TEI Technical Council F2F Meeting in Newcastle, 11–13 September 2022

Meeting Times

Date Time Meeting of Location
Sunday, Sept 11 09:30–12:30, break at ~10:45 14:00–17:30, break at ~15:45 Full Council meeting * atop meeting Percy Buildingin  G.09
Dinner TBC
Monday, Sept 12 09:30–13:00, break at ~10:45 14:30–17:30, break at ~16:00 Full Council meeting Armstrong Building in G.09.KLLT
17:30–18:30 Joint Board/Council meeting
Dinner TBC
Tuesday, Sept 13 09:00–13:00, break at ~10:45 Full Council meeting Armstrong Building in G.09.KLLT

Present:

Apologies:

Council Members running workshops

SUNDAY, September 11

> Morning session (9:30–12:30)

>> Approval of the agenda

>> Topics for Board consideration

>> Invite potential Council members to the current F2F meeting

>> Council’s report at annual general meeting

>> Tasks to takeover from PS

>> Moving from OxGarage to TEIGarage

>> ROMA

>> CMC

>> Documentation

> Afternoon session (14:00–17:30)

>> Preparation for the next release

>> Housekeeping

>> Review of Guidelines Issues

MONDAY, September 12

> Morning session (09:30–13:00)

>> Invited guest: TOC and NL

>> Discussion and review of Guidelines Issues

> Afternoon session (14:30–17:30)

>> ATOP report:

>> Stylesheets issues

>> Discussion of Guidlines issues

>> i18n report

> Board/Council meeting (17:30–18:30)

TUESDAY, September 13

> Morning session (9:00–13:00)

>> Discussion of Guidelines issues

Guidelines issues from SVF2F meeting in October 2021

Issues from the October 2021 meeting discussed by group B (European Group: MS, PS, MT, HBS) but on which no final decision was by the plenary.

#2136 B Use PureODD inside attribute lists Subgroup recommends to deprecate @org on <attList> and use schematron rules instead. A few elements use <attList org=""> at the moment, e.g. <relation>. These would have to be re-written. SB: There is no comment on the ticket to this effect, so I am commenting here. I think this [using Schematron instead] is a terrible idea. One can express this (alternation of attributes) in RELAX NG, XML Schema, and even in ODD (except that it does not work in classes). There is no reason it should not remain expressible in ODD. 2022-09-11: SB left a comment in GitHub; issue closed, Stylesheets issue reopened

Guidelines issues from SVF2F meeting in April 2022

Issues from the SVF2F meeting in April 2022 discussed by group B (European Group: MS, PS, MT, HBS) but on which no final decision was made by the plenary. Group A: (HC), EBB, JJ, SB, RV

Group B: PS, HBS, MS, SS, EB

No. Group Title Comment
#2137 B bookmark links in examples pages have no mouse-over text Subgroup proposes to 1) add @rel="bookmark" and @aria-label on HTML output <a>. Consider opening a new issue for citation recommendation (issue with this is which link should be in the citation recommendation, the current release or stable link to the Vault). MT: Discuss issues arising from the request to offer citation buttons / bookmarks: tension between human readable and significant citation, e.g. TEI Guidelines, Reference for element <abbr>, Example 3 vs using a generated example id, direct bookmark link; latest vs fixed version (permalink) HC: Could we solve the citation problem (that the example id changes every version) by redirecting /latest/ to the current version?
#2140 B Example needed to model attList for delimiting alternative groupings of attributes Subgroup thinks this is great and should be added to the Guidelines after the Stylesheets have been prepared to support both the HTML output and the overriding of @type & @subtype as mentioned on the ticket. For the HTML output there’s already a ticket but for the overriding we need one. 2022-09-11: green, EBB will provide a reduced example
#2144 B wrong attribution for example Subgroup proposes (1) to change the whitespace to a hyphen; and (2) to add the same example with @ref instead of @key plus URL as value. 2022-09-11:SB will come up with an improved example
#2190 Should we revise datatypes related to sex and gender? This is related to issue #2189: Create a gender element which is yellow and green 2022-09-11:CLOSED with all the action on gender branch.
#2213 att.datatable.custom is under-documented in the Guidelines This issue will be discussed on Saturday in the full Council session Where we introduced @datingPoint in the prose of the Guidelines (with ref to <calendar> element isn't helpful yet. 2022-09-11: EB and MH will come up with an example. Work is proceeding on this branch: https://github.com/TEIC/TEI/tree/issue-2213

Guidelines pull requests

No. Group Title Comment
#2245 Translation from CarmendeSantiago/TEI Review HBS
#2281 Translation from knagasaki/TEI Review with MH, but probably ready to pull in
#2294 Translation from knagasaki/TEI Review with MH, but probably ready to pull in
#2320 Translation from dh-miami/TEI Review HBS
#2321 Translation from joloor2/TEI Review HBS
#2323 Translation from casanso/TEI Review HBS
#2328 rename unboundedInt to unboundedCount Review EBB and EB

Guidelines issues

Group A: HB, EBB, HC, RV, NL Group B: SB, MT, TOC, PS

Group C: MH, EB, MS, SS

—-

Group D: SB, HB, SS

Group E: MT, MH, EBB

Group F: HC, MS, PS

No. Group Comment
#2215 A att.pointing for <app>? Subgroup: There’s no need to do this. Close won’t-fix. Maybe consider tweaking the note on @from. Council discussion: Green for EBB to tweak note and provide a simple example on the Spec page.
#2216 B Example in rhyme.xml looks weird Assigned to JJ to get in touch with Christian Wittern.
#2221 C Why doesn't divGen have the same content model as div? Subgroup: <divGen> is not supposed to have any content at all: it is supposed to be empty. So technically it should not contain <head> either, but in view of the Birnbaum doctrine we cannot make its data model empty. We should just close the ticket. Ticket is closed.
#2222 A tagdoc for egXML speaks of attr in TEI namespace Subgroup is repairing this on the spot. Error appears on the <egXML> spec, in the example. “used in the TEI namespace” was removed and the verb now agrees with the subject. Ticket is closed.
#2225 B missing space Leave for JJ - add space after or before <cl> and surround the example with <l> because it is verse (hence the uppercase T in “They”). The same problem occurs in the following example (which is in French). Green for JJ to fix.
#2227 C Update ISOCat reference to DatCatInfo MS: there is a related issue – I hope that MH can help with this Does login for https://datcatinfo.net/ work for everybody? It seems that attempts to access actual content in datcatinfo.net require logins, so we can’t link there. Linking directly to terms prompts login. Council is not in favor of using DatCatInfo. Reach out to LingSIG and Piotr Banski to come up with some wording to indicate the need to refer to a standard (and also to suggest the standard, if there is one). We need to remove all links to ISOCat anyway.
#2228 A figure element in model.titlepagePart class? Subgroup: Wordsmithing required to correct the false impression that <figure> is a member of model.titlePart. Subgroup is checking in this change now. Closed.
#2242 B Another lg content model deficiency Subgroup discussed several options: a) allow for <supplied> since it’s already allowed within <ellipsis>; or b) do not allow for <supplied> since it allows for too many elements not originally allowed within <lg>; or c) propose a new element <suppliedSpan> that would allow the editors to supply structural information (like lines) without allowing any further elements; d) make model.pPart.transcriptional part of the <lg> content model (suggested by RV; see below). Council discussion: If we go with a), provide strong guiding text in the Guidelines: What goes in <supplied> should be what belongs in <lg>.RV suggests we can add model.pPart.transcriptional to <lg>’s content model. This gets approval.
#2246 C att.spanning value targeting an external document Subgroup checked the current schematron rule in att.spanning.xml and, following HC’s suggestion to change it, we suggest the following: assert test="not(starts-with(@spanTo, '#')) or(id(substring(@spanTo,2)) and following::*[@xml:id=substring(current()/@spanTo,2)])" This change requires testing of course. Green for MT.
#2247 A <handShift/> should be a member of model.global Subgroup thinks `@hand` should suffice, and possibly that att.written might be expanded if someone needs more elements to be included. RV opens a new ticket to add a sentence to chapter 11, making the use of @hand and <handShift> more clear.
#2248 B textLang should be able to contain bibl Subgroup suggests to a) keep the current macro and add only <bibl>, or b) widen the content model to macro.specialPara (like other header elements). Council decides on b) and marks the issue as green.
#2249 C Update data.temporal.iso Subgroup: The first (and easiest) thing to do would be to simply remove the remark, which is what we did. The second action point would be to update to ISO8601:2019 and allow users to only use @when-iso, while still permitting the old expressions like @from-iso, etc. https://tei-c.org/release/doc/tei-p5-doc/en/html/ref-teidata.temporal.iso.html. It’s not clear to us that the semantics of @notBerfore-iso and @notAfter-iso are actually supported by ISO 8601’s Extended Date-Time Format; if they are, could someone who knows 8601 well explain how?
#2251 A Need to update ODD transformation scenarios Subgroup asks: was this fixed already? Check with MH then close. Closed.
#2252 B Misuse of @xml:lang in an example for msPart Mark as green
#2254 C <measure> should be a member of att.ranging Work in progress, SB and EB will come back to Council as soon as they have a proposal.
#2258 A msIdentifier constraint should test node instead of its name; and does it make sense? Subgroup thinks SB should make the Schematron test for the node rather than local-name() or name(). 1. We should improve the prose of the Schematron message. For example, this message (“An <msIdentifier> must contain either a repository or location.”) fires if we introduce an empty <collection>, when just adding contents to the element would make it valid. 2. The test shouldn’t force that the elements are not empty (as SB showed, examples with @key are perfectly valid) Be clear what we’re testing for, or whether there’s a need for it at all. This test might have been overly strict in the first place. Mark as green.
#2261 B no need to specify scheme when deleting a constraint specification Subgroup agrees that @scheme should not be required (and perhaps not allowed) if @mode is “delete”, but also wonders about a different solution (see ticket). Furthermore this issue raises the “what happens if you have multiple <constraintSpec>s in chained ODDs that have the same @ident?” question. Council agrees with the subgroup’s proposal, green for SB.
#2262 C need examples of changing or deleting constraint specifications Council discussion: we should provide an example that demonstrates and discusses deletion of any kind of spec (not just constraintSpec). Nothing too long. Discussion: Suggestion to create an example demonstrating how to work with att.combinable. mode="change" is much more complicated than mode="delete". But mode="change" is better exemplified. Green + Yellow for EB to work on examples for deletion and replacement.
#2267 A detest target needs to be more comprehensible SB added a note some days ago asking if this can be closed. Subgroup thinks, probably yes? Council: PR is merged, issue closed.
#2271 A broken links on BIB.html HC suggests checking for the broken links on the Internet archive and updating them. Subgroup greenlights.
#2272 A various broken links to external resources Subgroup thinks similar recommendation as for #2271 (check for where these external resources have moved).
#2273 D Dictionaries:Homograph number is a property of the headword SB suggests to ask  @chr-emil: “Do you still want <hom> as a child of <form>, or have you decided to use some other method? If you still need it, we (Council) think we should put <hom> into model.formPart.”
#2274 E example of decls= shows incorrect non-usage of default= Subgroup thinks this is more complicated than SB’s ticket implies. Should the @xml:lang on <metDecl> or the elements effectively decide which one is default? Should we arbitrarily declare one default? Why must the Guidelines require a default? Default should be for guiding the processing, but it may not be required if it’s following the language declared. The example already illustrates how to reference the rhyming pattern explicitly for a piece.
#2275 F Nabble TEI-L Archive No Longer Exists Fixed
#2276 D shouldn’t the class= attribute of the attRef element be a name? Subgroup made this GREEN for EB to implement.
#2277 E Seeking examples for att.written on stage MH has added an example on the ticket to get us started. JJ and EBB to work on this in a branch.
#2279 F Constrain translatable elements in TEI ODDs Almost done; duplicate French gloss is actually mislabeled Spanish; mark green
#2282 D attributes of <attRef> Subgroup thinks both attributes should be required. Council: GREEN to require both. <attList org="choice"> <attRef class="att.typed"/> <attRef class="att.datable"/> </attList>
#2285 E more on <altIdent> Subgroup believes that: For the ODD stage where RNG is being generated, there should only be a single <altIdent> at most. However, earlier in the processing chain, it is reasonable to have different <altIdent>s for different languages; it makes sense, for instance, to have Arabic <altIdent>s for convenience when working RTL, with another LTR language <altIdent> alongside it. So we think that the requirement should be that if you have multiple <altIdent>s, they must all have different @xml:lang values. The processor would then select the correct language <altIdent> before generating the PLODD (to use ATOP terminology) and then have no further decision to make when generating the RNG.On the issue of elements where <altIdent> can never end up in the schema at all, we see no reason to retain <altIdent>, and it would be clearer to deprecate and remove it.
#2286 F errors in the generation of templates Mark as green; ask SB about the current status.
#2292 D Allow <listBibl> in <abstract> Subgroup suggests to allow <listBibl> either in <abstract> or to add it to model.listLike (group B is partial to the latest if it doesn’t cause too many problems). Council agrees to SS’ proposal. GREEN.
#2295 E Add an organism element Subgroup thinks that rather than creating elements for more specific types of things, it would be better to create a more generic <entity> element, of which all the specifics we have (<person>, <org> etc.) are all syntactic sugar. This way, whatever your ontological preferences in dividing the world, you can use <entity type="organism"> or whatever.  SB:+1 Council decision: Close this ticket and open a new one to propose <entity>, <listEntity>, and create a content model. Should there be a task force? YES Do not merge with #367, says PS <entity> can be used as a more generic and customizable mechanism when existing “ography” constructions don’t work. Issue closed, new issue #2341 opened.
#2298 (continue with review in October 2022 meeting) F description vs gloss in @type of <divGen> Mark as green; go ahead
#2300 D restrict embedded transcription to sourceDoc / exclude text in zone and line as part of facsimile Subgroup suggests marking it as green.
#2303 E URLs of projects using TEI are not displayed on the individual project info pages Subgroup inquires: where is the TEI underlying the projects page? And where is the XSLT that processes it? For some entries (probably older ones), it looks like some TEI elements (like <ptr>) are embedded in the HTML, while others are just HTML. The TEI needs to be cleaned out of the page.
#2306 F tagdocs elements are available in silly contexts Subgroup: that would be a breaking change and we’re not sure whether it’s worth the effort. Might make a case for an ATOP-specific application profile (customization)? Does SB remember why the membership in model.inter came about?
#2314 D TEI example is old-formatted Subgroup suggests marking it as green.
#2315 E example example discusses a possible future feature Subgroup agrees that changing the item is a good idea to encourage best practices.
#2316 F <speaker> as content of <lem> Subgroup: we’re not against it, but it seems too hard to implement (i.e. too much of a stretch for the content models of <lem> and <wit>); and for the currently provided example there are several workarounds in the comments.
#2317 D <app> as sole content of a <sp> element Subgroup thinks this should be allowed, but note that since <app> is a member of model.global.edit it is going to be very difficult (if not impossible) to do. <app> should have a warning informing encoders that it’s their responsibility that the contents of each <app>/<rdg> or <app>/<lem> is valid within the parent of the <app>.
#2326 E constraints on <dataFacet> and @restriction should be explicit PS has already solved and closed this.
#2327 F teidata.unboundedInt should be named teidata.unboundedCount Seems sensible. Subgroup thinks we should merge the PR and close. Do we need deprecation of teidata.unboundedInt before nuking it?Council: no, deprecation is not necessary.
#2330 D schemaSpec should provide a mechanism for specifing Schematron query language binding
#2331 E bad link → consistent referencing needed to ISO 24611 Subgroup suggests green for go to create a new bibliography entry and fix all references to point to it.
#2332 F How are we to refer to an <egXML>? (Or <figure> or <table>?) Subgroup: This should be addressed and added to TCW20.