May 2015 - Tei-council - lists.tei-c.org

ODD question
by Hugh Cayless 22 Jul '15

22 Jul '15

Hello all, Hope your trips home went well (or are going well). I’m having a hard time defining a new spec for <secl>. Does anybody know why https://github.com/hcayless/TEI-Guidelines/blob/secl/P5/Source/Specs/secl.x… <https://github.com/hcayless/TEI-Guidelines/blob/secl/P5/Source/Specs/secl.x…> wouldn’t be picked up by any of the build processes I’m running? It’s as though the file is invisible. There’s clearly some incantation I’m not including, but I don’t know what it is… Thanks, Hugh /** * Hugh A. Cayless, Ph.D * hugh.cayless(a)duke.edu * Duke Collaboratory for Classics Computing (DC3) * http://blogs.library.duke.edu/dcthree/ **/

9 36

New Jenkins build jobs for bleeding-edge and stable plugin
by Martin Holmes 04 Jun '15

04 Jun '15

Hi all, I have what looks like a working setup on my Jenkins box for building two versions of the Oxygen TEI plugin. The original build job, "oxygen-tei", to which some of you may have subscribed, is now replaced by "oxygen-tei-bleeding", to which you can subscribe here: <http://teijenkins.hcmc.uvic.ca/job/oxygen-tei-bleeding/lastSuccessfulBuild/…> while there is also now "oxygen-tei-stable", to which if you want you can subscribe here: <http://teijenkins.hcmc.uvic.ca/job/oxygen-tei-stable/lastSuccessfulBuild/ar…> The idea is that oxygen-tei-bleeding builds with the latest plugin code against the latest trunk builds of P5 and the Stylesheets, enabling us to see if any change to any of these has broken the plugin; oxygen-tei-stable builds with the latest plugin code, but against the current release versions of P5 and the Stylesheets, enabling us to see what a release version of the plugin would look like if either we or Syncro built one right now. The build is now being done with ant, and in addition to these two targets, there's also: ant syncro which builds a version for the Syncro folks using the latest stable P5 and Stylesheets, as they would do when creating a build for a new Oxygen release. Up to now they've been using our original update-and-upload.sh script, and then preventing it from uploading to SourceForge as it tries to do. That script also requires that you _know_ the correct versions of P5 and the Stylesheets, and supply them correctly on the command line; the new ant build script works that information out for itself. Finally, there is: ant stable which is what our release technician would use during a P5 release process, or what we would use to release an official build of our version of the plugin. This builds in the same way as the version above, but also offers to upload the result to SourceForge, and creates a new version of the updateSite.oxygen file which it offers to upload to tei-c.org. This should make releasing a new plugin version much easier, I hope (although it can only truly be tested for real when we come to release a version, so I'd like to be the one who runs it for real the first time in case something goes wrong). I've added the new job configs for Jenkins to SVN, where James can find them for the Oxford Jenkins, when he has the time. This is all brand-new and only gently tested, so let me know if you try anything, and please report any problems with as much detail as you can. Cheers, Martin

3 4

Re: [tei-council] [tei:feature-requests] #546 <defaultVal> should be removed from all specs
by Syd Bauman 03 Jun '15

03 Jun '15

Martin -- We'll talk tomorrow, I'm sure, but I'm not sure this is OK. I know you hate default values, as do I, and we want to get rid of them. But I'm not sure it's acceptable to yank them out w/o warning. DTD users might be upset. Besides, changing the semantics of a value like that is the sort of thing users need warning about. Those who actually have part="N" explicitly expressed may well be using it to say "I can't tell if this is a fragment or not". MH> att.fragmentable's @part now has no defaultVal, and the "N" value MH> now actually means "no", as it should. Rev 13232.

2 6

dataSpec content
by Lou Burnard 01 Jun '15

01 Jun '15

I'm still poking away at my datatype replacement ticket ... We agreed that <dataSpec> should be used to fully specify a TEI-defined datatype and also noted in passing that its existing <content> child needed additional constraints (e.g, it shouldnt allow you to supply elementRefs) In fact it seems to me that the "content" of a dataSpoe should be much more restrictive: I'd like for it to contain just <alternate>, <sequence>, <valList>, or <dataRef>, with the additional constraint that <alternate> and <sequence>s it contains can contain only one or more <dataRef>s If no-one has any counterproposal that's what i am planning to try to implement anyway.

2 1

Fast-track a simple ticket
by Martin Holmes 31 May '15

31 May '15

Hi all, This ticket: <https://sourceforge.net/p/tei/feature-requests/561/> arose out of our re-working yesterday of Laurent's standoff proposal. I'd like to fast-track it if we can; it seems straightforward to me, and doesn't involve any Birnbaum issues. While I get around to figuring out the actual change to the Spec file, could you all check to see if you have any objections to it in principle? Cheers, Martin

6 7

rng:group but no rng:choice
by Peter Stadler 31 May '15

31 May '15

I just stumbled across the content model of climate (and there are more …) which looks like <content> <rng:group> <rng:zeroOrMore> <rng:ref name="model.headLike"/> </rng:zeroOrMore> <rng:choice> <rng:group> <rng:oneOrMore> <rng:ref name="model.pLike"/> </rng:oneOrMore> </rng:group> <rng:group> <rng:oneOrMore> <rng:ref name="model.labelLike"/> </rng:oneOrMore> </rng:group> </rng:choice> <rng:zeroOrMore> <rng:choice> <rng:ref name="model.noteLike"/> <rng:ref name="model.biblLike"/> </rng:choice> </rng:zeroOrMore> <rng:zeroOrMore> <rng:ref name="climate"/> </rng:zeroOrMore> </rng:group> </content> My question is: is there some magic to the surrounding rng:group or could that be deleted? (I don’t intend to do so but was just wondering …) Best Peter

2 2

rework of standoff proposal
by Peter Stadler 31 May '15

31 May '15

Dear all, a Council working group (PFS, LB, MH, FC, SM, PWS) just created an alternative proposal as the "Ann Arbor" branch at https://github.com/laurentromary/stdfSpec/tree/AnnArbor. The changes in detail: * killed mapStruct * renamed annotationGrp —> listAnnotation * renamed model.annotationPart —> model.annotation * got rid of dash in module name * removed class att.confidence in favor of https://sourceforge.net/p/tei/feature-requests/561/ * added some more elements to model.annotation * removed standOffHeader in favor of teiHeader(!!) We did not update the examples yet. I’m curious about Laurent’s reaction and will probably have a conf call in the next week with him (— Laurent?!) Best Peter

5 12

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

Next defaultVal to be nuked
by Martin Holmes 30 May '15

30 May '15

I hereby nominate att.divLike/@org's defaultVal: <http://www.tei-c.org/release/doc/tei-p5-doc/en/html/ref-att.divLike.html> This has its own novel brand of silliness: There are two values, and the first value, "composite", means "no claim is made about the sequence in which the immediate contents of this division are to be processed, or their inter-relationships." Yet THIS IS NOT THE DEFAULT VALUE; the _other_ one, "uniform", which _does_ make a claim, is the default. Well FFS. It shall die unless I hear otherwise in the next few hours. cheers, Martin

3 5

eXist-db password changed
by Peter Stadler 30 May '15

30 May '15

Sid argued strongly that we should change the admin password for the eXist-db that serves Roma (the issue arose around the discussion of tickets [feature-requests:#527] and [bugs:#747]). Hence, I duly changed the password and documented it as described in [1] and briefly checked that Roma still works. If you encounter any side effects, please let me know! Best Peter [1] http://www.tei-c.org/Activities/Council/Working/tcw22.xml

2 1

Reconvene at 14:30
by Hugh Cayless 29 May '15

29 May '15

Sent from my phone.

1 0

First defaultVal for the chopping block
by Martin Holmes 29 May '15

29 May '15

My first candidate for an annoying and pointless defaultVal is @part. This has the defaultVal "n", which means: (no) either the element is not fragmented, or no claim is made as to its completeness. [Default] If it means "No claim is made as to its completeness", then there's no point whatsoever -- and actually the fact that this means _both_ "not fragmented" _and_ "I'm not saying whether this is fragmented or not" is just silly. Unless anyone gives me a good reason within the next 24 hours, I'm going to delete the defaultness and remove the latter part of the definition. I don't believe anyone has ever explicitly applied this attribute with the value "no" to mean the second part of its definition, so I don't see that any difference could be made to anyone's existing projects. Shout if you think I'm missing something. Cheers, Martin

1 0

Minutes Doc
by Hugh Cayless 28 May '15

28 May '15

https://docs.google.com/document/d/1V7RB54nYMKeXPqPAGxjbKmXwViz0vLFfwzGo4Ql… <https://docs.google.com/document/d/1V7RB54nYMKeXPqPAGxjbKmXwViz0vLFfwzGo4Ql…> Stefanie, ping me on skype when you’re able to join in. /** * Hugh A. Cayless, Ph.D * hugh.cayless(a)duke.edu * Duke Collaboratory for Classics Computing (DC3) * http://blogs.library.duke.edu/dcthree/ **/

3 4

DTW-Ann Arbor ride
by Lou Burnard 28 May '15

28 May '15

Anyone else arriving about 18h15 (or as the Americans say "6.30 pm") in DTW next Weds? Interested in sharing a cab rather than waiting for the 19h45 bus (or bemoaning the departure of the 18h02 one)?

12 46

Stand-off proposal
by Martin Holmes 27 May '15

27 May '15

Very minor niggle on the stand-off spec: <https://github.com/laurentromary/stdfSpec/blob/master/Specification/standof…> If the intention is to create a new module in TEI rather than add the new elements to an existing module, the module ident should be "standoff" rather than "stand-off", to align with existing modules ("namesdates", "textstructure"). Cheers, Martin

1 1

TEI Simple: Executive summary
by James Cummings 27 May '15

27 May '15

Dear Council, Please find attached an executive summary of the work of TEI Simple project and its proposal to Council. (The proposal really being the proto-ODDs on the github website.) Best wishes, James -- Dr James Cummings, James.Cummings(a)it.ox.ac.uk Academic IT Services, University of Oxford

1 0

Agenda for F2F
by Hugh Cayless 27 May '15

27 May '15

Hi all, I’m putting together a draft agenda at http://wiki.tei-c.org/index.php/Council_Agenda_2015-05 <http://wiki.tei-c.org/index.php/Council_Agenda_2015-05>. Items for discussion are to be found in the Minutes Google Doc at https://docs.google.com/document/d/1V7RB54nYMKeXPqPAGxjbKmXwViz0vLFfwzGo4Ql… Please add any additional business items you have to that list. I’ll be working more on it and the agenda today and tomorrow. Martin had proposed talking about how to build the Oxygen plugin. Do we have any other possible technical talks? Anything for TEI Simple or Pure ODD, for example? Looking forward to seeing you on Thursday! Hugh

2 1

UVic Jenkins temporarily down
by Martin Holmes 27 May '15

27 May '15

Just for info: the UVic Jenkins machine is currently down because it's being cloned to create a new Jenkins build server for other projects. Should be up again soon. Cheers, Martin

1 0

Release of the Oxygen plugin
by Martin Holmes 26 May '15

26 May '15

Hi all, After some more testing with Ron, I've just released a new version of the Oxygen plugin, incorporating the latest Stylesheets release (7.35.0) and some changes to the plugin codebase that we had had to back out of before Oxygen 17 was released because they depended on Stylesheets changes which hadn't been released at that point. If you're subscribed to this version of the plugin, please let me know if you see any problems. One of the confusing issues I'll talk about when I do my little show-and-tell on the Oxygen plugin in Michigan will be version numbers. Right now, the plugin filename is derived from the P5 and Stylesheets versions incorporated in it: teioxygen-2.8.0-7.35.0.zip = P5 2.8.0 and Stylesheets 7.35.0. However, there also has to be a true version number for the plugin, which is inserted into the updateSite.oxygen XML file that handles the plugin updates for users. Oxygen's system only allows a version number consisting of three dotted digits: 1.2.3 Our practice up to now has been to name the plugin release after the P5 release on which it's based, so this would be: 2.8.0 but because the last one was already 2.8.0, I've had to increment it: 2.8.1 If we now go ahead and release a minor P5 update, which would be 2.8.1, we would confusingly have to increment the plugin version number to 2.8.2. So I'd like to suggest that we detach the plugin versioning count entirely from the P5 version, and simply increment it independently; but with the second and third digits representing collapsed versions of the P5 and Stylesheets versions on which it's based: 3.280.7350 This would mean that we could have meaningful plugin version numbers simply incrementing in position 1, and still record the versions of P5 and the Stylesheets on which they're based. Any see any problem with this? We could make the change at the same time that we switch the file naming from teioxygen to oxygen-tei, as we've discussed before. Cheers, Martin

7 14

Building (e.g. TEI All schema)
by Hugh Cayless 26 May '15

26 May '15

Can someone explain to me like I’m an idiot how to use the Stylesheets to generate, say, a TEI All schema using the Guidelines specs? I’m afraid the documentation doesn’t give me enough to go on. I thought I understood it, but I don’t. My understanding is that you use the transformtei script, but what inputs do you give it? Thanks, Hugh

6 12

Re: [tei-council] ROMA does not work
by Martin Holmes 26 May '15

26 May '15

Once the build process has gone through on Jenkins successfully, I'll test my new oxygen-tei plugin build script by doing "ant release" to see what happens. :-) Cheers, Martin On 15-05-26 09:09 AM, Sebastian Rahtz wrote: > Roma should be working again now. I made a new Stylesheets release and deployed it around. > > Many apologies for that. > > Sebastian Rahtz > Chief Data Architect > University of Oxford IT Services > +44 1865 283431 >

1 0

bad date example
by Syd Bauman 26 May '15

26 May '15

We have a single example (used twice with slight differences) that shows use of dating attributes in a way I think may be inappropriate: <residence from="1857-03-01" notAfter="1857-04-30">From the 1st of March to some time in April of 1857.</residence> I don't like the combination of @from and @notAfter. As I understand (or at least imagine) it: @from and @to: indicate a range of dates during which something happened -- the thing happens during the entire range. <persName from="1940-10-09" to="1969-04-21">Lennon, John Winston</> <persName from="1969-04-22" to="1980-12-08">Lennon, John Winston Ono</> or <occupation from="1991" to="1994">Provost, Duke University</> For the *entire period* from 09 Oct 40 to 21 Apr 69 his name was "John Winston Lennon", and for the *entire period* from the day he legally changed his name (22 Apr 69) to his death (08 Dec 80) his name was "John Winston Ono Lennon". Similarly, Tommy Langford was Provost from 1991 to 1994. We don't know what day he became Provost or what day he stepped down, so the values of @from and @to are given only to year precision. Fair enough. (If we knew more, we could provide more precision.) @notBefore and @notAfter: indicate the earliest and latest endpoints for a single event to have occurred. Typically, but not at all necessarily, that event is measured in the precision of a single day. E.g. <death notBefore="1831" notAfter="1840"/> indicates that Mary Clarke [nee Carr] died sometime between those dates. She died on one day[1] during those days, she did not spend all 9 years in the throes of death. I think these notions of @from and @to indicating a *duration* during which something happened continuously, and @notBefore and @notAfter indicating endpoints within which range something happened at a given point, is borne out by the definitions in the tagdocs: - notBefore: specifies the earliest possible date for the event - notAfter: specifies the latest possible date for the event - from: indicates the starting point of the period - to: indicates the ending point of the period Thus my unease with the example <residence from="1857-03-01" notAfter="1857-04-30">From the 1st of March to some time in April of 1857.</residence> To me, the @from indicates that she lived there for a 1-2 month long period starting 01 March; but the @notAfter (at least, taken alone) implies that she lived there for one day[2] some day before 01 May. So I'm sure I don't like this, although I'm not sure how to solve the problem, i.e. how this information should be indicated. Possibilities include: * State explicitly in the prose what @from & @notAfter or @notBefore & @to mean. (To wit, an entire duration of vague length, specified by one endpoint and a point over which the duration did not extend.) * Indicate the duration with a start point and a duration, then indicate the (lack of) precision of the duration: <residence xml:id="R" from="1857-03-01" dur="P46D"> From the 1st of March to some time in April of 1857. </residence> <precision target="#R" match="@dur" atLeast="31" atMost="61"/> Don't like this as it sits, because it only works because we presume the unit @atLeast & @atMost refer to is days. Makes sense for this case, but if the dur= were "P1M15D" it would be less obvious. I presume if it were "P1.5M", one would have to use atLeast=1 atMost=2; i.e. the precision attrs apply to the lowest order component. So I guess we could make this work by making that the rule. * Fix <precision> so that it allows representation of precision (meaning the unit or number of significant digits to which a value may be measured reliably) rather than just precision (meaning vaguely how exact or accurate). Then <residence xml:id="R" from="1857-03-01" to="1857-04-15"> From the 1st of March to some time in April of 1857. </residence> <precision target="#R" match="@to" dur="P15D"/> or whatever. * Create new attributes (egads!) to mean exactly what we want. E.g., @fromAfter and @toBefore. * Don't fix it. Caveat encoder. People tend to know what you mean. Besides, what possible automated processing will handle this stuff? Notes ----- [1] Actually one second, but let's not go nuts. [2] Or only one second, but let's not go nuts.

5 8

agenda item / bug
by Syd Bauman 26 May '15

26 May '15

Julia just reminded me that <event> still is not allowed as a child or grandchild of <org>. Off the top of my head, I don't see why <listEvent> is not a member of model.orgPart. Not quite so obvious how to make <event> an eligible direct child of <org>, but it seems like a reasonable request. I plan to submit a bug report shortly, with hopes that we can tackle this and solve it for next release. (Although I have my suspicions that this will bring up other thorny issues.)

3 3

Transcription of spoken language draft
by Martin Holmes 21 May '15

21 May '15

Hi all, I was just following up with Thomas Schmidt on this ticket: <http://sourceforge.net/p/tei/feature-requests/511/> which is one of at least two relating to the ISO/TEI spoken language proposal, and he's sent me a draft dated April 27, which is later than the last one mentioned on Lou's ticket: <http://sourceforge.net/p/tei/feature-requests/539/> He's still not happy with this being posted publicly, so we need a method of distributing it which is not public. The simplest thing would perhaps be to wait until the FtF and then just email it around. Does that work, or should be devise a password-protected location for it? Cheers, Martin

3 4

duplicate @rend error in generated RNG schema
by James Cummings 20 May '15

20 May '15

I was just bitten by the side effect of moving @rend to att.global.rendition. In taking an old and perfectly fine TEI ODD, and generating a fresh schema from it, the RNG was invalid because of a duplicate definition of @rend. Changing it so that I modified @rend on att.global.rendition rather than att.global solved this. Took me a good 10 minutes staring at it to realise what was wrong. doh. I'm just mentioning it to put it in the back of your brain in case anyone asks in the future. -James -- Dr James Cummings, James.Cummings(a)it.ox.ac.uk Academic IT Services, University of Oxford

3 4