Quantcast

use cases for <space>

classic Classic list List threaded Threaded
8 messages Options
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

use cases for <space>

Martin Mueller

The sources of TCP texts include forms that deliberately leave a space to be filled in by somebody,  as in A83003 (Wing E1791), where the phrase “Aſſeſſors and Collectors of” is followed by a space to let somebody write down what they were assessing or collecting.  In this particular TCP transcription the space is represented by a <gap> element. That doesn’t seem right.  There is no lacuna there; it is, in Wallace Stevens’ lovely phrase a “nothing that is.”

 

Would <space> be the appropriate element for representing this?  It’s from the transcription module, which deals with the physical aspects of documentary edition, which is not what the TCP texts are about. And the description of the <space> element in the Guidelines does not include this, although I’d think of this as a very obvious use case that it might good to have mentioned there, if <space> is thought of as the right solution for this sort of thing.

 

And what about a name like “B     “, not uncommon in satirical texts?

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

Christian Thomas

Dear Martin, maybe that older discussion is helpful for you:
https://listserv.brown.edu/archives/cgi-bin/wa?A2=tei-l;832300b1.1502

Esp. the second example
[...] use [<metamark> and] @function="placeholder" here, as the space indeed is the mark or sign indicating the insecurity and it was left as a kind of place holder to be filled in later:

Baco von Verulam in seinem Werke: 
<metamark function="placeholder"><space dim="horizontal" unit="words" quantity="1"/></metamark>.

Best wishes
Christian




Am 13.02.2017 um 16:10 schrieb Martin Mueller:

The sources of TCP texts include forms that deliberately leave a space to be filled in by somebody,  as in A83003 (Wing E1791), where the phrase “Aſſeſſors and Collectors of” is followed by a space to let somebody write down what they were assessing or collecting.  In this particular TCP transcription the space is represented by a <gap> element. That doesn’t seem right.  There is no lacuna there; it is, in Wallace Stevens’ lovely phrase a “nothing that is.”

 

Would <space> be the appropriate element for representing this?  It’s from the transcription module, which deals with the physical aspects of documentary edition, which is not what the TCP texts are about. And the description of the <space> element in the Guidelines does not include this, although I’d think of this as a very obvious use case that it might good to have mentioned there, if <space> is thought of as the right solution for this sort of thing.

 

And what about a name like “B     “, not uncommon in satirical texts?


Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

Paul Schaffner
In reply to this post by Martin Mueller
I'm off to a meeting, so I won't trouble to defend <gap/>, but
would for the moment point out that there are at least four
situations, some of which should perhaps be distinguished:

1. A book leaves a blank because the information is not
(yet?) available. E.g. in lists of office holders or attendees at
a meeting, if the name is forgotten or similar you'll see

   John Smith
   Bill Bones
   James ________
   Richard Roe

(gaps left for decorative initials to be supplied later might
fall into this category too.)

2. A form or form book leaves blanks intended to be put in
by hand when used

   This __ day of ___ 16__, the ship ______ was examined
   and found to draw ____ feet of water.

3. Polite elision, occasional before 1700, endemic in the 18th
century. A blank intended to be filled in *mentally* but not
usually physically.

   Mr. B***** of the town of M****** wrote to me...

4. An example of one of these (especially no.2) which has
in fact been filled in.

Also that the 'blank' itself might be literally blank, or might be
represented by some character or character string, such as
----- _____ ***** or .....  -- which might be worth preserving.
I used to think this info not worth preserving, until I received
queries from three different people trying to find patterns in
the characters used to represent polite elision.

More later (maybe).

pfs


On Mon, Feb 13, 2017, at 10:10, Martin Mueller wrote:

> The sources of TCP texts include forms that deliberately leave a space to
> be filled in by somebody,  as in A83003 (Wing E1791), where the phrase
> “Aſſeſſors and Collectors of” is followed by a space to let somebody
> write down what they were assessing or collecting.  In this particular
> TCP transcription the space is represented by a <gap> element. That
> doesn’t seem right.  There is no lacuna there; it is, in Wallace Stevens’
> lovely phrase a “nothing that is.”
>
> Would <space> be the appropriate element for representing this?  It’s
> from the transcription module, which deals with the physical aspects of
> documentary edition, which is not what the TCP texts are about. And the
> description of the <space> element in the Guidelines does not include
> this, although I’d think of this as a very obvious use case that it might
> good to have mentioned there, if <space> is thought of as the right
> solution for this sort of thing.
>
> And what about a name like “B     “, not uncommon in satirical texts?
--
Paul Schaffner  Digital Library Production Service
[hidden email] | http://www.umich.edu/~pfs/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

Martin Mueller
An excellent taxonomy, and a useful point of departure for some advice  on how to encode stuff that is not (quite) there, but is not a lacuna in the ordinary sense of the word. Michael Oakeshott in On Human Conduct distinguishes between procedures, which always are evidence of intelligence, and processes, which are not. There is a big difference between a wink or blink of an eye. Gaps are blinks rather than winks.

So one might argue that the <gap> element should never be used for absences that result from some choice and absences that are not intentional. Whose intention? A book that has come down to us in a censored version has gaps that the author did not intend. It gets complicated at the edges, but may still be useful in determining whether <gap> should or should not be used.

On 2/13/17, 1:25 PM, "Paul Schaffner" <[hidden email]> wrote:

    I'm off to a meeting, so I won't trouble to defend <gap/>, but
    would for the moment point out that there are at least four
    situations, some of which should perhaps be distinguished:
   
    1. A book leaves a blank because the information is not
    (yet?) available. E.g. in lists of office holders or attendees at
    a meeting, if the name is forgotten or similar you'll see
   
       John Smith
       Bill Bones
       James ________
       Richard Roe
   
    (gaps left for decorative initials to be supplied later might
    fall into this category too.)
   
    2. A form or form book leaves blanks intended to be put in
    by hand when used
   
       This __ day of ___ 16__, the ship ______ was examined
       and found to draw ____ feet of water.
   
    3. Polite elision, occasional before 1700, endemic in the 18th
    century. A blank intended to be filled in *mentally* but not
    usually physically.
   
       Mr. B***** of the town of M****** wrote to me...
   
    4. An example of one of these (especially no.2) which has
    in fact been filled in.
   
    Also that the 'blank' itself might be literally blank, or might be
    represented by some character or character string, such as
    ----- _____ ***** or .....  -- which might be worth preserving.
    I used to think this info not worth preserving, until I received
    queries from three different people trying to find patterns in
    the characters used to represent polite elision.
   
    More later (maybe).
   
    pfs
   
   
    On Mon, Feb 13, 2017, at 10:10, Martin Mueller wrote:
    > The sources of TCP texts include forms that deliberately leave a space to
    > be filled in by somebody,  as in A83003 (Wing E1791), where the phrase
    > “Aſſeſſors and Collectors of” is followed by a space to let somebody
    > write down what they were assessing or collecting.  In this particular
    > TCP transcription the space is represented by a <gap> element. That
    > doesn’t seem right.  There is no lacuna there; it is, in Wallace Stevens’
    > lovely phrase a “nothing that is.”
    >
    > Would <space> be the appropriate element for representing this?  It’s
    > from the transcription module, which deals with the physical aspects of
    > documentary edition, which is not what the TCP texts are about. And the
    > description of the <space> element in the Guidelines does not include
    > this, although I’d think of this as a very obvious use case that it might
    > good to have mentioned there, if <space> is thought of as the right
    > solution for this sort of thing.
    >
    > And what about a name like “B     “, not uncommon in satirical texts?
    --
    Paul Schaffner  Digital Library Production Service
    [hidden email] | https://urldefense.proofpoint.com/v2/url?u=http-3A__www.umich.edu_-7Epfs_&d=CwIFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=rG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=Xjy5Q1JvULQg5EHa_clELImZ4XVfKjgw_cugcVHsx2U&s=MpTdbfsBzFC1RWy2STHoAVOTHgurMwv8nNTkzviAHSI&e= 
   
   
   

Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

C. M. Sperberg-McQueen
In reply to this post by Paul Schaffner
> On Feb 13, 2017, at 12:25 PM, Paul Schaffner <[hidden email]> wrote:
>
> I'm off to a meeting, so I won't trouble to defend <gap/>, but
> would for the moment point out that there are at least four
> situations, some of which should perhaps be distinguished:
>
> 1. A book leaves a blank because the information is not
> (yet?) available. ...
>
> 2. A form or form book leaves blanks intended to be put in
> by hand when used ...
>
> 3. Polite elision, occasional before 1700, endemic in the 18th
> century. A blank intended to be filled in *mentally* but not
> usually physically. ...
>
> 4. An example of one of these (especially no.2) which has
> in fact been filled in.

These are all worth distinguishing, but ‘gap’ is defined as
marking a point in a transcript at which the transcriber has
chosen not to transcribe something present in the exemplar.

Since 1 and 2 both involve meaningful whitespace in the
exemplar, both seem to me to fall within the defined meaning
of ‘space’, and outside the defined meaning of ‘gap’.  

Paul, if you have a way to defend the use of ‘gap’ for these,
I for one would be interested to hear it.  

Since 3 does not involve whitespace in the exemplar but
instead a kind of obfuscation, it doesn’t seem to me to fit
either ‘space’ or ‘gap’.  I don’t know whether there is a
pre-defined element for it.

For 4, my first thought (not necessarily the best thought)
would be to distinguish multiple hands (taking that term
in a general sense as meaning ‘act of inscription’, so that
the printer producing a pre-printed form has a ‘hand’.
Or perhaps it’s as simple as a ‘space’ with an adjacent
(or embedded?) ‘add’?

My two cents.

Michael


********************************************
C. M. Sperberg-McQueen
Black Mesa Technologies LLC
[hidden email]
http://www.blackmesatech.com
********************************************
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

Lou Burnard-6
See  http://tei-l.970651.n3.nabble.com/Telegrams-td2349233.html for some
prior discussion of case (2).

I suppose one might defend the use of <gap> here by arguing that the
document being transcribed is the idealised completed version of the
form, and hence that the blanks are actually intentional gaps. Well, maybe.

The trouble with using <space> is that it doesn't distinguish these and
other possible readings of the white space or "polite elision". I think
there's a good case for a specific element (anyone for <blank/> ?) for
the latter (Paul's case 3), if only because it is so conventionalised.


On 13/02/17 22:04, C. M. Sperberg-McQueen wrote:

>> On Feb 13, 2017, at 12:25 PM, Paul Schaffner <[hidden email]> wrote:
>>
>> I'm off to a meeting, so I won't trouble to defend <gap/>, but
>> would for the moment point out that there are at least four
>> situations, some of which should perhaps be distinguished:
>>
>> 1. A book leaves a blank because the information is not
>> (yet?) available. ...
>>
>> 2. A form or form book leaves blanks intended to be put in
>> by hand when used ...
>>
>> 3. Polite elision, occasional before 1700, endemic in the 18th
>> century. A blank intended to be filled in *mentally* but not
>> usually physically. ...
>>
>> 4. An example of one of these (especially no.2) which has
>> in fact been filled in.
> These are all worth distinguishing, but ‘gap’ is defined as
> marking a point in a transcript at which the transcriber has
> chosen not to transcribe something present in the exemplar.
>
> Since 1 and 2 both involve meaningful whitespace in the
> exemplar, both seem to me to fall within the defined meaning
> of ‘space’, and outside the defined meaning of ‘gap’.
>
> Paul, if you have a way to defend the use of ‘gap’ for these,
> I for one would be interested to hear it.
>
> Since 3 does not involve whitespace in the exemplar but
> instead a kind of obfuscation, it doesn’t seem to me to fit
> either ‘space’ or ‘gap’.  I don’t know whether there is a
> pre-defined element for it.
>
> For 4, my first thought (not necessarily the best thought)
> would be to distinguish multiple hands (taking that term
> in a general sense as meaning ‘act of inscription’, so that
> the printer producing a pre-printed form has a ‘hand’.
> Or perhaps it’s as simple as a ‘space’ with an adjacent
> (or embedded?) ‘add’?
>
> My two cents.
>
> Michael
>
>
> ********************************************
> C. M. Sperberg-McQueen
> Black Mesa Technologies LLC
> [hidden email]
> http://www.blackmesatech.com
> ********************************************
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

Paul Schaffner
In reply to this post by C. M. Sperberg-McQueen
A. The defense of <gap/> is quadripartite:

(1) Expedience.

In our limited tag set of about 60 elements, it was
the only one even vaguely appropriate (we lack <space/>
and at the time lacked even <seg>)

(2) Versatility

As an empty tag with all-purpose @desc and @reason
attributes, alongside the rendition elements, it
could be used for all the likely variants, e.g.

<gap reason="politeElision" rend="*****"/>
<gap reason="formBlank" rend="_____">
<gap reason="tbd" rend="    ">

(3) Practical processing. Since we already had a
decent process for rendering other sorts of gaps
(music, maths, non-latin alphabets) by means of
the character strings stored in the @rend attribute,
it was easy to use the same for this different kind
of gap.

(4) Perverse rationalization. One could perversely
justify using <gap/> to capture *every* sort of textual
feature. Start out by declaring that you intend to
capture this text by capturing none of it, and then
class the various features by means of the @reason
or @desc attributes of <gap/>, like this:

<l>
<gap reason="word" desc="name" rend="Mary"/>
<gap reason="word" desc="verb" rend="had"/>
<gap reason="word" desc="article" rend="a"/>
<gap reason="word" desc="adjective" rend="little"/>
<gap reason="word" desc="noun" rend="lamb"/>
<gap reason="punctuation" desc="full_stop" rend=".">
</l>

B. More legitimate options include:


1. Capture as literals

    Robert _____
    I, _____ _____
    Mr. B***** of M*****
 
2. Do the same, but use <g> mechanism to define special characters

3. Do (1) or (2) but wrap in <seg> or some equivalent noncommittal
element.

    Robert <seg type="leftBlank">_____</seg>
    I, <seg type="formBlank">_____</seg> <seg
    type="formBlank">_____</seg>
    Mr B<seg type="politeElision">*****</seg> of M<seg
    type="politeElision">*****</seg>
 
4. Use <space> and move the literal rendition into one of the @rend-type
attributes e.g.

   Robert <space type="tbd" rend="_____"/>
   I, <space type="fillable" rend="_____"/> <space type="fillable"
   rend="_____"/>
   Mr B<space type="elision" subtype="polite" rend="*****"/> of M<space
   type="elision" subtype="polite" rend="*****"/>
   
When the space is filled in, you could opt to

1. Put the space (however represented) and the addition into del
   and add tags, which effectively says that to fill in a blank
   is implicitly to delete the blankness of it.
   
   I,
     <subst>
       <del><space/></del>
       <add>Paul</add>
     </subst>
     <subst>
        <del><space/></del>
        <add>Schaffner</add>
     </subst>
     
2. Just use <add> and put everything else in attributes

   I, <add type="formFill" place="onBlank">Paul</add>
      <add type="formFill" place="onBlank">Schaffner</add>
     
   This does, however, make it difficult to describe the
   typographic form of the original 'blank'.
   
3. Use <choice> and <seg> rather than <subst> and <del><add>.

pfs  
     
     
--
Paul Schaffner  Digital Library Production Service
[hidden email] | http://www.umich.edu/~pfs/
Reply | Threaded
Open this post in threaded view
|  
Report Content as Inappropriate

Re: use cases for <space>

Martin Mueller
In reply to this post by Lou Burnard-6
On the other hand, if there is an element that says “the space I represent was put there on purpose” that may be good enough for most purposes. The context will help to disambiguate, and if you want to be more precise you could use attributes.

On 2/13/17, 5:21 PM, "TEI (Text Encoding Initiative) public discussion list on behalf of Lou Burnard" <[hidden email] on behalf of [hidden email]> wrote:

    See  https://urldefense.proofpoint.com/v2/url?u=http-3A__tei-2Dl.970651.n3.nabble.com_Telegrams-2Dtd2349233.html&d=CwIFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=rG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=qHb54EaiIR0sNQzy5iyYcjNue0ZH7Fw491LAaHtVko0&s=o0tCj3BSTurHr2oiN1lPypNAxXDdnAn7GkakC7n26as&e=  for some
    prior discussion of case (2).
   
    I suppose one might defend the use of <gap> here by arguing that the
    document being transcribed is the idealised completed version of the
    form, and hence that the blanks are actually intentional gaps. Well, maybe.
   
    The trouble with using <space> is that it doesn't distinguish these and
    other possible readings of the white space or "polite elision". I think
    there's a good case for a specific element (anyone for <blank/> ?) for
    the latter (Paul's case 3), if only because it is so conventionalised.
   
   
    On 13/02/17 22:04, C. M. Sperberg-McQueen wrote:
    >> On Feb 13, 2017, at 12:25 PM, Paul Schaffner <[hidden email]> wrote:
    >>
    >> I'm off to a meeting, so I won't trouble to defend <gap/>, but
    >> would for the moment point out that there are at least four
    >> situations, some of which should perhaps be distinguished:
    >>
    >> 1. A book leaves a blank because the information is not
    >> (yet?) available. ...
    >>
    >> 2. A form or form book leaves blanks intended to be put in
    >> by hand when used ...
    >>
    >> 3. Polite elision, occasional before 1700, endemic in the 18th
    >> century. A blank intended to be filled in *mentally* but not
    >> usually physically. ...
    >>
    >> 4. An example of one of these (especially no.2) which has
    >> in fact been filled in.
    > These are all worth distinguishing, but ‘gap’ is defined as
    > marking a point in a transcript at which the transcriber has
    > chosen not to transcribe something present in the exemplar.
    >
    > Since 1 and 2 both involve meaningful whitespace in the
    > exemplar, both seem to me to fall within the defined meaning
    > of ‘space’, and outside the defined meaning of ‘gap’.
    >
    > Paul, if you have a way to defend the use of ‘gap’ for these,
    > I for one would be interested to hear it.
    >
    > Since 3 does not involve whitespace in the exemplar but
    > instead a kind of obfuscation, it doesn’t seem to me to fit
    > either ‘space’ or ‘gap’.  I don’t know whether there is a
    > pre-defined element for it.
    >
    > For 4, my first thought (not necessarily the best thought)
    > would be to distinguish multiple hands (taking that term
    > in a general sense as meaning ‘act of inscription’, so that
    > the printer producing a pre-printed form has a ‘hand’.
    > Or perhaps it’s as simple as a ‘space’ with an adjacent
    > (or embedded?) ‘add’?
    >
    > My two cents.
    >
    > Michael
    >
    >
    > ********************************************
    > C. M. Sperberg-McQueen
    > Black Mesa Technologies LLC
    > [hidden email]
    > https://urldefense.proofpoint.com/v2/url?u=http-3A__www.blackmesatech.com&d=CwIFaQ&c=yHlS04HhBraes5BQ9ueu5zKhE7rtNXt_d012z2PA6ws&r=rG8zxOdssqSzDRz4x1GLlmLOW60xyVXydxwnJZpkxbk&m=qHb54EaiIR0sNQzy5iyYcjNue0ZH7Fw491LAaHtVko0&s=do_D0JeQ_Jh_bRqR2Jyp4N6BRFwJFQ2Tb5gdvAEU25s&e= 
    > ********************************************
   

Loading...