I agree. They should all indicate beginnings. (If I have a multi-page document, I will begin the first page with a <pb/> (in order to give it a page numbers) as well as all subsequent pages.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The Guidelines for <lb> should give guidance on whether, when, and how <lb> should be included in texts marked with <p>. Some guidance on <lb> vs. <l> would also be helpful.
Guidelines should make clear whether all of these practices are condoned.
Frederic's:
<p>
<lb/>mais le principal est de l’appliquer bien. Les plus grandes âmes
<lb n="5"/>sont capables des plus grands vices aussi bien que des plus
<lb/>grandes vertus ; et ceux qui ne marchent que fort lentement
<lb/>peuvent avancer beaucoup davantage, s’ils suivent toujours le
<lb/>droit chemin, que ne font ceux qui courent et qui s’en éloignent.
</p>
Oxford Text Archive (as also eMUNCH and others):
<p>
THE scurvy, though a disease very common to the inhabitants of this island, have occa
<lb rend="hidden" type="hyphenInWord"/>
sioned it to be confounded with almost every chronic disease, that I shall recommend my readers to their observations, and at pre
<lb rend="hidden" type="hyphenInWord"/>
sent dedicate this essay to the service and attention of those who seem infatuated with the belief that sea-water is their only remedy.
</p>
Livingstone:
<lb n="19"/><p>He did wrong in that. He loses
<lb n="20"/>all his beads and many friends in
<lb n="21"/>consequence.</p>
<milestone unit="section" rend="line"/>
<lb n="22"/><p>An old man called Kabobo
<lb n="23"/>came for his old wife. I asked her
<lb n="24"/>if this were her husband she went
And something I haven't seen, paragraph breaks between line breaks:
<lb n="19"/>He did wrong in that.</p><p>He loses
<lb n="20"/>all his beads and many friends in
<lb n="21"/>consequence.</p>
<milestone unit="section" rend="line"/>
<lb n="22"/><p>An old man called Kabobo
<lb n="23"/>came for his old wife. I asked her
<lb n="24"/>if this were her husband she went
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I would say that an initial linebreak is assumed or understood in the cases of <p>, <lg> and <l>. Perhaps there should be a setting in the header that makes this explicit, or perhaps we just need to identify all cases where there's a similar assumed relationship between milestone elements and block-level elements. There are lots of such cases -- <div>, <head> and many others.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
> I would say that an initial linebreak is assumed or understood
> in the cases of <p>, <lg> and <l>.
Wouldn't that generate three linebreaks here
<p>
<lb n="4"/>mais le principal est de l’appliquer bien. Les plus grandes âmes
<lb n="5"/>sont capables des plus grands vices aussi bien que des plus
instead of two?
I suspect no proposal that says "These elements presume an initial linebreak" will work. That suggests: "When you see a <p> treat it as if it include a <lb> within it. Then if you see another <lb> that one is additional."
I think the right model is like HTML's block model, where multiple blocks nest by overlap. In that model the <lb n="4"> above merely falls on top of the <p> and the same number of lines would have been encoded had <lb="4"> not been included.
I think the right language will be based on "collapsing" breaks. This
<gb/><pb n=1"/><cb n="A"/><lb/><p>said the man
encodes one break.
I might propose, "All contiguous, unrepeated breaks, and openings of blocks, collapse." But I'd need to sketch out several possibilities to be sure that is right. I'm unsure, for example, of </lb><p><lb/>
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
The inconsistency is what I said was a bug, and that is easily fixed. Before doing so however, we should decide on whether or not the idea of "collapsing" multiple breaks is a good one.
What does <gb/><pb/><lb/> mean? a gathering cannot start without ipso facto starting a page, and a page cannot start without ipso facto starting a line, so the presence of <pb/> and <lb/> in the sequence is arguably redundant. However, if the three levels are being explicitly recorded separately via their @n attributes, the tags are also clearly useful. It seems reasonably to say therefore that no *additional* whitespace should be generated.
The interaction with <p> <div> etc seems different, because these things are not so clearly linked directly to line or page starts. Yes, usually a div starts a new page, but not necessarily. <div> and <p>represent structural components, not layout.
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
At http://sourceforge.net/p/tei/code/11794
I've made the descs for pb and cb consistent with those for lb and gb, but am leaving ticket open to see if there are other changes needed
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
I agree. They should all indicate beginnings. (If I have a multi-page document, I will begin the first page with a <pb/> (in order to give it a page numbers) as well as all subsequent pages.
The Guidelines for <lb> should give guidance on whether, when, and how <lb> should be included in texts marked with <p>. Some guidance on <lb> vs. <l> would also be helpful.
Guidelines should make clear whether all of these practices are condoned.
Frederic's:
<p>
<lb/>mais le principal est de l’appliquer bien. Les plus grandes âmes
<lb n="5"/>sont capables des plus grands vices aussi bien que des plus
<lb/>grandes vertus ; et ceux qui ne marchent que fort lentement
<lb/>peuvent avancer beaucoup davantage, s’ils suivent toujours le
<lb/>droit chemin, que ne font ceux qui courent et qui s’en éloignent.
</p>
Oxford Text Archive (as also eMUNCH and others):
<p>
THE scurvy, though a disease very common to the inhabitants of this island, have occa
<lb rend="hidden" type="hyphenInWord"/>
sioned it to be confounded with almost every chronic disease, that I shall recommend my readers to their observations, and at pre
<lb rend="hidden" type="hyphenInWord"/>
sent dedicate this essay to the service and attention of those who seem infatuated with the belief that sea-water is their only remedy.
</p>
Livingstone:
<lb n="19"/><p>He did wrong in that. He loses
<lb n="20"/>all his beads and many friends in
<lb n="21"/>consequence.</p>
<milestone unit="section" rend="line"/>
<lb n="22"/><p>An old man called Kabobo
<lb n="23"/>came for his old wife. I asked her
<lb n="24"/>if this were her husband she went
And something I haven't seen, paragraph breaks between line breaks:
<lb n="19"/>He did wrong in that.</p><p>He loses
<lb n="20"/>all his beads and many friends in
<lb n="21"/>consequence.</p>
<milestone unit="section" rend="line"/>
<lb n="22"/><p>An old man called Kabobo
<lb n="23"/>came for his old wife. I asked her
<lb n="24"/>if this were her husband she went
I would say that an initial linebreak is assumed or understood in the cases of <p>, <lg> and <l>. Perhaps there should be a setting in the header that makes this explicit, or perhaps we just need to identify all cases where there's a similar assumed relationship between milestone elements and block-level elements. There are lots of such cases -- <div>, <head> and many others.
> I would say that an initial linebreak is assumed or understood
> in the cases of <p>, <lg> and <l>.
Wouldn't that generate three linebreaks here
<p>
<lb n="4"/>mais le principal est de l’appliquer bien. Les plus grandes âmes
<lb n="5"/>sont capables des plus grands vices aussi bien que des plus
instead of two?
I suspect no proposal that says "These elements presume an initial linebreak" will work. That suggests: "When you see a <p> treat it as if it include a <lb> within it. Then if you see another <lb> that one is additional."
I think the right model is like HTML's block model, where multiple blocks nest by overlap. In that model the <lb n="4"> above merely falls on top of the <p> and the same number of lines would have been encoded had <lb="4"> not been included.
I think the right language will be based on "collapsing" breaks. This
<gb/><pb n=1"/><cb n="A"/><lb/><p>said the man
encodes one break.
I might propose, "All contiguous, unrepeated breaks, and openings of blocks, collapse." But I'd need to sketch out several possibilities to be sure that is right. I'm unsure, for example, of </lb><p><lb/>
The inconsistency is what I said was a bug, and that is easily fixed. Before doing so however, we should decide on whether or not the idea of "collapsing" multiple breaks is a good one.
What does <gb/><pb/><lb/> mean? a gathering cannot start without ipso facto starting a page, and a page cannot start without ipso facto starting a line, so the presence of <pb/> and <lb/> in the sequence is arguably redundant. However, if the three levels are being explicitly recorded separately via their @n attributes, the tags are also clearly useful. It seems reasonably to say therefore that no *additional* whitespace should be generated.
The interaction with <p> <div> etc seems different, because these things are not so clearly linked directly to line or page starts. Yes, usually a div starts a new page, but not necessarily. <div> and <p>represent structural components, not layout.
At http://sourceforge.net/p/tei/code/11794
I've made the descs for pb and cb consistent with those for lb and gb, but am leaving ticket open to see if there are other changes needed
Have any other changes that are needed turned up? If not, we should close this ticket.
Closing as per council decision 2013-11 face-to-face