>formulas, tables, figures; more generally multimedia documents <li>canonical reference systems <li>language corpora </ul> <p> Membership changes were discussed. Libermann and Thaller will leave the committee; Cencioni, Cover, DeRose, Hockey, Huitfeldt, Mylonas, and Renear should stay. Status of Chesnutt, Ott, and Johansson is unclear; the latter two appear tempted to step down now. Some possible nominees for the character set subgroup were discussed, including a French representative from Grenoble, Harry Gaylord of Groningen, and Ed Hart from the Johns Hopkins Applied Physics Laboratory. Makato Nagao has also expressed interest in having Japanese participation; this has technical as well as political ramifications, as some approaches to character-set problems are supported along national lines, and it may be necessary to include Chinese participants as well or risk premature choice among the various suggestions now being discussed. <p> To replace Thaller and ensure treatment of historical text types, LB suggested, a nascent committee of the Association for History and Computing on the requirements of historical computing should perhaps be consulted. This group will meet in Montpelier in September, at the AHC conference. <!> <h2>6.4. Organization of Committees in General <p> The committee discussed for some time the proper organization of the coming cycle. Consensus was reached that most of the work should be done in small work groups (corresponding to the subcommittees of the first cycle). It was not clear whether the committees of the first cycle should continue to have any prominent role; TD and ML will continue to exist, but there is some possibility that TR and AI will exist only as coordinating bodies to which work groups will report. During the discussion, the consensus appeared to be that they will have no central role, but no formal decision was made to that effect. <p> The work groups should have few members (two or three should be the usual size), short lives (as little as three or four months), and focused charges and work plans. Many should have only one meeting; others might have several, if they could make a case for it and earlier meetings had been productive. <p> It was agreed that the steering committee must decide on a mechanism for deciding on the topics of work to be assigned to the work groups and monitoring their progress. Two proposals were made: that the steering committee should appoint an ad hoc committee (or more than one) to specify the areas needing work, and that the steering committee itself should decide on the areas needing work, possibly forming work groups to prepare advisory papers on the issue. No decision was made. <p> AZ asked the editors to prepare a paper listing the areas they thought needed work groups. They agreed to do this. <action> <who>editors <act>to draft list of work groups needed at present <duedate>asap </action>     <!> <h1>7. Affiliated Projects <p> NI urged that a clear understanding be achieved with the affiliated projects as to what exactly they and we should be providing each other, specifying deadlines for the completion of tagging of a certain amount of text and for reports. Reports should summarize problems encountered using the TEI scheme to express required information and methods used for data capture or tag generation. If possible, there should also be a money payment to the projects. <p> It was agreed that each affiliated project and the TEI should name a contact person to which the other could always send documents or questions; the possibility that there should be a single TEI contact person for all affiliated projects was discussed but not accepted. It was also agreed that a written plan with specifications of data and reports to be submitted, with deadlines, should be negotiated with each project's representative. <p> AZ described some sample types of interaction with the projects: initial technical meeting, consulting and advice during the implementation phase, second technical meeting after the implementation phase for debriefing, reception and routing of output from the project, and reaction to the results. All agreed that each project should have a single TEI contact person to answer questions and give advice (the TEI <term>uncle</term>). MSM urged that the projects not be restricted to contact with the uncle; it should be part of the responsibility of any work group head or member to assist any affiliated project as far as possible, and the projects should have the right to call anyone in the project directly, not just through their uncle. The uncle might suggest, also, that the technical meetings be handled by some other TEI participant rather than the uncle. This would be particularly true where the uncle was a participant in both the TEI and the affiliated project (true of several projects). AZ suggested that the projects might be grouped according to the type of work they are doing, and each group be assigned a common uncle. This will affect and be affected by the work-group structure. <p> NI was asked to draft a sample written plan specifying details of cooperation, to serve as the basis for talks with each project. This written plan should provide further detail within the framework established by the basic agreement with the affiliated projects, document TEI SCP1. <action> <who>NI <act>to draft detailed proposal for cooperation between TEI and affiliated projects <docref>TEI SCA7 <duedate>10 July [received -Ed.] </action> <p> New affiliated projects were formally agreed to: the Nietzsche project of Malcolm Brown at Stanford University, and the Network of European Corpora (contact person: AZ). <!> <h1>8. Final Report to EEC <p> AZ explained that the final report to the EEC should have three parts: an introduction describing the project and the results, the guidelines themselves (version 1), and the financial summary. A substantial amount of money remains unclaimed in the Pisa account; if it is not claimed soon, it will be lost. Methods of dealing with this situation were discussed at some length. [NI left the meeting during this discussion.] <!> <h1>9. Administrative Support <p> MSM described the inconclusive state of his negotiations with UIC management to free more of his time; various scenarios were discussed and their costs were estimated. The budget submitted to NEH with the proposal of September 1989 calls for 50% time for MSM, a continuation of the current level of undergraduate clerical help, and a graduate assistant at 50% time. If 100% of MSM's time is purchased instead, it will raise the relevant salary costs about $50,000 including overhead for the two years. If the line for a graduate assistant is changed into a 50%-time administrative assistant or secretary, it will raise the cost by perhaps as much as $25,000. If two graduate assistants are hired instead of one, the cost will rise about $25,000. The preferred solution is to release the rest of MSM's time; either of the other two solutions may be chosen otherwise, depending upon whom it seems possible to hire. <p> MSM was urged to list in some detail the responsibilities of the editors and those which may and may not be delegated to others. <action> <who>MSM <act>to draft summary of editorial responsibilities for cycle 2 <duedate>unspecified </action> <!> <h1>10. Schedule for Cycle II and Official Responses to Comments <p> The schedule included in the agenda was approved as shown above. <p> The committee agreed that public comments and queries should receive official TEI responses. These should be drafted by the competent working committee or by an editor; a working committee response is official if approved as such by either editor. If either editor sanctions a response as official without first consulting the other, he is to notify the other within one day. The editors are expected to inform the steering committee regularly of such officially sanctioned responses and may consult with them as appropriate. <p> All responses are to be logged in the comment-log database. <!> <h1>11. Software <p> The editors summarized their contacts with software vendors. SEMA group (through Yard Software) has generously made Mark-It available to anyone associated with the TEI, and both editors have copies. LB also has SEMA's Write-It. Software Exoterica has provided evaluation copies of CheckMark (a validating editor for the Macintosh) and XTRAN (a translator for other platforms; editors have the MS-DOS version). Datalogics have promised copies of their Edit Station software but have since fallen silent; MSM should nudge them. SoftQuad has resisted donations on the grounds that they cannot afford them, but their UK distributor seems to be taking a more liberal approach. LB also has a copy of Daphne from the Deutsches Forschungsnetz, but this has a hard-wired DTD. The Turing Institute in Glasgow has promised source code for TOLK, their SGML parser; this supports CONCUR <q>but not in the the way you'd expect.</q> TOLK has a strong Hypercard and database interface. LB and MSM have both been pursuing Michael Cowlishaw's LEXX editor originally developed for OUP and the OED project, in DOS and VM versions, respectively, so far without luck. <!> <h1>12. Unfinished Business <p> The meeting adjourned sine die without acting on the following items of the agenda: <ul> <li>future structure of TR and AI committees and subgroups <li>future role of TR and AI committee heads <li>topics of working groups and names of heads <li>clarification of EEC payments through Pisa <li>publication and distribution of final report <li>Rockefeller grant and conference </ul> <!> </body> <!> </gdoc>