Amen. Only @ref should be encouraged. -James On 21/07/16 16:40, Magdalena Turska wrote:

As the stated aim of Simple was to 'minimize ambiguity for the encoder' I believe this is an omission and only @ref should be encouraged (or even allowed perhaps).

On 21 July 2016 at 12:36, Lou Burnard wrote:

...

I am still toiling in this vineyard. Here's a new question for you to consider.

The Simple schema makes available both @ref and @key on both <name> and <rs>. In the prose however only <rs key="arbitrarytoken"> is presented. This seems counter to our current dogma, which would prioritize <name ref="#pointer"> wherever possible, I believe.

Since I've now added <particDesc> and hence <person> is available, it would seem better to rework some (or all?) the examples to use that mechanism, e.g. <name ref="#BENM2">Mrs Bennet</name> rather than <rs key="BENM2">

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services, University of Oxford

Well, that would certainly be possible. I note only that there are projects who would like to use Simple who do use @key (Obvil springs to mind) to supply a regularised form of a name, however much we might deprecate it. <name key="Burnard, Louis (b.1946)">Lou</name> Their argument is that that's a nice simple well-understood way of providing a standardised form of a name without all the folderoll of defining and maintaining a separate authority file. No-one seems to have noticed that my question also relates to the relative priorities we should give <rs> and <name> On 21/07/16 18:27, Elisa wrote: major

So, how about making it only @ref with hashtag in value, and leaving @key entirely out of Simple for simplicity's sake? Keeping both and giving examples is what we'd expect in the TEI All, but the multiple possible approaches here in Simple seem like just the sort of thing to stall a new user unnecessarily...

Elisa Typeset by hand on my iPad

...

On Jul 21, 2016, at 11:55 AM, James Cummings wrote:

Amen. Only @ref should be encouraged.

-James

...

On 21/07/16 16:40, Magdalena Turska wrote: As the stated aim of Simple was to 'minimize ambiguity for the encoder' I believe this is an omission and only @ref should be encouraged (or even allowed perhaps).

...

On 21 July 2016 at 12:36, Lou Burnard wrote:

I am still toiling in this vineyard. Here's a new question for you to consider.

The Simple schema makes available both @ref and @key on both <name> and <rs>. In the prose however only <rs key="arbitrarytoken"> is presented. This seems counter to our current dogma, which would prioritize <name ref="#pointer"> wherever possible, I believe.

Since I've now added <particDesc> and hence <person> is available, it would seem better to rework some (or all?) the examples to use that mechanism, e.g. <name ref="#BENM2">Mrs Bennet</name> rather than <rs key="BENM2">

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services, University of Oxford -- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

If Obvil is one of them, which I think it is, then no you can't. On 21/07/16 18:37, Magdalena Turska wrote:

I'd be in favor of doing as Elisa suggests, only thing is to check if we can still accommodate our reference corpora after ruling out @key altogether...

On 21 July 2016 at 17:27, Elisa wrote:

...

So, how about making it only @ref with hashtag in value, and leaving @key entirely out of Simple for simplicity's sake? Keeping both and giving examples is what we'd expect in the TEI All, but the multiple possible approaches here in Simple seem like just the sort of thing to stall a new user unnecessarily...

Elisa Typeset by hand on my iPad

...

On Jul 21, 2016, at 11:55 AM, James Cummings wrote:

Amen. Only @ref should be encouraged.

-James

...

On 21/07/16 16:40, Magdalena Turska wrote: As the stated aim of Simple was to 'minimize ambiguity for the encoder' I believe this is an omission and only @ref should be encouraged (or even allowed perhaps).

...

On 21 July 2016 at 12:36, Lou Burnard wrote: I am still toiling in this vineyard. Here's a new question for you to consider.

The Simple schema makes available both @ref and @key on both <name> and <rs>. In the prose however only <rs key="arbitrarytoken"> is presented. This seems counter to our current dogma, which would prioritize <name ref="#pointer"> wherever possible, I believe.

Since I've now added <particDesc> and hence <person> is available, it would seem better to rework some (or all?) the examples to use that mechanism, e.g. <name ref="#BENM2">Mrs Bennet</name> rather than <rs key="BENM2">

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services, University of Oxford -- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived -- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

Well, by 'accommodating' I meant to be able to losslessly convert from reference corpora into Simple, not to allow anything that is used there. Thus Simple prescribes use of <name> instead of <persName> or <placeName>, even though these elements are obviously used in some places, same goes for @rend. For @key it is conceivable to have it automatically converted into @ref with simultaneously created <person> or <place> records, which should solve the issue? Opinions? On 21 July 2016 at 17:57, Lou Burnard wrote:

Biut in any case, if we applied that principle everywhere, Simple would rapidly become even more complicated. For example, most of the OTA texts use @rend all over the place.

On 21/07/16 18:43, Lou Burnard wrote:

...

If Obvil is one of them, which I think it is, then no you can't.

On 21/07/16 18:37, Magdalena Turska wrote:

...

I'd be in favor of doing as Elisa suggests, only thing is to check if we can still accommodate our reference corpora after ruling out @key altogether...

On 21 July 2016 at 17:27, Elisa wrote:

So, how about making it only @ref with hashtag in value, and leaving @key

...

entirely out of Simple for simplicity's sake? Keeping both and giving examples is what we'd expect in the TEI All, but the multiple possible approaches here in Simple seem like just the sort of thing to stall a new user unnecessarily...

Elisa Typeset by hand on my iPad

On Jul 21, 2016, at 11:55 AM, James Cummings <

...

James.Cummings@it.ox.ac.uk>

wrote:

...

Amen. Only @ref should be encouraged.

-James

On 21/07/16 16:40, Magdalena Turska wrote:

...

As the stated aim of Simple was to 'minimize ambiguity for the encoder'

I

...

believe this is an omission and only @ref should be encouraged (or even

...

allowed perhaps).

On 21 July 2016 at 12:36, Lou Burnard > wrote:

...

I am still toiling in this vineyard. Here's a new question for you to

...

> consider. > > The Simple schema makes available both @ref and @key on both <name> > and > <rs>. In the prose however only <rs key="arbitrarytoken"> is > presented. > This seems counter to our current dogma, which would > prioritize <name ref="#pointer"> wherever possible, I believe. > > Since I've now added <particDesc> and hence <person> is available, it > would seem better to rework some (or all?) the examples to use that > mechanism, e.g. > <name ref="#BENM2">Mrs Bennet</name> rather than <rs key="BENM2"> > > > > > > > -- > tei-council mailing list > tei-council@lists.tei-c.org > http://lists.lists.tei-c.org/mailman/listinfo/tei-council > > PLEASE NOTE: postings to this list are publicly archived >

-- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services,

University of Oxford

...

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

For @key it is easy to convert to something in @ref where previously something might have said key="ABC123" then we just prefix it as a local private URI named after the corpus. so obvil:ABC123 -- there is no creating of person or place records needed. While it _should_ be expanded with a prefixDef, not doing so is no more or less obtuse than just having and undocumented @key. (and if there is a schematron constraint on @ref to only allow it to start with # or http* then, that is what I'd change.) -James On 21/07/16 18:14, Magdalena Turska wrote:

Well, by 'accommodating' I meant to be able to losslessly convert from reference corpora into Simple, not to allow anything that is used there. Thus Simple prescribes use of <name> instead of <persName> or <placeName>, even though these elements are obviously used in some places, same goes for @rend.

For @key it is conceivable to have it automatically converted into @ref with simultaneously created <person> or <place> records, which should solve the issue? Opinions?

On 21 July 2016 at 17:57, Lou Burnard wrote:

...

Biut in any case, if we applied that principle everywhere, Simple would rapidly become even more complicated. For example, most of the OTA texts use @rend all over the place.

On 21/07/16 18:43, Lou Burnard wrote:

...

If Obvil is one of them, which I think it is, then no you can't.

On 21/07/16 18:37, Magdalena Turska wrote:

...

I'd be in favor of doing as Elisa suggests, only thing is to check if we can still accommodate our reference corpora after ruling out @key altogether...

On 21 July 2016 at 17:27, Elisa wrote:

So, how about making it only @ref with hashtag in value, and leaving @key

...

entirely out of Simple for simplicity's sake? Keeping both and giving examples is what we'd expect in the TEI All, but the multiple possible approaches here in Simple seem like just the sort of thing to stall a new user unnecessarily...

Elisa Typeset by hand on my iPad

On Jul 21, 2016, at 11:55 AM, James Cummings <

...

James.Cummings@it.ox.ac.uk>

wrote:

...

Amen. Only @ref should be encouraged.

-James

On 21/07/16 16:40, Magdalena Turska wrote: > As the stated aim of Simple was to 'minimize ambiguity for the > encoder' > I believe this is an omission and only @ref should be encouraged (or even > allowed perhaps). > > > On 21 July 2016 at 12:36, Lou Burnard > wrote: I am still toiling in this vineyard. Here's a new question for you to >> consider. >> >> The Simple schema makes available both @ref and @key on both <name> >> and >> <rs>. In the prose however only <rs key="arbitrarytoken"> is >> presented. >> This seems counter to our current dogma, which would >> prioritize <name ref="#pointer"> wherever possible, I believe. >> >> Since I've now added <particDesc> and hence <person> is available, it >> would seem better to rework some (or all?) the examples to use that >> mechanism, e.g. >> <name ref="#BENM2">Mrs Bennet</name> rather than <rs key="BENM2"> >> >> >> >> >> >> >> -- >> tei-council mailing list >> tei-council@lists.tei-c.org >> http://lists.lists.tei-c.org/mailman/listinfo/tei-council >> >> PLEASE NOTE: postings to this list are publicly archived >> -- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services,

University of Oxford

...

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services, University of Oxford

Well that is what we did with @rend if you had rend="b" rend="bold" rend="Bold" rend="boldtext" or whatever, they all got changed to rendition="simple:bold". This seems quite reasonable to me. Happy to comment on any such list you want to produce.... -James On 21/07/16 18:28, Lou Burnard wrote:

Ah, well if *that*s your definition of "accomodating", then life suddenly got a whole lot simpler. Since, as any fule kno, you can convert anything into anything using XSLT...

But I think we might want to produce a document listing the 99 most frequent problems you will find...

On 21/07/16 19:14, Magdalena Turska wrote:

...

Well, by 'accommodating' I meant to be able to losslessly convert from reference corpora into Simple, not to allow anything that is used there. Thus Simple prescribes use of <name> instead of <persName> or <placeName>, even though these elements are obviously used in some places, same goes for @rend.

For @key it is conceivable to have it automatically converted into @ref with simultaneously created <person> or <place> records, which should solve the issue? Opinions?

On 21 July 2016 at 17:57, Lou Burnard wrote:

...

Biut in any case, if we applied that principle everywhere, Simple would rapidly become even more complicated. For example, most of the OTA texts use @rend all over the place.

On 21/07/16 18:43, Lou Burnard wrote:

...

If Obvil is one of them, which I think it is, then no you can't.

On 21/07/16 18:37, Magdalena Turska wrote:

...

I'd be in favor of doing as Elisa suggests, only thing is to check if we can still accommodate our reference corpora after ruling out @key altogether...

On 21 July 2016 at 17:27, Elisa wrote:

So, how about making it only @ref with hashtag in value, and leaving @key

...

entirely out of Simple for simplicity's sake? Keeping both and giving examples is what we'd expect in the TEI All, but the multiple possible approaches here in Simple seem like just the sort of thing to stall a new user unnecessarily...

Elisa Typeset by hand on my iPad

On Jul 21, 2016, at 11:55 AM, James Cummings < > James.Cummings@it.ox.ac.uk> > wrote:

> Amen. Only @ref should be encouraged. > > -James > > > On 21/07/16 16:40, Magdalena Turska wrote: >> As the stated aim of Simple was to 'minimize ambiguity >> for the >> encoder' >> > I > believe this is an omission and only @ref should be > encouraged (or even >> allowed perhaps). >> >> >> On 21 July 2016 at 12:36, Lou Burnard >> >> wrote: > I am still toiling in this vineyard. Here's a new > question for you to >>> consider. >>> >>> The Simple schema makes available both @ref and @key on >>> both <name> >>> and >>> <rs>. In the prose however only <rs >>> key="arbitrarytoken"> is >>> presented. >>> This seems counter to our current dogma, which would >>> prioritize <name ref="#pointer"> wherever possible, I >>> believe. >>> >>> Since I've now added <particDesc> and hence <person> is >>> available, it >>> would seem better to rework some (or all?) the examples >>> to use that >>> mechanism, e.g. >>> <name ref="#BENM2">Mrs Bennet</name> rather than <rs >>> key="BENM2"> >>> >>> >>> >>> >>> >>> >>> -- >>> tei-council mailing list >>> tei-council@lists.tei-c.org >>> http://lists.lists.tei-c.org/mailman/listinfo/tei-council >>> >>> PLEASE NOTE: postings to this list are publicly archived >>> > -- > Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT > Services, > University of Oxford

> -- > tei-council mailing list > tei-council@lists.tei-c.org > http://lists.lists.tei-c.org/mailman/listinfo/tei-council > > PLEASE NOTE: postings to this list are publicly archived > -- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- tei-council mailing list tei-council@lists.tei-c.org http://lists.lists.tei-c.org/mailman/listinfo/tei-council

PLEASE NOTE: postings to this list are publicly archived

-- Dr James Cummings, James.Cummings@it.ox.ac.uk Academic IT Services, University of Oxford