February 2015 - Tei-council

Recommendations on values of @xml:id (FR #540)
by Raffaele Viglianti 31 May '15

31 May '15

Hi all, Martin Mueller asked on the list for recommendations on good uses of xml:ids, which resulted in a good discussion and a FR. I summarized the main suggestions on the ticket on SourceForge, but I post them here as well because I think we need to discuss this further, see below. The Guidelines already have some recommendations (numbering based on doc structure): * The Guidelines already suggest numbering based on doc structure ( http://www.tei-c.org/release/doc/tei-p5-doc/en/html/CO.html#CORS2) * Use 3 letters (e.g. from the title)+ 3 digits, incremental. E.g. HOL001, HOL002, etc.. * Same as above, but no fix number of digits. E.g. HOL1, HOL2, etc. * Prefix an id with name of element (this is simpler version of what the Guidelines already recommend) * Give tei:TEI an id and prefix every other id in the document with it (to guarantee cross-corpus uniqueness) These are all reasonable suggestions and there can be plenty more - which is why I think that the TEI should *not* give any recommendation on best practices for xml:id because it's a project management issue, not an encoding one. Myself, I prefer random ids, a practice that avoids introducing yet another level of complexity and data management. I understand the human readability, but sequences are too easily broken. And when parsing, relying on ID content instead of TEI content sounds like a bad idea.

7 16

MsPart Comments
by Hugh Cayless 13 Mar '15

13 Mar '15

Hoping to spark some comment. As you might expect, I'm *strongly* in favor of permitting the use of msPart for describing fragments, so all I have are quibbles: 1) I don't thing the distinction should be "composite" vs. "dispersed". The fragments of a broken up (e.g. papyrus) text aren't necessarily scattered (though they may be), but the text *is* in bits rather than joined or bound together, and those bits may each have their own identifiers and their own histories. 2) In the proposed text rewording "only remaining fragment of a former codex", the word "codex" excludes papyri and inscriptions. 3) I'm not sure I understand the altIdentifier/msPart issue...

7 17

correspAction@type
by Peter Stadler 13 Mar '15

13 Mar '15

The last changes made by Lou introduced the suggested values „sentBy“ „deliveredTo“ etc. to correspAction@type [1]. I’d like to propose skipping the preposition and simply call those „sent“, „delivered“ etc. because we usually describe three things within <correspAction>, i.e. the date, the place and the actor. Since there is not one preposition that (grammatically) fits all items, and because I like the analogy with email terminology, I think it’s best to only have „sent“, „received“, „forwarded“, „redirected“ and „transmitted“. Cheers Peter [1] http://teijenkins.hcmc.uvic.ca/job/TEIP5/lastSuccessfulBuild/artifact/relea…

3 3

next oXygen release
by James Cummings 28 Feb '15

28 Feb '15

As I was asking them about something they told me would be in the oXygen 17 release (being able to populate oXygen forms properties from remote files, e.g. in lists of attribute values from a remote file being used in a dropdown menu) I happen to ask them when oXygen 17 was planned. George answered saying: On 28/02/15 16:16, oXygen XML Editor Support (George Bina) wrote:> Hi James, > > We plan oXygen 17 for end of April / beginning of May. > > Best Regards, > George So that is after our intended release... so let's make that release on time!! -James -- Dr James Cummings, James.Cummings(a)it.ox.ac.uk Academic IT Services, University of Oxford

2 1

msDesc proposal
by Martin Holmes 28 Feb '15

28 Feb '15

I seem to be unable to comment on the msDesc proposal here: <https://docs.google.com/document/d/1bpBPughQp5sOoB13iTMHMeRd88Mjv12_eGZiQXY…> so I'll raise a quick question here. I'm a bit confused about the second definition; where it says: "In this case, the original manuscript is being encoded with msPart in order to provide information about the dispersed parts, or one distinct part is being encoded." Does this mean that, conceivably, the "larger original manuscript" is being encoded with a _single_ msPart, or that it's being encoded with a _series_ of msPart elements, one for each component? Cheers, Martin

5 9

Friday's Telecon; Agenda Items?
by Hugh Cayless 27 Feb '15

27 Feb '15

Dear all, A reminder that our next teleconference will be this Friday, February 27, at 05:30 PST | 08:30 EST | 13:30 GMT | 14:30 CET. I've posted a draft agenda at http://wiki.tei-c.org/index.php/Council_agenda_2015-02 Please add, remove, or modify agenda items as necessary. There are a couple of items we could profitably discuss beforehand on-list, including the transcription of spoken language draft and the suggested msPart updates. All the best, Hugh

5 6

My Conference Info
by Hugh Cayless 27 Feb '15

27 Feb '15

Hi all, I’m having trouble finding the list of international numbers on the Economy Conference Call site, but here’s what was in the email they sent me. You should be able to dial the toll free (US, so prefix +1) number using Skype. Toll-Free: 866-906-0040 Toll: 857-288-2640 Code: 5753850 Here’s the set of number changes that John forwarded me: Locale Old Dialed Number New Dialed Number GERMANY (Land) 08001820270 08001800565 GERMANY (Mobile) 08001820270 08001814978 UK (London Local) 442034785604 442033186423

3 3

Agenda for Friday's Telecon
by Hugh Cayless 27 Feb '15

27 Feb '15

Hi all, I’ve put up an agenda for our teleconference on Friday, Jan 30 at 14:00 GMT: http://wiki.tei-c.org/index.php/Council_agenda_2015-01 <http://wiki.tei-c.org/index.php/Council_agenda_2015-01> It’s mostly a copy of the previous one, on the assumption that we probably didn’t make much progress over the holidays and the review will likely be good for all of us, especially our new members. Please do add new items for discussion (or remove items there’s nothing to report on). It will take me a while to figure out how things work, so reminders of anything I may have missed, or suggestions for how things might be done better are most welcome! All the best, Hugh

8 17

Minutes Link
by Hugh Cayless 27 Feb '15

27 Feb '15

https://docs.google.com/document/d/1bfUCykhiABF7JP_KqSyprrI8n-qgcnMstJhBOuV…

1 0

feedback requested
by Lou Burnard 26 Feb '15

26 Feb '15

Could someone please comment on my proposed actions concerning this longstanding FR? https://sourceforge.net/p/tei/feature-requests/480/ It's the one about classifying the use of @hand better...

2 1

F2F poll results
by Hugh Cayless 25 Feb '15

25 Feb '15

Looks like we have (sort of) a winner: the week of April 27–May 1. Raffaele marked "if need be" for the 27th–29th and Lou and Fabio for the 29th–May 1. How much of a problem are those conflicts? In particular, would travel to the US or Europe make them impossible? If I’m not mistaken, these are the offers to host we’ve had so far: ## US Boston (Syd) College Park (Raffaele) Ann Arbour (Paul) ## Canada Victoria (Martin) ## Europe Mainz (Peter) Estonia (maybe…) Is that correct? Other takers or ideas? Hugh /** * Hugh A. Cayless, Ph.D * hugh.cayless(a)duke.edu * Duke Collaboratory for Classics Computing (DC3) * http://blogs.library.duke.edu/dcthree/ **/

11 59

Suggested typology for correspAction
by Lou Burnard 18 Feb '15

18 Feb '15

I'd like to propose some change to this. Firstly I find the current descriptions of the various valItems a bit cumbersome and non-idiomatic. e.g. "identifies a/the sending action of the message" The use of "a/the" for "a or the" is not something we do elsewhere, nor do I think it necessary here. It's not obvious what "the sending action of the message" means either -- the message is not carrying out the sending! The description should explain what this value of the @type attribute implies, and therefore in what circumstances it should be chosen. I suggest something like "information concerning the creation or authorship of a message". I'd prefer to see the suggested values as nouns rather than gerunds, as elsewhere. So "despatch" rather than "sending", "receipt" rather than "receiving". (But I appreciate that this may cause some upsets in existing data) I would like to see a suggestion for the "authoring" correspAction, since the text does talk about this (when introducing use of @sameAs) The distinction between "receiving" and "transmitting" needs to be clarified. I would say the same for the distinction between "redirecting" and "forwarding", except that I am not sure that I know what the distinction is in that case. So, my revised valList for correspAction/@type looks like this: <valList type="semi" mode="add"> <valItem ident="creation"> <desc versionDate="2015-02-09" xml:lang="en"> information concerning the creation or authorship of a message</desc> </valItem> <valItem ident="despatch"> <desc versionDate="2015-02-09" xml:lang="en"> information concerning the sending or despatch of a message</desc> </valItem> <valItem ident="receipt"> <desc versionDate="2015-02-09" xml:lang="en"> information concerning the receipt of a message</desc> </valItem> <valItem ident="transmission">  <desc versionDate="2015-02-09" xml:lang="en"> information concerning the transmission of a message after it has been despatched</desc> </valItem> <valItem ident="redirection">  <desc versionDate="2015-02-09" xml:lang="en"> information concerning the redirection or forwarding of a message </desc> </valItem> </valList>

6 14

correspAction example
by Lou Burnard 18 Feb '15

18 Feb '15

In Shakespeare's Twelfth Night, the clown Feste carries a letter to the steward Malvolio. The letter is unsigned, but Malvolio thinks it is from the lady Olivia. In fact, it was written by her maid Maria, in collaboration with Sir Toby Belch. So... <correspAction type="deliveredTo"> <persName>Malvolio</persName> </correspAction> <correspAction type="deliveredBy"> <persName>Feste</persName> </correspAction> <correspAction type="sentBy" subtype="mistaken"> <persName>Olivia</persName> </correspAction> <correspAction type="sentBy" subtype="actual"> <persName>Maria</persName> </correspAction> <correspAction type="sentBy" subtype="actual"> <persName>Sir Toby Belch</persName> </correspAction> Does that sound plausible? Can someone come up with another better example showing a variety of correspActions?

1 0

Spring meeting
by Hugh Cayless 16 Feb '15

16 Feb '15

Can I encourage those of you who haven't participated in the "let's just choose a date and place" poll at http://doodle.com/fk8qbq6rrrn7qugq to do so if you care at all about the outcome? Right now, we're leaning (by one vote) toward Durham at the end of April. I'll plan on closing the poll on Monday. All the best, Hugh

7 14

correspDesc implementation
by Peter Stadler 11 Feb '15

11 Feb '15

Dear all, a brief report on my work on correspDesc. (I’ll respond to Lou’s input separately) (I’m a little bit behind with documentation because my children passed their illnesses on to me …) So, I implemented correspDesc in revision 13129 and only needed three shots to make the Jenkins servers build the Guidelines (Yeah!). That said, I needed several tries locally before (which to my surprise worked almost flawlessly). I found it especially useful to run only the 'p5.xml‘ target which is quite quick and will create p5subset.xml along with p5.xml. This was helpful, because my naive assumption was that everything under P5/Source/Specs would automagically be included. It is not, you rather need a <specGrp> (in some Guidelines chapter?) with Xincludes of the files. * So, question 1: is this a bug or a feature? if it’s a feature (what I do think, because we do not implement elements without accompanying prose, do we?) it should probably be documented in TCW20 or did I missed it?! * question 2: how are the xml:ids (for <specGrp>s, <div>s etc) within the Guidelines chapters created? I just made them up … Best Peter

3 7

Backlinks at element description
by Peter Stadler 11 Feb '15

11 Feb '15

How do I get those nice backlinks (see the link "[2.4.4 Abstracts]“ at http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-abstract.html) into my correspDesc element description. Is this listRef/ptr-magic? Best Peter

2 1

correspDesc desc update
by Lou Burnard 11 Feb '15

11 Feb '15

As most of you will have noticed, over the weekend Peter quite rightly added a new section to the Header chapter describing correspDesc, along with associated ODDs to implement the proposals. I feel very guilty for not having found time to feedback suggestions for the new text (which I promised to do months ago).,.. but here they are now! I haven't checked the following in yet, but am happy to do so. As the section is quite short, I am just pasting it into this message. We're all grown ups here and can cope with a few TEI tags, right? My revisions are mostly to enlarge and make a bit more explicit what the examples are doing, plus a few other things changed for conformity with house style. I have however reordered the original, putting correspContext first, since it is simpler and shorter than the rest and I didn't want to introduce new subsections. I am not certain that I've properly understood the last and first examples, so please read my rewording carefully! -------- <div type="div3" xml:id="HD44CD"> <head>Correspondence Description</head> The <gi>correspDesc</gi> element is used within the <gi>profileDesc</gi> element to provide detailed correspondence-specific metadata, concerning in particular the communicative aspects (sending, receiving, forwarding etc.) associated with acts of of correspondence. This information is intended to complement the description of the objects recording correspondence activities (such as letters), typically provided by the <gi>sourceDesc</gi> element. <specList> <specDesc key="correspDesc"/> </specList> The <gi>correspDesc</gi> element contains the elements <gi>correspAction</gi> and <gi>correspContext</gi>, describing the actions identified and the context in which the correspondence occurs respectively. <specList> <specDesc key="correspAction" atts="type"/> <specDesc key="correspContext"/> </specList> Acts of correspondence typically do not occur in isolation from each other. The <gi>correspContext</gi> element is used to group references relevant to the item of correspondence being described, typically to other items such as the item to which it is a reply, or the item which replies to it: <egXML xmlns="http://www.tei-c.org/ns/Examples"> <correspContext> <ref type="replyTo" target="#CLF0102">Previous letter of Chamisso to de La Foye: 16 January 1807</ref> <ref type="replyFrom" target="#CLF0104">Next letter of Chamisso to de La Foye: 07 May 1810</ref> </correspContext> </egXML> Many types of correspondence action may be distinguished. The <att>type</att> attribut should be used to indicate the type of action being documented, using values such as those suggested above. The following basic example uses <gi>correspAction</gi> to describe the sending of a letter by Adelbert von Chamisso from Vertus at 29 January 1807 to Louis de La Foye at Caen. The date of reception is unknown: <egXML xmlns="http://www.tei-c.org/ns/Examples"> <correspAction type="sending"> <persName>Adelbert von Chamisso</persName> <placeName>Vertus</placeName> <date when="1807-01-29">29 January 1807</date> </correspAction> <correspAction type="receiving"> <persName>Louis de La Foye</persName> <placeName>Caen</placeName> <date>unknown</date> </correspAction> </egXML> Several senders, recipients, etc. can be specified for a single <gi>correspAction</gi> if the action is considered to apply to them all acting as a single group. In the following example three people are considered to have received the communication. <egXML xmlns="http://www.tei-c.org/ns/Examples"> <correspAction type="receiving"> <persName>Hermann Hesse</persName> <persName>Ninon Hesse</persName> <placeName>Montagnola</placeName> </correspAction> </egXML>  The <att>subtype</att> attribute may be used to further distinguish types of action. In the following example, an e-mail is sent to two people directly and to a third by blind copy (<term>cc</term>). <egXML xmlns="http://www.tei-c.org/ns/Examples"> <correspAction type="sending"> <persName>PN0001</persName> <date when="1999-06-02">Wed, 02 Jun 1999</date> </correspAction> <correspAction type="receiving" subtype="to"> <persName>PN0002</persName> </correspAction> <correspAction type="receiving" subtype="to"> <persName>PN0003</persName> </correspAction> <correspAction type="receiving" subtype="cc"> <persName>PN0004</persName> </correspAction> </egXML> The same person may often be associated with many actions. For example, it will often be the case that the author and sender of a message are identical, and that many individual letters will need to be associated with the same person. The <att>sameAs</att> attribute defined in section <ptr target="#SAIE"/> may be used to indicate that the same name applies to many actions. Its value will usually be the identifier of an element defining the person or name concerned, which is supplied elsewhere in the document. <egXML xmlns="http://www.tei-c.org/ns/Examples"> <correspAction type="sending"> <name sameAs="#author"/> </correspAction> </egXML>  It is assumed that each correspondence action applies to a single act of communication. It may however be the case that the same physical object is involved in several such acts, if for example person A sends a letter to person B, who then annotates it and sends it on to person C, or if persons A and B both use the same document to convey quite different messages. In such situations, multiple <gi>correspDesc</gi> elements should be supplied, one for each communication. In the following example, the same document contains distinct messages, sent by two different people. to the same destination: <gi>correspDesc</gi> is used for each separately: <egXML xmlns="http://www.tei-c.org/ns/Examples"> <correspDesc xml:id="message1"> <correspAction type="sending"> <persName>John Gneisenau Neihardt</persName> <placeName>Branson (Montgomery)</placeName> <date>17 December 1932</date> </correspAction> <correspAction type="receiving"> <persName xml:id="JTH">Julius Temple House</persName> <placeName>New York</placeName> </correspAction> </correspDesc> <correspDesc xml:id="message2"> <correspAction type="sending"> <persName>Enid Neihardt</persName> <placeName>Branson (Montgomery)</placeName> <date>17 December 1932</date> </correspAction> <correspAction type="receiving"> <persName sameAs="#JTH"/> <placeName>New York</placeName> </correspAction> </correspDesc> </egXML>

2 3

Bug and FR assignments
by Hugh Cayless 10 Feb '15

10 Feb '15

Hi all, As promised, I’ve gone through the unassigned bugs and feature requests and assigned them, mostly arbitrarily based on who’s currently working on the fewest things. For the new folks: being assigned a Feature Request doesn’t mean just go do it :-), it means you’re responsible for making sure it gets discussed adequately and then implemented, revised, broken up into smaller implementable chunks, rejected, quietly neglected for years, etc. Do speak up if you think an assignment is grossly unfair! I haven't assigned the latest FR, https://sourceforge.net/p/tei/feature-requests/546/ <https://sourceforge.net/p/tei/feature-requests/546/> yet, because it looked like there’s much active discussion with possibly some consensus emerging. All the best, Hugh

1 0

Development version of oxygen-tei plugin available
by Martin Holmes 08 Feb '15

08 Feb '15

Hi all, I now have a development build of the oxygen-tei add-on working. You can subscribe from here: <http://teijenkins.hcmc.uvic.ca/job/oxygen-tei/lastSuccessfulBuild/artifact/…> If you're brave enough to try it, uninstall the regular version of the add-on before installing this one. I can also confirm that James was right: you can provide access to multiple versions of the plugin, allowing people to roll back to a previous version if something breaks, by providing multiple <xt:extension> elements in the updateSite.oxygen file. This seems to be working well with the URL above: if, when you're subscribing to the plugin (Help / Install New Add-Ons) you uncheck the "Show only the latest versions of the add-ons", you'll see there are three versions available. I have it set up so we'll keep the last ten versions available. I've subscribed myself, so I should be able to see if anything goes wrong. If you do test it, be aware that if you have transformation scenarios (for instance) that depend on the absolute location of the TEI extension framework, those paths will no longer work; Oxygen stores the add-on data in a path which includes a modified version of the URL of the updateSite.oxygen file. It's easy to change back to the standard TEI release, though. A new version of the plugin will be built and "released" every time the Stylesheets or P5 are rebuilt; the former triggers the latter, so the overall process takes quite a while, but the build of the oxygen-tei plugin itself takes only a few seconds. I'm copying Sebastian on this because my Jenkins is still reporting build failures and successes to him, as the Oxford one does to me, and I want to confirm that this should continue (or if it shouldn't, then what should happen instead). I'd also like to get this setup working on the Oxford Jenkins so we have some redundancy. The next step is to create some tests we can run in Oxygen to check core functionality of the add-on. I envisage this as an Oxygen project with a bunch of sample files that are transformed or otherwise manipulated and the results checked. These files would presumably live in the oxygen-tei repo. Cheers, Martin

5 27

getting mad with makefile
by Fabio Ciotti 07 Feb '15

07 Feb '15

Hi all, I'm struggling with Makefile from svn repo. Any time I try to make I have this error BUILD FAILED /media/Archivio/src/tei-code/P5/antbuilder.xml:102: Fatal error during transformation using /usr/share/xml/tei/stylesheet/odds/odd2relax.xsl: A sequence of more than one item is not allowed as the result of function tei:generateRefPrefix() ("TEI_model.contentPart", "TEI_model.contentPart", ...) ; SystemID: file:/usr/share/xml/tei/stylesheet/odds/teiodds.xsl; Line#: 2339; Column#: -1 The first time I've run it this morning (after implementing a ticket) it has worked. Then before to check I thought to implement a couple of other tickets, make again to see if something had broken and the error jumped out, but it has apparently nothing to do with my modifications. I have already tried to: make clean 1000 times delete and download again my stylesheets deb try to use online stylesheets as referenced in the makefile (nope the URL is broken) re-checkout the trunk and try the build with an untouched svn local copy curse all the possible deities hit the pc with a baseball bat any hint??? f

3 4

Conferencing system
by Fabio Ciotti 06 Feb '15

06 Feb '15

Hi all, this link could be of some interest. http://crln.acrl.org/content/76/2/98.full Our voice conf system works only by wired phone, and nowadays it is very hard to find one if you are out of office. It is quite annoying to beg for a wired phone when you for ex. are at a conference, where usually you have a fast access to the net (like me last week!). Fabio

5 11

Teleconference dates
by Hugh Cayless 04 Feb '15

04 Feb '15

We settled on the last Friday of the month, at 13:30 GMT, but note the weirdness over time changes in March—I can’t do 09:30 EDT, as it will be then. These are the teleconference dates and times, though August: February 27, 05:30 PST | 08:30 EST | 13:30 GMT | 14:30 CET March 27 05:30 PDT | 08:30 EDT | 12:30 GMT | 13:30 CET April 24 05:30 PDT | 08:30 EDT | 13:30 BST | 14:30 CEST May 29 05:30 PDT | 08:30 EDT | 13:30 BST | 14:30 CEST June 26 05:30 PDT | 08:30 EDT | 13:30 BST | 14:30 CEST July 31 05:30 PDT | 08:30 EDT | 13:30 BST | 14:30 CEST August 28 05:30 PDT | 08:30 EDT | 13:30 BST | 14:30 CEST We’ll cancel one of April, May, or June because the F2F should fall in one of those months. If you haven’t done so yet, please put in your availability for the poll I posted yesterday. If the time shift in March will cause problems, please object now so we can reschedule. I do wish they’d get rid of Daylight Savings Time entirely, but even if they won’t, I wish we could have it at least be coordinated…sigh. Thanks, Hugh

2 1

Re: [tei-council] Development version of oxygen-tei plugin available
by Martin Holmes 03 Feb '15

03 Feb '15

Hi there, On 15-02-03 09:44 AM, Sebastian Rahtz wrote: > happy to set up the job on oxjenkins, shall I do that now or wait a > bit? Go for it. I just tested some of my normal stuff using the new plugin, and it seems to work a treat so far. You'll find the config.xml file for Jenkins in the usual place. The job has to be called "oxygen-tei" (there are some paths in the build-for-jenkins.sh script that depend on that). If you read that script, you'll see it's heavily based on your update-and-upload.sh script, but there are some nifty shenanigans around making the updateSite.oxygen script permanently available at the same URL, while behind the scenes it's shifting its location and has to be regenerated every time based on the previous version. But as far as I know, if you just install the config.xml in a job called oxygen-tei, it'll work. I'm keeping ten versions of the plugin, because we need to let people roll back when things go wrong, and because it gets built often (due to changes in itself, the Stylesheets, or P5). I know your Jinks occasionally has space issues, so that might be something to watch out for, but it's not huge. > the work monitoring oxjenkins is fairly small, so leave that with me > for now. OK. >> The next step is to create some tests we can run in Oxygen to check >> core functionality of the add-on. I envisage this as an Oxygen >> project with a bunch of sample files that are transformed or >> otherwise manipulated and the results checked. > > do you know how to drive oxygen from the command line, then? cool. No, I was planning to write some ant tasks that would be run by a user in the Oxygen interface. Along the lines of: - Create sample file based on template X. - Transform using Stylesheets transformation Y. - Validate result with schema Z. That sort of thing. Not very sophisticated, but the idea would be that when we're about to release a new version of the plugin, someone subscribed to the latest Jinks build would run the tests and confirm nothing is wildly broken, then we'd go ahead and build/upload to the release location. It would be nicer to have automated testing happening as part of the build, of course. If I can learn how to do that, I'll have a go at it. But I don't know where to start right now. Cheers, Martin > > -- Sebastian Rahtz Chief Data Architect University of Oxford IT > Services 13 Banbury Road, Oxford OX2 6NN. Phone +44 1865 283431 > > >

1 0

Spring F2F Poll
by Hugh Cayless 02 Feb '15

02 Feb '15

Hi all, I've set up a poll at http://doodle.com/2qzywq8c379atu87 in the window April 13–May 15. Let's see what we can do, and I'll adjust as needed.

7 11

re corresp oddity
by Lou Burnard 01 Feb '15

01 Feb '15

I note the following things in the definition for <correspAction> in the corresp ODD 1) you should not use @mode="add" together with @module on an <elementSpec> -- it's meaningless, since you are not allowed to add new things to existing TEI modules 2) if you want your valList to appear in the schema you must put @type="closed" on it. otherwise your suggested values appear in the schema only as comments 3) since you are replacing the existing (vacuous) declaration for the possible values of @type, you must put @mode="replace" on your valList With these changes, things work as expected for elements in the TEI namespace. I am still puzzling over why they don't for non-TEI namespaced elements, but have to go out now.

3 2