On 25/05/11 17:39, Paul F. Schaffner wrote:
> On Wed, 25 May 2011, Espen S. Ore wrote:
>> In the P4 version I am converting they created a new element,<printer>  as a
>> parallel to<publisher>  - printer is here ment as a printing firm or
>> printer's shop.
> Speaking in ignorance, as usual...
> Is there any body of practice in the application of TEI to
> early modern bibliography?

I am sure there is quite a bit of established practice out there in the 
wild, but I don't know that it's been fed back into the Guidelines. I 
know we've two or three times tried to set up a workgroup on related 
topics, but they don't seem to have led to updates to the Guidelines, if 
updates are needed.

But are they actually needed? Here's my understanding of the issues 
about marking up these things using currently available TEI elements.

Let's begin by making distinctions though... The following are all, in 
my view, different animals:

a) the printer (etc) when included in a bibliographic record, for 
example inside a <sourceDesc> for a particular exemplar being encoded.

b) the printer (etc) as it appears verbatim at some point on a title 
page or similar

c) a name which (we know) is that of a printer (etc) which might appear 
anywhere (including a priori the preceding two cases)

For the first case, the TEI provides a few special purpose elements, 
mostly corresponding with things that have an established place in 
modern bibliographic records -- <publisher> being the most obvious. It 
doesn't provide <printer> or <bookseller> or <illustrator> or <engraver> 
etc but it *does* provide a generic <respStmt> for them.  So in a 
bibliographic record, e.g. inside a <bibl> or <biblStruct> you can 
certainly say <respStmt><resp>printer</resp><name>Blackberry 
Press</name></respStmt> etc. And you can of course add your own 
<my:printer> element to the model.biblPart class if the generic case 
seems too much of a faff for your immense bibliography of 18th c. 
printing. And you can also lobby for such an element to be added 
explicitly to the Guidelines in due course.

For the second case, the TEI again provides some generic elements to 
mark up the function of particularly, um, functional parts of a title 
page: <docAuthor>, <docEdition>, <docTitle>, <docImprint> and <docDate>. 
The third of these is where one would normally expect to find such 
things as printers being mentioned, of course. Its content model permits 
text, the usual phrase-level elements, and a few hand-picked components 
from the model.biblPart class -- notably <pubPlace> and <publisher> -- 
which would otherwise not be there.

Arguably, this is a mistake -- it might have been cleaner not to permit 
them at all, since they really belong to the world of bibliography, not 
the world of transcribed text, but there they are. If you want to mark 
up  a docImprint which contains a publisher, you are at liberty to do so 
using the explicit <publisher> tag -- but you cannot put in a <respStmt> 

My preference however, given that the things in a docImprint are really 
not parts of a modern bibliographic description, but just bits of text 
that are being associated with some function, would be to use one or 
more of the facilities provided by TEI for naming things. So I would 
rather say <name role="printer"> or even <persName 
<surname>Tottle</surname></persName> etc.

This has the great merit that I can define as many new roles as I find 
necessary for my materials; it also means that I can identify these 
named roles correctly wherever they appear in my text -- and do so 
consistently. I can also use the @ref and @key attributes to resolve 
ambiguous references to the same named individual without a great deal 
of additional retagging.

Hoorah! I'll have a crack at some of Paul's list of excellent examples...

> AT LONDON, Printed by FELIX KYNGSTON for Thomas Man. 1607.

    AT <name type="place" role="pubPlace">LONDON</name>, Printed by 
<name role="printer">FELIX KYNGSTON</name> for <name 
role="bookseller">Thomas Man</name>. <docDate>1607.</docDate></docImprint>

Note that I've inferred here that Mr Man who paid for it to be printed 
was probably a bookseller. Some other value such as "commissioner" or 
even "publisher" might be better.

> LONDON, Printed for IOHN BARTLET, at the gilt Cup in Cheape-side 1628.

<name type="place" role="pubPlace">LONDON</name>, Printed for <name 
role="bookseller">IOHN BARTLET</name>, <name type="place" 
role="pubAddress">at the gilt Cup in Cheape-side</name> 

Same remark as above about the "bookseller". Here I've assumed that "at 
the gilt cup in cheapeside" is effectively the bookseller's address.

> LONDON, Printed by G. D. 1652.

Note that in this one, if I ever found out who G.D. was, I could do 
something like  <name role="printer" 

I leave the rest to the interested reader...

> AT LEYDEN, Printed by VVilliam Christian. 1634.
> IMPRINTED AT LONDON by Henrie Bynneman, dwelling in Knightrider strete,
>    at the signe of the Mermayde.
> London : Printed for A. Millar; W. Strahan; J. Rivington; J. Newbery; R.
>     Baldwin; S. Crowder, and Co.; T. Caslon; B. Law; M. Richardson; and B.
>     Collins. 1767.
> Paris. Printed by the Widow Chrestien. 1670.
> LONDON Printed by M.S. and are to be sold by Livewell Chapman, at the
>     Crowne in Popes head Alley. 1653.
> LONDON, Printed by Ja. Cottrel, for Hen. Fletcher at the three Gilt Cups
>     in St. Pauls Church-yard, 1661.
> LONDON, Septemb. 19. Printed by T.F. for H.H. 1642.
> LONDON, Printed by John Bill, Thomas Newcomb, and in
>     Henry Hills, Printers to the Kings most Excellent Majesty,
>     1679/80.
> LONDON, Printed for T. Parkhurst at the Bible on London-Bridg, and
>     G. Calvert at the Golden-Ball in Duck-Lane.