Gibbons

1092 views
Skip to first unread message

Vince

unread,
May 2, 2020, 2:37:03 PM5/2/20
to Standard Ebooks
While I’m continuing to proof L’Mort de Arthur

I think I may regret this later (maybe even sooner), but I’ve always wanted to read Gibbons’ History of the Decline and Fall of the Roman Empire, so I thought I’d take a look at what it would take to produce it. I’ve already run into a problem with the PG transcription, which they fixed, but they broke something else in the process, and it’s going to be “many days” before they get it resolved. But that’s for them; for us, I have some questions.

1. Although it’s always printed in multiple volumes, it doesn’t appear to be structured in volumes (and the number of volumes is all over the map—four, six, eight, and I’ve seen as many as twelve). The chapters are numbered sequentially from 1 to 71. I think, therefore, that it should just have a single level, i.e. chapter. Does that sound OK?

2. Gibbons included a boatload (technical term) of footnotes. Many of the now PD editions, including the one PG used for transcription, also include notes by others. For example, in the PG edition, there are notes by three people: Gibbons, Milman, and Guizot. This in and of itself is NBD.

However, in some cases, the new notes are notes about Gibbon’s notes. In other words, an endnote has an endnote reference in it. For example, see here; footnote 2 has two note references within it, both towards the end. The notes those references refer to are printed after the note, below a <hr>-style line.
I’m not sure how (or if) our popup notes would handle that. If we tap on note 2, then the popup displays it. If that note in turn has a note reference in it, what happens when we tap on it? Do we get another popup? Does the existing popup change to display the new note? How (or can we) get back to the original note?

In short, how would we want to handle that? I suspect the best way would be to merge the additional notes into the original note, and just use superscripted letter(s) to show the notes within the note. 

(From my short review, all of the notes are worthwhile; I don’t think we want to drop them. And I definitely don’t want to have to wade through them all and figure out what to keep.)

3. There is heavy use of ligatures in the book. Substituting them should mostly be OK, but in a few cases, I think we might want to correct for modern spelling. For example, I’ve seen Æthiopia; I assume we want to replace that with Ethiopia, not Aethiopia, correct? Same for æra; we want that to be era, not aera, correct? Those are the two obvious ones I’ve found so far. I wouldn't change them unless they’re really obvious like those two.

4. Do our endnote tools (renumber and reorder) handle non-numeric notes? Some of the notes are labeled in the PG files as, e.g. 2c, or 14a. Will those get renumbered to be numbers, e.g. if I have a note 1, 2, 2a, and 3, will renumber turn those into 1,2,3,4?

5. I’ve found plenty of scans, but getting all volumes of a single edition has proven to be challenging. Plus, finding ones for the edition that PG used for the transcription. I thought from the picture they have in the PG epub that they used this edition, but I’ve since discovered that there aren’t any Smith notes in PG. The HTML file says it’s an 1845 edition, but I haven’t found it yet. I do have a complete set of the edition from the earlier link I can use for now. (I’ve also asked PG what scans they used.)

6. The citations in the endnotes are going to be a problem for the SE tools, because they use abbreviations that look like roman numbers that aren’t. For example, notes 1-3 from the link in #2 above has this information…

1 Dion Cassius (l. liv. p. 736,)…
2 Strabo (l. xvi. p. 780,)…
3 …and Vellius Paterculus, l. ii. c. 117…

In each case, the l. isn't 50, and the c. isn't 100. (I’m not sure what they do mean; I haven’t been able to find anything in a search yet.)

This is exacerbated by a few instances where what follows the L. really is 50, e.g. “…especially L. l. c. ix.” In that case, the capital L (I don’t know why it was capitalized there) is the whatever the other l.’s are (voLume?), while the lowercase l is fifty.

So, a couple of questions.
First, how do we deal with the periods on roman numerals in these citations? Are they still extraneous and we should remove them, or do we leave them to be consistent? IOW, should it be
    Vellius Paterculus l. ii. c. 117
or 
    Vellius Paterculus l. ii c. 117
And
    …especially L. l. c. ix.
or
    …especially L. l c. ix. (the one after ix is a sentence-terminating period)

Second, any ideas on preventing typogrify from wrapping all of those l’s and c's so I don’t have to go hand-fix them all? My first thought is to do a search and replace them with something weird that doesn’t occur elsewhere. The problem is that I’d either have to leave them like that for the entire production until the very end (to allow multiple typogrify’s), or I’d have to do that (and remember to do that) every time I typogrify. Either way, that would be a problem on review, because the reviewer couldn’t run typogrify to see if anything changed, because there would be too much noise to see anything else.

7. There are multiple prefaces (you can see them beginning here). There’s a preface by Milman the editor (Preface by Dean Milman), a preface by Gibbons (Preface of the Author), a preface to the 4th volume of the quarto (“Preface to the Fourth Volume of the Original Quarto Edition). From what I’ve seen, all of those are include in all the various editions. How would we name those various files; they can’t all be preface.xhtml. Would we do the usual and make them the same as the title, e.g. preface-by-dean-milman, preface-of-the-author, and preface-to-the-fourth-volume-of-the-original-quarto-edition?

That should do it for now. :)

Vince

unread,
May 2, 2020, 6:12:58 PM5/2/20
to Standard Ebooks
A few more questions, and updates on a couple of the previous ones.

8. Almost all of the editions I’ve looked at, in one way or another, show the year A.D. as it changes throughout the book. Some of them do it inline, almost all do it in the ToC. For example, see here. The ToC has very detailed entries for each chapter, and shows the year as it changes.

We’re obviously not going to have that detail in the ToC (it’s not represented in the text itself), but I really hate to lose the year information. It’s pretty valuable as you go through the history to know where you are. Could I, at the very least, perhaps add the year range for a chapter to the chapter's epigraph (every chapter has one)? E.g., after the existing epigraph text in chapter 7 in the above link, add something like “(A.D. 235–248)”. There wouldn’t be one for chapters that didn’t have any years. (E.g., chapter 9 in the link, “The State of Germany…”, has no years in the ToC, and so wouldn’t have any in the epigraph.)

9. Some of the footnotes have Greek text (i.e. quoting a Greek source) in them. As far as I can tell, PG didn’t include the Greek text in the transcription; the rest of the note is there, just not the Greek portion. I hate to be 99% of the way there; is there any objection to me putting the Greek text back in the notes? (I haven’t yet seen an instance of him using it in the book text itself, although he certainly could have.)

10. There are also Latin quotes in a number of the footnotes, but the Latin is almost never italicized. (This is across all editions I’ve seen.) I assume we do want to italicize it, per normal SE policies, correct?

11. Do we semanticate books referenced in a footnote? I’ve never had that happen before, but almost every note here has a reference to one or more books. (Well, I assume they’re books.) For example, here are portions of a couple of notes from the first chapter…
    Dion Cassius, (l. liv. p. 736,) with the annotations of Reimar…
    Strabo, (l. xvi. p. 780,) Pliny the elder, (Hist. Natur. l. vi. c. 32, 35, [28, 29,]) and Dion Cassius, (l. liii. p. 723, and l. liv. p. 734,) have left us…

“Dion Cassius” appears to be referring to Lucius Cassius Dio (don’t where the ’n’ came from); he wrote an 80(!)-volume history of Rome. So he’s using “Dion Cassius” as a shorthand for the book, but he gives the actual reference in the history (l. liv. p. 736,).

Same for “Strabo”; although he also wrote a history, almost nothing survives of it, so I’m assuming the note is referring to his Geographica. (The note here is the first mention of Strabo in D&F, so there’s not a more detailed earlier citation showing the book name; Gibbons just assumes everyone knows what he’s talking about, and probably everybody reading this back then did.)

In Pliny’s case it refers to the actual book name (Hist. Natur., an abbreviation for his Naturalis Historia).

Do any/all of those need to be wrapped in a <span epub:type="epub:type="se:name.publication.book”>? (Or, actually, an <i epub:type…?) If so, that might tip me over the edge :); I don’t see any way to do that with a regex. In this case, I assume at least Dion Cassius and Strabo would not be, since they’re not the actual book names. But they have detailed references attached to them. So there’s no way I can see to tell the difference between them and the Pliny reference, regex-wise.


As more information on #2 (the notes on notes), it appears that PG did in the transcription what I suggested, in several but not all places—they merged a note on a footnote into the footnote itself, and just asterisked the reference and note.

<p class=“foot”>This is a footnote with an extra* note on it. *This is the extra note.</p>


As more information for #5 (scans), the above link in question 8 is from a complete five-volume edition I found that matches the PG scans, even though the PG transcription was done from a six-volume edition. The five-volume is one with just Milman’s notes, it’s post-1845 so it has the adjustments that Milman in that edition, and everything I’ve checked so far, where I’ve found differences in other editions, matches the transcription.



--
You received this message because you are subscribed to the Google Groups "Standard Ebooks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to standardebook...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/standardebooks/E25C54CE-27D6-4558-9306-0F0D0B5A0389%40letterboxes.org.

David Grigg

unread,
May 2, 2020, 7:37:29 PM5/2/20
to standar...@googlegroups.com
You're a braver man than I am, Vince!

I can answer (4): Yes, renumber-endnotes will renumber endnotes in a simple sequence. All that's required on the first pass is that notes have UNIQUE noterefs in both the body and in endnotes.xhtml. It can be any kind of code. For example in Pilgrim's Progress there were initially two sets of notes, one for each Part, and so I just made the noterefs "1-34", "2-15" etc. Then the renumber function goes through and matches them up and rebuilds endnotes.xhtml.

Vince Rice

unread,
May 2, 2020, 9:55:32 PM5/2/20
to standar...@googlegroups.com
I think the word you’re looking for is “dumber.” :)

Excellent, that’s good to know. And it goes in order of appearance, so if I have note 1, 15, and 4, then the renumber makes it 1 (former 1), 2 (former 15), 3 (former 4), correct? (I ask because sometimes a “subnote” is numbered by just appending numbers, e.g. note 61, note 611, note 62.) And the numbers start over in every chapter, so I’ll have to do what you did, prefix the note# with the chapter number.

And you remind me that they have to be in the same order in both the body and endnotes.xhtml (duh!), so that probably answers #2 as well; I think footnotes that have footnotes are going to have to be merged into a single note, with some kind of notation approved by Alex to indicate the footnote (superscripted letters? numbers? asterisks? …), otherwise renumber notes is going to get confused. The notation inside the note can be display-only, since we won’t be tapping/clicking on it.

Here’s a picture of what a note with notes might look like (this is a real example from chapter 1); I used letters for this example, but they could be anything. They would only be styled with <sup>a</sup>, since they’re only for display. (These superscripts show smaller than a regular <sup></sup> would.)


David Grigg

unread,
May 2, 2020, 10:17:20 PM5/2/20
to standar...@googlegroups.com
> you remind me that they have to be in the same order in both the body and endnotes.xhtml (duh!)

Not really. So long as the routine can match up a noteref anywhere in endnotes.xhtml it will link it to the note it is renumbering in the body, and will output an endnotes.xhtml in the correct numerical order.

It IS important to have correctly ordered the spine in content.opf, though, because that’s the order of files that the routine will follow.
On 3 May 2020, 11:55 AM +1000, Vince Rice <v...@therices.name>, wrote:
I think the word you’re looking for is “dumber.” :)

Excellent, that’s good to know. And it goes in order of appearance, so if I have note 1, 15, and 4, then the renumber makes it 1 (former 1), 2 (former 15), 3 (former 4), correct? (I ask because sometimes a “subnote” is numbered by just appending numbers, e.g. note 61, note 611, note 62.) And the numbers start over in every chapter, so I’ll have to do what you did, prefix the note# with the chapter number.

And you remind me that they have to be in the same order in both the body and endnotes.xhtml (duh!), so that probably answers #2 as well; I think footnotes that have footnotes are going to have to be merged into a single note, with some kind of notation approved by Alex to indicate the footnote (superscripted letters? numbers? asterisks? …), otherwise renumber notes is going to get confused. The notation inside the note can be display-only, since we won’t be tapping/clicking on it.

Here’s a picture of what a note with notes might look like (this is a real example from chapter 1); I used letters for this example, but they could be anything. They would only be styled with <sup>a</sup>, since they’re only for display. (These superscripts show smaller than a regular <sup></sup> would.)

<PastedGraphic-13.tiff>

Alex Cabal

unread,
May 3, 2020, 2:07:27 PM5/3/20
to standar...@googlegroups.com
This is probably going to be the one of the most complex productions we
have, probably Pepys-level or Quixote-level or worse. It's going to take
a really long time. Good luck!

IIRC the PG transcription is a blend of various editions? You may never
find exact scans. Also the book went through a lot of editions and
revisions, and there have been tons of printings, as it was super
popular. We can even use multiple scans ourselves... if for example PG
book 1 is a blend of various editions, we can include scans of any
number of those editions for book 1 to check against.

Yes we can simply put all the volumes into one book, with no divisions.
It was printed in volumes because the whole thing takes up a whole
bookshelf.

Re. notes, I think we should just treat each subnote as an additional
endnote with its own number. When clicking links in an endnote I think
most reading systems open the link in the same popup window. But I'm
sure each one does its own thing so it's not something we can really
target. Each note is semantically its own thing so I think we do want to
keep them separate.

Agree on ligatures.

Not sure what the endnotes tools will do. You can try it. But if we're
renumbering notes to be sequential anyway (like in all our books) there
may not be a reason to keep the a-b-c subdesignations. Can we just give
them their own number?

Re citations, "l" probably stands for "liber" (book) and "c" must be
"chapter". semanticate will incorrectly mark those as you mentioned, but
you have to go in and fix what it does. I don't think there's a nice way
to get around this because I imagine there's going to be a lot of real
Roman numerals in there. Maybe if "L" isn't a common numeral you can use
interactive-sr to regex-replace those quickly. We do want to remove
periods after Roman numerals in all cases.

Prefaces can be preface-1.xhtml, preface-2.xhtml, etc.

On 5/2/20 1:36 PM, Vince wrote:
> While I’m continuing to proof /L’Mort de Arthur/…
>
> I think I may regret this later (maybe even sooner), but I’ve always
> wanted to read Gibbons’ /History of the Decline and Fall of the Roman
> Empire/, so I thought I’d take a look at what it would take to produce
> it. I’ve already run into a problem with the PG transcription, which
> they fixed, but they broke something else in the process, and it’s going
> to be “many days” before they get it resolved. But that’s for them; for
> us, I have some questions.
>
> 1. Although it’s always printed in multiple volumes, it doesn’t appear
> to be /structured/ in volumes (and the number of volumes is all over the
> map—four, six, eight, and I’ve seen as many as twelve). The chapters are
> numbered sequentially from 1 to 71. I think, therefore, that it should
> just have a single level, i.e. chapter. Does that sound OK?
>
> 2. Gibbons included a boatload (technical term) of footnotes. Many of
> the now PD editions, including the one PG used for transcription, also
> include notes by others. For example, in the PG edition, there are notes
> by three people: Gibbons, Milman, and Guizot. This in and of itself is NBD.
>
> However, in some cases, the new notes are notes about Gibbon’s notes. In
> other words, an endnote has an endnote reference in it. For example, see
> here
> <https://archive.org/details/historyofdecline01gibbuoft/page/2/mode/2up>; footnote
> <https://archive.org/details/historydeclinea21gibbgoog/page/n6/mode/2up> edition,
> but I’ve since discovered that there aren’t any Smith notes in PG. The
> HTML file says it’s an 1845 edition, but I haven’t found it yet. I do
> have a complete set of the edition from the earlier link I can use for
> now. (I’ve also asked PG what scans they used.)
>
> 6. The citations in the endnotes are going to be a problem for the SE
> tools, because they use abbreviations that look like roman numbers that
> aren’t. For example, notes 1-3 from the link in #2 above has this
> information…
>
> 1 Dion Cassius (l. liv. p. 736,)…
> 2 Strabo (l. xvi. p. 780,)…
> 3 …and Vellius Paterculus, l. ii. c. 117…
>
> In each case, the l. isn't 50, and the c. isn't 100. (I’m not sure what
> they /do/ mean; I haven’t been able to find anything in a search yet.)
>
> This is exacerbated by a few instances where what follows the L. really
> is 50, e.g. “…especially L. l. c. ix.” In that case, the capital L (I
> don’t know why it was capitalized there) is the whatever the other l.’s
> are (voLume?), while the lowercase l is fifty.
>
> So, a couple of questions.
> First, how do we deal with the periods on roman numerals in these
> citations? Are they still extraneous and we should remove them, or do we
> leave them to be consistent? IOW, should it be
> Vellius Paterculus l. ii. c. 117
> or
> Vellius Paterculus l. ii c. 117
> And
>     …especially L. l. c. ix.
> or
>     …especially L. l c. ix. (the one after ix is a sentence-terminating
> period)
>
> Second, any ideas on preventing typogrify from wrapping all of those l’s
> and c's so I don’t have to go hand-fix them all? My first thought is to
> do a search and replace them with something weird that doesn’t occur
> elsewhere. The problem is that I’d either have to leave them like that
> for the entire production until the very end (to allow multiple
> typogrify’s), or I’d have to do that (and /remember/ to do that) every
> time I typogrify. Either way, that would be a problem on review, because
> the reviewer couldn’t run typogrify to see if anything changed, because
> there would be too much noise to see anything else.
>
> 7. There are multiple prefaces (you can see them beginning here
> <https://archive.org/details/historydeclinea21gibbgoog/page/n12/mode/2up>).
> There’s a preface by Milman the editor (Preface by Dean Milman), a
> preface by Gibbons (Preface of the Author), a preface to the 4th volume
> of the quarto (“Preface to the Fourth Volume of the Original Quarto
> Edition). From what I’ve seen, all of those are include in all the
> various editions. How would we name those various files; they can’t all
> be preface.xhtml. Would we do the usual and make them the same as the
> title, e.g. preface-by-dean-milman, preface-of-the-author, and
> preface-to-the-fourth-volume-of-the-original-quarto-edition?
>
> That should do it for now. :)
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> <https://groups.google.com/d/msgid/standardebooks/E25C54CE-27D6-4558-9306-0F0D0B5A0389%40letterboxes.org?utm_medium=email&utm_source=footer>.

Alex Cabal

unread,
May 3, 2020, 2:13:10 PM5/3/20
to standar...@googlegroups.com
On 5/2/20 5:12 PM, Vince wrote> 8. Almost all of the editions I’ve
looked at, in one way or another,
> show the year A.D. as it changes throughout the book. Some of them do it
> inline, almost all do it in the ToC. For example, see here
> <https://archive.org/details/historyofdecline55gibb/page/32/mode/2up>.
> The ToC has very detailed entries for each chapter, and shows the year
> as it changes.
>
> We’re obviously not going to have that detail in the ToC (it’s not
> represented in the text itself), but I really hate to lose the year
> information. It’s pretty valuable as you go through the history to know
> where you are. Could I, at the very least, perhaps add the year range
> for a chapter to the chapter's epigraph (every chapter has one)? E.g.,
> after the existing epigraph text in chapter 7 in the above link, add
> something like “(A.D. 235–248)”. There wouldn’t be one for chapters that
> didn’t have any years. (E.g., chapter 9 in the link, “The State of
> Germany…”, has no years in the ToC, and so wouldn’t have any in the
> epigraph.)

Sure. But those aren't epigraphs, they're bridgeheads.

> 9. Some of the footnotes have Greek text (i.e. quoting a Greek source)
> in them. As far as I can tell, PG didn’t include the Greek text in the
> transcription; the rest of the note is there, just not the Greek
> portion. I hate to be 99% of the way there; is there any objection to me
> putting the Greek text back in the notes? (I haven’t yet seen an
> instance of him using it in the book text itself, although he certainly
> could have.)

Yes of course. That's part of the point of SE!

> 10. There are also Latin quotes in a number of the footnotes, but the
> Latin is almost never italicized. (This is across all editions I’ve
> seen.) I assume we /do/ want to italicize it, per normal SE policies,
> correct?

Yes

> 11. Do we semanticate books referenced in a footnote? I’ve never had
> that happen before, but almost every note here has a reference to one or
> more books. (Well, I assume they’re books.) For example, here are
> portions of a couple of notes from the first chapter…
>     Dion Cassius, (l. liv. p. 736,) with the annotations of Reimar…
>     Strabo, (l. xvi. p. 780,) Pliny the elder, (Hist. Natur. l. vi. c.
> 32, 35, [28, 29,]) and Dion Cassius, (l. liii. p. 723, and l. liv. p.
> 734,) have left us…

Yes, if they are proper names of publications and not just "book 2 of
some larger book". In the above, I think only "Hist Natur" would get
italics/semantics. Author names like Dion Cassius don't get italics or
semantics.

Alex Cabal

unread,
May 3, 2020, 2:14:39 PM5/3/20
to standar...@googlegroups.com
This might be a good book to split up among various producers. This is a
4-6 month production at least IMHO.

On 5/2/20 5:12 PM, Vince wrote:
> A few more questions, and updates on a couple of the previous ones.
>
> 8. Almost all of the editions I’ve looked at, in one way or another,
> show the year A.D. as it changes throughout the book. Some of them do it
> inline, almost all do it in the ToC. For example, see here
> <https://archive.org/details/historyofdecline55gibb/page/32/mode/2up>.
> The ToC has very detailed entries for each chapter, and shows the year
> as it changes.
>
> We’re obviously not going to have that detail in the ToC (it’s not
> represented in the text itself), but I really hate to lose the year
> information. It’s pretty valuable as you go through the history to know
> where you are. Could I, at the very least, perhaps add the year range
> for a chapter to the chapter's epigraph (every chapter has one)? E.g.,
> after the existing epigraph text in chapter 7 in the above link, add
> something like “(A.D. 235–248)”. There wouldn’t be one for chapters that
> didn’t have any years. (E.g., chapter 9 in the link, “The State of
> Germany…”, has no years in the ToC, and so wouldn’t have any in the
> epigraph.)
>
> 9. Some of the footnotes have Greek text (i.e. quoting a Greek source)
> in them. As far as I can tell, PG didn’t include the Greek text in the
> transcription; the rest of the note is there, just not the Greek
> portion. I hate to be 99% of the way there; is there any objection to me
> putting the Greek text back in the notes? (I haven’t yet seen an
> instance of him using it in the book text itself, although he certainly
> could have.)
>
> 10. There are also Latin quotes in a number of the footnotes, but the
> Latin is almost never italicized. (This is across all editions I’ve
> seen.) I assume we /do/ want to italicize it, per normal SE policies,
> correct?
>
> 11. Do we semanticate books referenced in a footnote? I’ve never had
> that happen before, but almost every note here has a reference to one or
> more books. (Well, I assume they’re books.) For example, here are
> portions of a couple of notes from the first chapter…
>     Dion Cassius, (l. liv. p. 736,) with the annotations of Reimar…
>     Strabo, (l. xvi. p. 780,) Pliny the elder, (Hist. Natur. l. vi. c.
> 32, 35, [28, 29,]) and Dion Cassius, (l. liii. p. 723, and l. liv. p.
> 734,) have left us…
>
> “Dion Cassius” appears to be referring to Lucius Cassius Dio (don’t
> where the ’n’ came from); he wrote an 80(!)-volume history of Rome. So
> he’s using “Dion Cassius” as a shorthand for the book, but he gives the
> actual reference in the history (l. liv. p. 736,).
>
> Same for “Strabo”; although he also wrote a history, almost nothing
> survives of it, so I’m assuming the note is referring to his
> /Geographica. /(The note here is the first mention of Strabo in D&F, so
> there’s not a more detailed earlier citation showing the book name;
> Gibbons just assumes everyone knows what he’s talking about, and
> probably everybody reading this back then did.)
>
> In Pliny’s case it refers to the actual book name (Hist. Natur., an
> abbreviation for his /Naturalis Historia/).
>
> Do any/all of those need to be wrapped in a <span
> epub:type="epub:type="se:name.publication.book”>? (Or, actually, an <i
> epub:type…?) If so, that might tip me over the edge :); I don’t see any
> way to do that with a regex. In this case, I assume at least Dion
> Cassius and Strabo would /not/ be, since they’re not the actual book
> names. But they have detailed references attached to them. So there’s no
> way I can see to tell the difference between them and the Pliny
> reference, regex-wise.
>
>
> As more information on #2 (the notes on notes), it appears that PG did
> in the transcription what I suggested, in several but not all
> places—they merged a note on a footnote into the footnote itself, and
> just asterisked the reference and note.
>
> <p class=“foot”>This is a footnote with an extra* note on it. *This is
> the extra note.</p>
>
>
> As more information for #5 (scans), the above link in question 8 is from
> a complete five-volume edition I found that matches the PG scans, even
> though the PG transcription was done from a six-volume edition. The
> five-volume is one with just Milman’s notes, it’s post-1845 so it has
> the adjustments that Milman in that edition, and everything I’ve checked
> so far, where I’ve found differences in other editions, matches the
> transcription.
>
>
>
>> On May 2, 2020, at 1:36 PM, Vince <al...@letterboxes.org
>> <mailto:al...@letterboxes.org>> wrote:
>>
>> While I’m continuing to proof /L’Mort de Arthur/…
>>
>> I think I may regret this later (maybe even sooner), but I’ve always
>> wanted to read Gibbons’ /History of the Decline and Fall of the Roman
>> Empire/, so I thought I’d take a look at what it would take to produce
>> it. I’ve already run into a problem with the PG transcription, which
>> they fixed, but they broke something else in the process, and it’s
>> going to be “many days” before they get it resolved. But that’s for
>> them; for us, I have some questions.
>>
>> 1. Although it’s always printed in multiple volumes, it doesn’t appear
>> to be /structured/ in volumes (and the number of volumes is all over
>> the map—four, six, eight, and I’ve seen as many as twelve). The
>> chapters are numbered sequentially from 1 to 71. I think, therefore,
>> that it should just have a single level, i.e. chapter. Does that sound OK?
>>
>> 2. Gibbons included a boatload (technical term) of footnotes. Many of
>> the now PD editions, including the one PG used for transcription, also
>> include notes by others. For example, in the PG edition, there are
>> notes by three people: Gibbons, Milman, and Guizot. This in and of
>> itself is NBD.
>>
>> However, in some cases, the new notes are notes about Gibbon’s notes.
>> In other words, an endnote has an endnote reference in it. For
>> example, see here
>> <https://archive.org/details/historyofdecline01gibbuoft/page/2/mode/2up>;
>> <https://archive.org/details/historydeclinea21gibbgoog/page/n6/mode/2up> edition,
>> but I’ve since discovered that there aren’t any Smith notes in PG. The
>> HTML file says it’s an 1845 edition, but I haven’t found it yet. I do
>> have a complete set of the edition from the earlier link I can use for
>> now. (I’ve also asked PG what scans they used.)
>>
>> 6. The citations in the endnotes are going to be a problem for the SE
>> tools, because they use abbreviations that look like roman numbers
>> that aren’t. For example, notes 1-3 from the link in #2 above has this
>> information…
>>
>> 1 Dion Cassius (l. liv. p. 736,)…
>> 2 Strabo (l. xvi. p. 780,)…
>> 3 …and Vellius Paterculus, l. ii. c. 117…
>>
>> In each case, the l. isn't 50, and the c. isn't 100. (I’m not sure
>> what they /do/ mean; I haven’t been able to find anything in a search
>> yet.)
>>
>> This is exacerbated by a few instances where what follows the L.
>> really is 50, e.g. “…especially L. l. c. ix.” In that case, the
>> capital L (I don’t know why it was capitalized there) is the whatever
>> the other l.’s are (voLume?), while the lowercase l is fifty.
>>
>> So, a couple of questions.
>> First, how do we deal with the periods on roman numerals in these
>> citations? Are they still extraneous and we should remove them, or do
>> we leave them to be consistent? IOW, should it be
>> Vellius Paterculus l. ii. c. 117
>> or
>> Vellius Paterculus l. ii c. 117
>> And
>>     …especially L. l. c. ix.
>> or
>>     …especially L. l c. ix. (the one after ix is a
>> sentence-terminating period)
>>
>> Second, any ideas on preventing typogrify from wrapping all of those
>> l’s and c's so I don’t have to go hand-fix them all? My first thought
>> is to do a search and replace them with something weird that doesn’t
>> occur elsewhere. The problem is that I’d either have to leave them
>> like that for the entire production until the very end (to allow
>> multiple typogrify’s), or I’d have to do that (and /remember/ to do
>> that) every time I typogrify. Either way, that would be a problem on
>> review, because the reviewer couldn’t run typogrify to see if anything
>> changed, because there would be too much noise to see anything else.
>>
>> 7. There are multiple prefaces (you can see them beginning here
>> <https://archive.org/details/historydeclinea21gibbgoog/page/n12/mode/2up>).
>> There’s a preface by Milman the editor (Preface by Dean Milman), a
>> preface by Gibbons (Preface of the Author), a preface to the 4th
>> volume of the quarto (“Preface to the Fourth Volume of the Original
>> Quarto Edition). From what I’ve seen, all of those are include in all
>> the various editions. How would we name those various files; they
>> can’t all be preface.xhtml. Would we do the usual and make them the
>> same as the title, e.g. preface-by-dean-milman, preface-of-the-author,
>> and preface-to-the-fourth-volume-of-the-original-quarto-edition?
>>
>> That should do it for now. :)
>>
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Standard Ebooks" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an email to standardebook...@googlegroups.com
>> <mailto:standardebook...@googlegroups.com>.
>> <https://groups.google.com/d/msgid/standardebooks/E25C54CE-27D6-4558-9306-0F0D0B5A0389%40letterboxes.org?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/E2823CC4-8005-4A05-B50F-4598BCAE9C6B%40letterboxes.org
> <https://groups.google.com/d/msgid/standardebooks/E2823CC4-8005-4A05-B50F-4598BCAE9C6B%40letterboxes.org?utm_medium=email&utm_source=footer>.

Alex Cabal

unread,
May 3, 2020, 2:24:58 PM5/3/20
to standar...@googlegroups.com
Re. the maps in the volume, I assume they're all historical maps and
thus of interest to the text itself. We probably want to keep them. But,
they are good candidates for SVG. The trick would be removing the yellow
stains before tracing to SVG... that might be tricky but probably not
impossible. Maybe David Grigg can help you with that, he converted a lot
of really good maps in Ulysses' S Grant's memoirs.

On 5/2/20 8:55 PM, Vince Rice wrote:
> I think the word you’re looking for is “dumber.” :)
>
> Excellent, that’s good to know. And it goes in order of appearance, so
> if I have note 1, 15, and 4, then the renumber makes it 1 (former 1), 2
> (former 15), 3 (former 4), correct? (I ask because sometimes a “subnote”
> is numbered by just appending numbers, e.g. note 61, note 611, note 62.)
> And the numbers start over in every chapter, so I’ll have to do what you
> did, prefix the note# with the chapter number.
>
> And you remind me that they have to be in the same order in both the
> body and endnotes.xhtml (duh!), so that probably answers #2 as well; I
> think footnotes that have footnotes are going to have to be merged into
> a single note, with some kind of notation approved by Alex to indicate
> the footnote (superscripted letters? numbers? asterisks? …), otherwise
> renumber notes is going to get confused. The notation inside the note
> can be display-only, since we won’t be tapping/clicking on it.
>
> Here’s a picture of what a note with notes might look like (this is a
> real example from chapter 1); I used letters for this example, but they
> could be anything. They would only be styled with <sup>a</sup>, since
> they’re only for display. (These superscripts show smaller than a
> regular <sup></sup> would.)
>
>
>> On May 2, 2020, at 6:32 PM, David Grigg <david...@gmail.com
>> <mailto:david...@gmail.com>> wrote:
>>
>> You're a braver man than I am, Vince!
>>
>> I can answer (4): Yes, renumber-endnotes will renumber endnotes in a
>> simple sequence. All that's required on the first pass is that notes
>> have UNIQUE noterefs in both the body and in endnotes.xhtml. It can be
>> any kind of code. For example in Pilgrim's Progress there were
>> initially two sets of notes, one for each Part, and so I just made the
>> noterefs "1-34", "2-15" etc. Then the renumber function goes through
>> and matches them up and rebuilds endnotes.xhtml.
>> On 3 May 2020, 8:12 AM +1000, Vince <al...@letterboxes.org
>> <mailto:al...@letterboxes.org>>, wrote:
>>> A few more questions, and updates on a couple of the previous ones.
>>>
>>> 8. Almost all of the editions I’ve looked at, in one way or another,
>>> show the year A.D. as it changes throughout the book. Some of them do
>>> it inline, almost all do it in the ToC. For example, see here
>>> <https://archive.org/details/historyofdecline55gibb/page/32/mode/2up>. The
>>> ToC has very detailed entries for each chapter, and shows the year as
>>> it changes.
>>>
>>> We’re obviously not going to have that detail in the ToC (it’s not
>>> represented in the text itself), but I really hate to lose the year
>>> information. It’s pretty valuable as you go through the history to
>>> know where you are. Could I, at the very least, perhaps add the year
>>> range for a chapter to the chapter's epigraph (every chapter has
>>> one)? E.g., after the existing epigraph text in chapter 7 in the
>>> above link, add something like “(A.D. 235–248)”. There wouldn’t be
>>> one for chapters that didn’t have any years. (E.g., chapter 9 in the
>>> link, “The State of Germany…”, has no years in the ToC, and so
>>> wouldn’t have any in the epigraph.)
>>>
>>> 9. Some of the footnotes have Greek text (i.e. quoting a Greek
>>> source) in them. As far as I can tell, PG didn’t include the Greek
>>> text in the transcription; the rest of the note is there, just not
>>> the Greek portion. I hate to be 99% of the way there; is there any
>>> objection to me putting the Greek text back in the notes? (I haven’t
>>> yet seen an instance of him using it in the book text itself,
>>> although he certainly could have.)
>>>
>>> 10. There are also Latin quotes in a number of the footnotes, but the
>>> Latin is almost never italicized. (This is across all editions I’ve
>>> seen.) I assume we /do/ want to italicize it, per normal SE policies,
>>> correct?
>>>
>>> 11. Do we semanticate books referenced in a footnote? I’ve never had
>>> that happen before, but almost every note here has a reference to one
>>> or more books. (Well, I assume they’re books.) For example, here are
>>> portions of a couple of notes from the first chapter…
>>>     Dion Cassius, (l. liv. p. 736,) with the annotations of Reimar…
>>>     Strabo, (l. xvi. p. 780,) Pliny the elder, (Hist. Natur. l. vi.
>>> c. 32, 35, [28, 29,]) and Dion Cassius, (l. liii. p. 723, and l. liv.
>>> p. 734,) have left us…
>>>
>>> “Dion Cassius” appears to be referring to Lucius Cassius Dio (don’t
>>> where the ’n’ came from); he wrote an 80(!)-volume history of Rome.
>>> So he’s using “Dion Cassius” as a shorthand for the book, but he
>>> gives the actual reference in the history (l. liv. p. 736,).
>>>
>>> Same for “Strabo”; although he also wrote a history, almost nothing
>>> survives of it, so I’m assuming the note is referring to his
>>> /Geographica./ (The note here is the first mention of Strabo in D&F,
>>> so there’s not a more detailed earlier citation showing the book
>>> name; Gibbons just assumes everyone knows what he’s talking about,
>>> and probably everybody reading this back then did.)
>>>
>>> In Pliny’s case it refers to the actual book name (Hist. Natur., an
>>> abbreviation for his /Naturalis Historia/).
>>>
>>> Do any/all of those need to be wrapped in a <span
>>> epub:type="epub:type="se:name.publication.book”>? (Or, actually, an
>>> <i epub:type…?) If so, that might tip me over the edge :); I don’t
>>> see any way to do that with a regex. In this case, I assume at least
>>> Dion Cassius and Strabo would /not/ be, since they’re not the actual
>>> book names. But they have detailed references attached to them. So
>>> there’s no way I can see to tell the difference between them and the
>>> Pliny reference, regex-wise.
>>>
>>>
>>> As more information on #2 (the notes on notes), it appears that PG
>>> did in the transcription what I suggested, in several but not all
>>> places—they merged a note on a footnote into the footnote itself, and
>>> just asterisked the reference and note.
>>>
>>> <p class=“foot”>This is a footnote with an extra* note on it. *This
>>> is the extra note.</p>
>>>
>>>
>>> As more information for #5 (scans), the above link in question 8 is
>>> from a complete five-volume edition I found that matches the PG
>>> scans, even though the PG transcription was done from a six-volume
>>> edition. The five-volume is one with just Milman’s notes, it’s
>>> post-1845 so it has the adjustments that Milman in that edition, and
>>> everything I’ve checked so far, where I’ve found differences in other
>>> editions, matches the transcription.
>>>
>>>
>>>
>>>> On May 2, 2020, at 1:36 PM, Vince <al...@letterboxes.org
>>>> <mailto:al...@letterboxes.org>> wrote:
>>>>
>>>> While I’m continuing to proof /L’Mort de Arthur/…
>>>>
>>>> I think I may regret this later (maybe even sooner), but I’ve always
>>>> wanted to read Gibbons’ /History of the Decline and Fall of the
>>>> Roman Empire/, so I thought I’d take a look at what it would take to
>>>> produce it. I’ve already run into a problem with the PG
>>>> transcription, which they fixed, but they broke something else in
>>>> the process, and it’s going to be “many days” before they get it
>>>> resolved. But that’s for them; for us, I have some questions.
>>>>
>>>> 1. Although it’s always printed in multiple volumes, it doesn’t
>>>> appear to be /structured/ in volumes (and the number of volumes is
>>>> all over the map—four, six, eight, and I’ve seen as many as twelve).
>>>> The chapters are numbered sequentially from 1 to 71. I think,
>>>> therefore, that it should just have a single level, i.e. chapter.
>>>> Does that sound OK?
>>>>
>>>> 2. Gibbons included a boatload (technical term) of footnotes. Many
>>>> of the now PD editions, including the one PG used for transcription,
>>>> also include notes by others. For example, in the PG edition, there
>>>> are notes by three people: Gibbons, Milman, and Guizot. This in and
>>>> of itself is NBD.
>>>>
>>>> However, in some cases, the new notes are notes about Gibbon’s
>>>> notes. In other words, an endnote has an endnote reference in it.
>>>> For example, see here
>>>> <https://archive.org/details/historyofdecline01gibbuoft/page/2/mode/2up>;
>>>> <https://archive.org/details/historydeclinea21gibbgoog/page/n6/mode/2up> edition,
>>>> but I’ve since discovered that there aren’t any Smith notes in PG.
>>>> The HTML file says it’s an 1845 edition, but I haven’t found it yet.
>>>> I do have a complete set of the edition from the earlier link I can
>>>> use for now. (I’ve also asked PG what scans they used.)
>>>>
>>>> 6. The citations in the endnotes are going to be a problem for the
>>>> SE tools, because they use abbreviations that look like roman
>>>> numbers that aren’t. For example, notes 1-3 from the link in #2
>>>> above has this information…
>>>>
>>>> 1 Dion Cassius (l. liv. p. 736,)…
>>>> 2 Strabo (l. xvi. p. 780,)…
>>>> 3 …and Vellius Paterculus, l. ii. c. 117…
>>>>
>>>> In each case, the l. isn't 50, and the c. isn't 100. (I’m not sure
>>>> what they /do/ mean; I haven’t been able to find anything in a
>>>> search yet.)
>>>>
>>>> This is exacerbated by a few instances where what follows the L.
>>>> really is 50, e.g. “…especially L. l. c. ix.” In that case, the
>>>> capital L (I don’t know why it was capitalized there) is the
>>>> whatever the other l.’s are (voLume?), while the lowercase l is fifty.
>>>>
>>>> So, a couple of questions.
>>>> First, how do we deal with the periods on roman numerals in these
>>>> citations? Are they still extraneous and we should remove them, or
>>>> do we leave them to be consistent? IOW, should it be
>>>> Vellius Paterculus l. ii. c. 117
>>>> or
>>>> Vellius Paterculus l. ii c. 117
>>>> And
>>>>     …especially L. l. c. ix.
>>>> or
>>>>     …especially L. l c. ix. (the one after ix is a
>>>> sentence-terminating period)
>>>>
>>>> Second, any ideas on preventing typogrify from wrapping all of those
>>>> l’s and c's so I don’t have to go hand-fix them all? My first
>>>> thought is to do a search and replace them with something weird that
>>>> doesn’t occur elsewhere. The problem is that I’d either have to
>>>> leave them like that for the entire production until the very end
>>>> (to allow multiple typogrify’s), or I’d have to do that (and
>>>> /remember/ to do that) every time I typogrify. Either way, that
>>>> would be a problem on review, because the reviewer couldn’t run
>>>> typogrify to see if anything changed, because there would be too
>>>> much noise to see anything else.
>>>>
>>>> 7. There are multiple prefaces (you can see them beginning here
>>>> <https://archive.org/details/historydeclinea21gibbgoog/page/n12/mode/2up>).
>>>> There’s a preface by Milman the editor (Preface by Dean Milman), a
>>>> preface by Gibbons (Preface of the Author), a preface to the 4th
>>>> volume of the quarto (“Preface to the Fourth Volume of the Original
>>>> Quarto Edition). From what I’ve seen, all of those are include in
>>>> all the various editions. How would we name those various files;
>>>> they can’t all be preface.xhtml. Would we do the usual and make them
>>>> the same as the title, e.g. preface-by-dean-milman,
>>>> preface-of-the-author, and
>>>> preface-to-the-fourth-volume-of-the-original-quarto-edition?
>>>>
>>>> That should do it for now. :)
>>>>
>>>>
>>>> --
>>>> You received this message because you are subscribed to the Google
>>>> Groups "Standard Ebooks" group.
>>>> To unsubscribe from this group and stop receiving emails from it,
>>>> send an email to standardebook...@googlegroups.com
>>>> <mailto:standardebook...@googlegroups.com>.
>>>> <https://groups.google.com/d/msgid/standardebooks/E25C54CE-27D6-4558-9306-0F0D0B5A0389%40letterboxes.org?utm_medium=email&utm_source=footer>.
>>>
>>>
>>> --
>>> You received this message because you are subscribed to the Google
>>> Groups "Standard Ebooks" group.
>>> To unsubscribe from this group and stop receiving emails from it,
>>> send an email to standardebook...@googlegroups.com
>>> <mailto:standardebook...@googlegroups.com>.
>>> To view this discussion on the web visit
>>> https://groups.google.com/d/msgid/standardebooks/E2823CC4-8005-4A05-B50F-4598BCAE9C6B%40letterboxes.org
>>> <https://groups.google.com/d/msgid/standardebooks/E2823CC4-8005-4A05-B50F-4598BCAE9C6B%40letterboxes.org?utm_medium=email&utm_source=footer>.
>>
>> --
>> You received this message because you are subscribed to the Google
>> Groups "Standard Ebooks" group.
>> To unsubscribe from this group and stop receiving emails from it, send
>> an email to standardebook...@googlegroups.com
>> <mailto:standardebook...@googlegroups.com>.
>> To view this discussion on the web visit
>> https://groups.google.com/d/msgid/standardebooks/43bad770-3a4d-498f-a1a5-d2f22f3eec28%40Spark
>> <https://groups.google.com/d/msgid/standardebooks/43bad770-3a4d-498f-a1a5-d2f22f3eec28%40Spark?utm_medium=email&utm_source=footer>.
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/0E97589D-F517-4CA3-A36D-66B344236F27%40therices.name
> <https://groups.google.com/d/msgid/standardebooks/0E97589D-F517-4CA3-A36D-66B344236F27%40therices.name?utm_medium=email&utm_source=footer>.

Vince Rice

unread,
May 3, 2020, 4:27:58 PM5/3/20
to standar...@googlegroups.com
> On May 3, 2020, at 1:13 PM, Alex Cabal <al...@standardebooks.org> wrote:
>
> On 5/2/20 5:12 PM, Vince wrote> 8. Almost all of the editions I’ve looked at, in one way or another,
>> show the year A.D. as it changes throughout the book. Some of them do it inline, almost all do it in the ToC. For example, see here <https://archive.org/details/historyofdecline55gibb/page/32/mode/2up>. The ToC has very detailed entries for each chapter, and shows the year as it changes.
>> We’re obviously not going to have that detail in the ToC (it’s not represented in the text itself), but I really hate to lose the year information. It’s pretty valuable as you go through the history to know where you are. Could I, at the very least, perhaps add the year range for a chapter to the chapter's epigraph (every chapter has one)? E.g., after the existing epigraph text in chapter 7 in the above link, add something like “(A.D. 235–248)”. There wouldn’t be one for chapters that didn’t have any years. (E.g., chapter 9 in the link, “The State of Germany…”, has no years in the ToC, and so wouldn’t have any in the epigraph.)
>
> Sure. But those aren't epigraphs, they're bridgeheads.

Yes, sorry, I can never keep those straight when talking about them. I was thinking of Mallory, which also has bridgeheads on every chapter, I just said the wrong word.

>> 11. Do we semanticate books referenced in a footnote? I’ve never had that happen before, but almost every note here has a reference to one or more books. (Well, I assume they’re books.) For example, here are portions of a couple of notes from the first chapter…
>> Dion Cassius, (l. liv. p. 736,) with the annotations of Reimar…
>> Strabo, (l. xvi. p. 780,) Pliny the elder, (Hist. Natur. l. vi. c. 32, 35, [28, 29,]) and Dion Cassius, (l. liii. p. 723, and l. liv. p. 734,) have left us…
>
> Yes, if they are proper names of publications and not just "book 2 of some larger book". In the above, I think only "Hist Natur" would get italics/semantics. Author names like Dion Cassius don't get italics or semantics.

Urgh. That’s going to be … interesting.

Vince

unread,
May 3, 2020, 4:40:14 PM5/3/20
to Standard Ebooks
Uggh. I hadn’t even noticed the images. Yucch.

So, David, how’s it go-ing? lol
I know you’re still working on Clarissa, which I was supposed to help with after O. Henry was finished and forgot about when our kids came in unexpectedly from overseas.
If you want to take a break here and there, I would love your help on these images. I have no talent or real interest in the imaging side of the books, so I would greatly appreciate any and all assistance. There are eight images, and it’s obviously no rush at all; as Alex said, this is going to be a several month project. If you would care to convert them to SVG, I’ll email them to you off-list.

Regardless, if you want to give me a Clarissa volume, I’ll start work on that.


To unsubscribe from this group and stop receiving emails from it, send an email to standardebook...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/standardebooks/daa111ac-b50d-e009-d682-b85721ad8433%40standardebooks.org.

David Grigg

unread,
May 3, 2020, 7:10:05 PM5/3/20
to standar...@googlegroups.com
Yes, I'd be more than happy to help convert the maps.

Vince Rice

unread,
May 3, 2020, 7:34:07 PM5/3/20
to standar...@googlegroups.com
On May 3, 2020, at 1:07 PM, Alex Cabal <al...@standardebooks.org> wrote:


Re. notes, I think we should just treat each subnote as an additional endnote with its own number. When clicking links in an endnote I think most reading systems open the link in the same popup window. But I'm sure each one does its own thing so it's not something we can really target. Each note is semantically its own thing so I think we do want to keep them separate.

I mocked up a single chapter to see what would happen. I only have Books on the Mac and iOS and the Kindle app on both as well to test. I don’t have or use Chrome, so I don’t have Readium. I used Marvin on iOS long ago, but Christian has apparently abandoned it, and even v3 didn’t support popup footnotes, so I gave it up for Books two or three years ago.

This is what I found. (I’ve attached the repository for this one chapter and endnote, if you want to play with it further. I only got it to the point I could build it to test this, so it’s not “clean”.)

renumber-endnote
This is the biggest issue right now. It doesn’t appear to look for endnotes in the endnotes.xhtml file (and, prior to this, there was no reason why it should), so the notes that are pointing to endnotes.xhtml instead of chapter-1.xhtml are deleted out of endnotes.xhtml on a renumber. I don’t know how feasible it is to have it look for, and properly renumber, a merger of notes in text files and notes in endnotes.xhtml, but I do know I don’t want to tackle this without it. :) There are 8600+ notes in this book, so not having renumber support is a deal-killer.

If it can be made to support it, then great!
But if it can't, then I think we have to merge notes on notes into the original note. The additional notes are clearly delineated (Milman puts “—M.” or “—M. 1845” at the end of his, and he has “—G.” at the end of Guizot’s.), so it will be obvious there are notes from multiple people involved. (In addition to whatever we use to identify the notes, i.e. a letter, number, asterisk, whatever.)

(For the readers, I’m not advocating for anything, i.e. I understand these are reader problems, not SE problems, I’m just noting what happens so we know what we’re dealing with.)

Books on the Mac
1. It does not appear to support formatting in the noteref popup (I’ve run into this before on other formatting in endnotes), so the note references in the note are not superscripted. See “2a” and “2b” in picture.
test_book.zip
PastedGraphic-15.tiff
PastedGraphic-16.tiff

David Grigg

unread,
May 3, 2020, 8:29:33 PM5/3/20
to standar...@googlegroups.com
Hmmm, interesting. In theory there’s no reason that renumber-endnotes shouldn’t treat endnotes.xhtml itself as another ‘source’ file for endnotes, and provided the note-refs are all unique it should find them; but the biggest problem is that the routine then overwrites the existing endnotes.xhtml which is likely to trash all of the renumbered note references within it, it’s a bit self-referential. Let me think about it.

David Grigg

unread,
May 4, 2020, 3:11:56 AM5/4/20
to standar...@googlegroups.com
OK. It's definitely do-able, but it requires a separate pass through endnotes.xhtml after it's dealt with all the other files. I'm running some trials with modifications to my original code (since I can't readily debug the version now in se tools) with a small dummy project. I should have an answer on a couple of days.

David Grigg

unread,
May 4, 2020, 7:58:44 AM5/4/20
to Standard Ebooks
Renumber endnotes with notes inside endnotes:

I have something which works, at least on a 'toy' project. Attached are before and after endnotes.xhtml. Needs much more testing, but in principle it works.

This is using my original, stand-alone python code. I'll have to look carefully at the code in the toolset, because Alex refactored it quite a bit.
Initial endnotes.xhtml
Processed endnotes.xhtml

David Grigg

unread,
May 4, 2020, 8:03:26 AM5/4/20
to Standard Ebooks
Sorry, I mucked up InitialEndnotes.xhtml a bit. Use the attached instead.


Initial endnotes.xhtml

Vince

unread,
May 4, 2020, 10:19:52 AM5/4/20
to standar...@googlegroups.com
The zip I attached to my last email should also be a pretty good test. It’s the first chapter from D&F, and has a half-dozen or so notes on notes. I’ll test it myself with your new code later (I’m about to head out).

Vince

unread,
May 4, 2020, 3:07:55 PM5/4/20
to standar...@googlegroups.com
Sorry, I only glanced at the emails on the way out the door and thought the attachments were the new code. Never mind!

On InitialEndNotes, I don’t know if it matters, but on the first one the id (note-3a) is not the same as the link (noteref-3). Those should be the same. (I’m looking at the second one you sent, not the first one.)

The results look good, with one exception. This looks like it's going to put all of the "notes on notes" at the end of endnotes, instead of next to the notes they’re referring to. For example, here note 3a should immediately follow note 3, i.e. turn into note 4, not note 6. Based on what I’ve seen so far, that’s going to make navigation even more complicated than it already is. It also makes it impossible to just read Endnotes and see the related notes next to what they’re about. (I understand the difficulty of doing that; it’s why I think the notes should just be combined.) I’ve found so far that trying to read each endnote as it appears is too interruptive, but there’s too much information to ignore them completely.

This is one of the rare times that a printed book has a real advantage over an e-book. With the printed book, you can read the facing pages, then read all of the notes for those pages at once (if you choose). Normally having popups is nice, but the fact that nothing (I have anyway) supports notes on notes very well makes it a lot more intrusive and more difficult to read.

Hold off on doing any more for a bit, I’m going to make another appeal to leave them combined.

Vince Rice

unread,
May 4, 2020, 3:40:21 PM5/4/20
to Standard Ebooks
> On May 3, 2020, at 1:07 PM, Alex Cabal <al...@standardebooks.org> wrote:
>
> …
> Re. notes, I think we should just treat each subnote as an additional endnote with its own number. When clicking links in an endnote I think most reading systems open the link in the same popup window. But I'm sure each one does its own thing so it's not something we can really target. Each note is semantically its own thing so I think we do want to keep them separate.

Alex, I would like to make an appeal for you to reconsider this one. After further investigation, there are a few problems, and they make things harder for both the producer and the reader, and there’s no corresponding benefit.

1. PG has combined, i.e. merged, the notes. The ones I found that weren’t merged were because those editor notes were on the text, not an existing note. (The majority of the editor notes are on Gibbons’ notes, but there are some that are on the text itself.) That means every note that has a note is going to have to be manually split out. There are going to be hundreds of them over the entire book. (There were an even dozen in the first of 71 chapters.) That is a lot of extra work.

(When I say merged, I mean they’re part of the same endnote. They’re still delineated by starting a new paragraph and start with an asterisk and "Note: " and have a “signature" at the end to indicate they’re an editor note, not part of the Gibbon note.)

2. The reader support (I have access to, anyway) makes it more difficult to read the notes (see my earlier email). None of the ones I’ve tested support the "note on note" in a popup. All of them make reading the notes on notes a lot more cumbersome than just reading a regular endnote.

3. Renumber-endnotes doesn’t currently support notes on notes, i.e. notes whose source is endnotes.xhtml instead of one of the chapters. David has started work on supporting it, but it looks like doing so is going to put all of the “notes on notes” at the end of endnotes, instead of next to the note they’re referring to. I don’t want think we want to lose that proximity; the notes in this book are so voluminous and useful (like Pepys, the notes are usually commentary, not just citations) that it can make more sense to read them separately, i.e. just go to Endnotes and read a chapter’s notes all at once.

Merging them (or, in this case, leaving them merged) thus makes things easier for the producer, easier for the reader, and doesn’t lose any information; each “note on note” is marked as to who made it, and it’s thus clear it’s from an editor, not Gibbon.

Conversely, splitting them out doesn’t provide any real benefit—it doesn’t make the notes easier to read and it doesn’t make them clearer.

Thanks for considering.

David Grigg

unread,
May 4, 2020, 7:46:01 PM5/4/20
to standar...@googlegroups.com
Renumber-endnotes can keep the associated notes and subnotes next to each other, if they are placed together in your original endnotes file. I've tried to be too clever in my example by putting a subnotes out of placde. If the note is followed by subnotes in the source, they will stay that way after renumber-endnotes has done its thing.
On

Alex Cabal

unread,
May 4, 2020, 7:46:06 PM5/4/20
to standar...@googlegroups.com
I get it, it's harder, that's true. But those notes are indeed separate
notes--they even have their own superscripts. We might get around it if
they were presented as inline editorial notes in brackets, but at least
in the examples you sent there can be multiple notes on a single endnote
and they can be quite lengthy. So while yes it's more difficult, they
are semantically separate notes and I think we should split them out.

Re. ereaders, we don't want to produce ebooks based on what ereaders
today are limited to doing. We want to produce ebooks that are
semantically correct, and hopefully ereaders will evolve to be able to
do the right thing with our ebooks in the future. In the meantime,
that's what the build process is for--to decompose 'correct' ebooks into
something more ereader friendly. Right now there's no logic for
notes-on-notes in there, but that doesn't mean there couldn't be.

In your example book, you put the subnotes as the next sibling li. But
we can't do that, because that will disrupt the ordering of the
subsequent notes (and also li items can't contain letters in their
ordinal). I think we have two options:

1. We could treat the subnotes as part of the natural endnote sequence.
That will result in endnote-2 having an endnote to endnote-8003, which
might be a little weird but not the end of the world.

2. We could have a 2nd or 3rd endnotes file just for the supplementary
endnotes, and preface noterefs in the original endnotes.xhtml with
something like "M" (or whatever the author's initial is). This is a
little like what happened in 10 Days that Shook the World, where there
is a set of actual endnotes, and then a set of noterefs prefaced with
"A" pointing to various appendices; the two aren't merged because the
appendices weren't actually endnotes, but more like supplementary chapters.

Vince Rice

unread,
May 4, 2020, 8:09:39 PM5/4/20
to Standard Ebooks
I put the notes as the next sibling because that’s where it is in the file. Again, PG made them all part of one big note; I just split them up into two or three notes.

I’m not sure what you mean by “disrupt the ordering of the subsequent notes”. We’re renumbering all of the notes anyway. We won’t keep the 2a-type noteref, with renumbering that would become 3, and 3 would become 4, and so on. So, instead of 2 having subnotes of 2a and 2b, as it does in my example (I left it that way because renumber-endnotes doesn’t support this scheme yet), 2 would have subnotes of 3 and 4, and the next “real” note would be 5. And renumber-endnotes would handle all of that.

The second endnotes file is an interesting possibility. I didn’t know we’d done that before; that’s the kind of thing I was talking about the other day when David was asking about endnotes vs scripture references. In this case, we could have an endnote file for Gibbons endnotes and a Editor endnote file for the other endnotes (Milman’s and Guizot’s). I don’t think there are enough Guizot notes to warrant a third endnote file; we could just mix M- and G- prefixes in the one. But it still loses proximity, and I personally think that’s a big loss. But I understand I might be alone in that thinking.

I understand our policy re ebooks and readers; I said so in my original email about it. But we also don’t want to make a book that’s untenable to read with what exists now, and this is, IMO, pretty close. The “harder” for the producer I can live with, but making it harder for the reader for no real benefit not so much (I’m a reader, and I dislike readers that make it harder for me to read a book). I want to produce this because I want to read it, and this makes for a much worse reading experience, IMO. Yes, they’re separate notes, but they would remain separate, each in its own paragraph and with the author mark. No information lost, no confusion, nothing but a much better reading experience now.

Nevertheless, it’s your call. I’ll have to think about this some more. Thanks for listening.


--
You received this message because you are subscribed to the Google Groups "Standard Ebooks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to standardebook...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/standardebooks/daad9937-6993-218d-223a-1818ebf1ebee%40standardebooks.org.

Alex Cabal

unread,
May 4, 2020, 8:16:43 PM5/4/20
to standar...@googlegroups.com
On 5/4/20 7:09 PM, Vince Rice wrote:
> I’m not sure what you mean by “disrupt the ordering of the subsequent
> notes”. We’re renumbering all of the notes anyway. We won’t keep the
> 2a-type noteref, with renumbering that would become 3, and 3 would
> become 4, and so on. So, instead of 2 having subnotes of 2a and 2b, as
> it does in my example (I left it that way because renumber-endnotes
> doesn’t support this scheme yet), 2 would have subnotes of 3 and 4, and
> the next “real” note would be 5. And renumber-endnotes would handle all
> of that.

In the current example, if you open the endnotes file in a web browser,
note 2a actually is listed as note 3. That's what I mean by disrupting
the order. That means that in the text, note 2 would be followed by note
5, which is incorrect. So like I said earlier, the actual result would
be the situation where note-2 has a noteref to note-8000, because the
reading order of the book has endnotes.xhtml at the end, and thus the
notes in endnotes.xhtml itself would be very last in that order.

Vince Rice

unread,
May 4, 2020, 8:23:18 PM5/4/20
to Standard Ebooks
Ah, OK. So if in InitialEndnotes they had been ordered 1,2,3,3a,5,7, then the result would have been 1(1), 2(2), 3(3), 4(3a), 5(5), 6(7)? (What’s in parentheses are the former number.)

What determines the order—the order of the notes in the text itself, or the order of the notes in endnotes? IOW, if the text order is 1,2,4,3; and the endnotes order is 4,2,1,3; then what order do they come out of renumber in? Would it be text order, i.e. 1(1), 2(2), 3(4), 4(3)?

And mixed in here are additional notes added in the text itself.

So, a real-world case, the text notes might look like this
1,2,3,31,4
and these would be in the endnotes
1,2,2a (or 2-1 or whatever),3,31,4

Would the end result be 1(1), 2(2), 3(2a), 4(3), 5(31), 6(4)?

Thanks, David.

David Grigg

unread,
May 4, 2020, 8:32:51 PM5/4/20
to standar...@googlegroups.com
Well it would have been good if what I had said in my last post was actually TRUE. But after thinking about it, in the present state of my code it's NOT true. However, I see clearly the need to MAKE it true. So, back to the drawing board for me!
--
You received this message because you are subscribed to the Google Groups "Standard Ebooks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to standardebook...@googlegroups.com.
To view this discussion on the web visit https://groups.google.com/d/msgid/standardebooks/35C286E9-5CFA-4A84-8348-AB7FBBA997F1%40therices.name.

Vince Rice

unread,
May 5, 2020, 12:24:00 AM5/5/20
to standar...@googlegroups.com
Well, in my view, yes, but not in Alex’s. He doesn’t want gaps in the footnote numbers in the text, which that would cause.

So, in the previous example, e.g. these in the text:
1,2,3,31,4
and these would be in the endnotes:
1,2,2a (or 2-1 or whatever),3,31,4

He wants the text ones to stay in order and the ones in the endnotes to be at the end. So, per him, this should end up:
1(1), 2(2), 3(3), 4(31), 4(5), 6(2a)

I’m not crazy about it, but it’s not my project. :)

Something else to consider—he mentioned the possibility of a second endnote file with a prefix to indicate, in this case, the author. So one endnote would be sequential numbers, the other would be something like M1, M2, M3. I don’t know if it’s feasible to support something like that…

But don’t worry about it for a day or two at least; I’m thinking about what I’m going to do.

Let me know if I can help with Clarissa, though!

Vince Rice

unread,
May 11, 2020, 2:13:41 PM5/11/20
to Standard Ebooks
For the sake of the archives… I’m passing on this, so Gibbons is available if someone else wants to do it.

David Grigg

unread,
May 11, 2020, 11:57:07 PM5/11/20
to standar...@googlegroups.com
If we could solve the endnotes problem, would you be interested in working on this again, perhaps as part of a multi-producer collaboration?
On 12 May 2020, 4:13 AM +1000, Vince Rice <v...@therices.name>, wrote:
For the sake of the archives… I’m passing on this, so Gibbons is available if someone else wants to do it.

--
You received this message because you are subscribed to the Google Groups "Standard Ebooks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to standardebook...@googlegroups.com.

Vince

unread,
May 12, 2020, 1:14:54 AM5/12/20
to Standard Ebooks
The problem is we can't solve the endnotes problem. We have to use endnotes on endnotes, and no reader supports them very well (that I've found, anyway). Trying to force it makes for a bad reading experience, IMO. I enjoy contributing through producing, and reading, but if the latter is a bad/mediocre experience it drags down the former.

But that's OK. There are plenty of books in the sea. :)

Vince

unread,
Aug 31, 2020, 1:01:47 PM8/31/20
to standar...@googlegroups.com
I find it hard to turn aside from a challenge once it worms its way into my head. I’m going to do this, with the notes in notes embedded in the original note while I’m proofing, but structured in a way that will make them easy to move to their own notes before I submit it. (Famous last words.)

This will eventually require that renumber-endnotes and reorder-endnotes support endnotes on endnotes, i.e. endnotes references in the endnotes file pointing to other endnotes. There are 640+ of them (in addition to 8500+ regular endnotes, so 9100+ endnotes total), so it’s too many to handle manually. However, I’m probably months away from needing it, so there’s no hurry. Per earlier discussion in this thread, Alex wants the endnotes in endnotes to be numbered at the end. Thus, if there are 100 endnotes in the book, and the second endnote has its own endnote, then that endnote on endnote would become endnote 101, not endnote 3. (I personally would like to see us use subnumbers here, i.e. the endnote in the above example would be endnote 2-1, rather than 101. That allows the endnotes in the main book to continue to be in order, but allows the endnotes on endnotes to remain in proximity to their source material, as well as indicate by their number what they’re attached to, but I don’t think Alex is keen on that.)

The PG transcriptions are not in very good shape, which is why it will be a while before the above is needed.
  • The one we’re using that has all the endnotes is missing all of the italics and diacritics (at least through chapter 50 [of 71]).
  • Any Greek used in the endnotes is missing, and verse in them is often not marked or formatted, and other languages are not marked, either. (The scans typically don’t italicize the Latin, either.)
  • I found a complete missing chapter when I looked at this originally, and I found several missing pages from a chapter last night, including all of the endnotes contained within those pages.
  • I’ve found a number of typos just in the process of moving all of the endnotes out of the main text to a separate file.
I have found scans of all volumes of the edition I believe PG used for their transcription (the one with endnotes), so that will be (and has been already) a huge help.

There will be lots of questions. :) Today’s revolve around dashes and asterisks in verse, and how we want to treat them. I know to indent verse, the questions all revolve around the dashes/asterisks and whether they should be formatted as 3em’s, or removed, or replaces with ellipses, or…?

1. For example, this from chapter 19. I’m not sure what those dashes are representing. Is it just a single word? Should I treat it as such and use a 3em-dash? Or is it just saying there’s more on the line, and it should be an ellipsis? (Idle rhetorical question: why is Google translate so horrible on Latin? You’d think a dead language would be the easiest to develop translations for.)


2. Then there’s this from chapter 20. This looks like vertical elision to me, and that’s how I’ll format it unless you think differently.


3. Then there’s this from chapter 22. He uses dots (ellipses) in other places, so I don’t know if the asterisks are significant. Maybe it means there was vertical elision and part of that line is missing? How should I format this?

4. This, from a little later in chapter 22, appears quite a bit. A number of dashes at the beginning of a line, but the line starts with a capital, which makes whether there is text prior questionable. And indenting is used in other places (e.g. Eunuchum in #1 above). Are these dashes filling in for words, or just spacing, or what? And how do we want to format it?

This, from chapter 29, has a little of everything. The first line formatting is also quite common (you can look at the few pages following that one and see several other instances), where it is both indented and with a series of leading dashes. It also has a couple of dashes at the end of the second line, and a line of dashes on a line by themselves. Is that line of dashes also vertical elision, i.e. serving the same purpose as the asterisks earlier?

Sometimes (from here) indent and leading dash is not on the first line.

Alex Cabal

unread,
Aug 31, 2020, 6:02:50 PM8/31/20
to standar...@googlegroups.com
On 8/31/20 12:01 PM, Vince wrote:
> I find it hard to turn aside from a challenge once it worms its way into
> my head. I’m going to do this, with the notes in notes embedded in the
> original note while I’m proofing, but structured in a way that will make
> them easy to move to their own notes before I submit it. (Famous last
> words.)

Glad to hear it, I've always wanted to read this one. Note that this was
an extremely successful book with a large number of editions. Before we
start work we should review those editions and decide which one to use.
PG's may not always be the best choice. I'm sure there are pretty big
differences between them all. Did we already do this when you started it
earlier?

> 1. For example, this from chapter 19
> <https://archive.org/details/historyofdecline54gibb/page/114/mode/2up>.
> I’m not sure what those dashes are representing. Is it just a single
> word? Should I treat it as such and use a 3em-dash? Or is it just saying
> there’s more on the line, and it should be an ellipsis?

This looks like a kind of ellipses to me. Maybe you can find the source
text of that quote?

> 2. Then there’s this from chapter 20
> <https://archive.org/details/historyofdecline54gibb/page/170/mode/2up>.
> This looks like vertical elision to me, and that’s how I’ll format it
> unless you think differently.

That is elision, you can find rules for that in the manual. Same with
the horizontal bars in the last image.

> 3. Then there’s this from chapter 22
> <https://archive.org/details/historyofdecline54gibb/page/274/mode/2up>.
> He uses dots (ellipses) in other places, so I don’t know if the
> asterisks are significant. Maybe it means there was vertical elision
> /and/ part of that line is missing? How should I format this?

That *might* mean that the source text was corrupt or missing. Again, I
would see if I can find that source somewhere online. That will clarify
whether it's an elision or whether it's missing text.

> 4. This, from a little later in chapter 22
> <https://archive.org/details/historyofdecline54gibb/page/284/mode/2up>,
> appears quite a bit. A number of dashes at the beginning of a line, but
> the line starts with a capital, which makes whether there is text prior
> questionable. And indenting is used in other places (e.g. Eunuchum in #1
> above). Are these dashes filling in for words, or just spacing, or what?
> And how do we want to format it?

I don't know. Again, you could try searching for it. Also, remember our
discussion on Wheatley earlier. This is a vast volume going through many
editions and back then typesetters had a lot of freedom to change things
as they wanted, or depending on if they ran out of characters that day.
I wouldn't be surprised to find a lot of inconsistencies in little
things like this, considering the size and complexity of the work. The
best approach would probably be to first try to find the source online,
then if you can't find it, compare different editions of Gibbon to see
if they differ, and maybe we can get a clue from that.

David Grigg

unread,
Aug 31, 2020, 7:01:47 PM8/31/20
to standar...@googlegroups.com
The chapter 22 one is a partial quote from Paradise Lost: The full line reads:

"The guarded Gold: So eagerly the fiend"

So the number of dashes doesn't correlate with the omitted words.


As for renumber endnotes it's clear we need to accommodate issues like this one, and I can come up with a plan once I'm certain what our preference is as to where and how the endnotes to endnotes should be treated
--
You received this message because you are subscribed to the Google Groups "Standard Ebooks" group.
To unsubscribe from this group and stop receiving emails from it, send an email to standardebook...@googlegroups.com.

Vince

unread,
Aug 31, 2020, 11:23:44 PM8/31/20
to Standard Ebooks
Very good, thanks, David. So it’s using dashes just represent that it’s starting in the middle of the line. That sounds like a job for a leading ellipsis. So, ellipses for the leading and trailing dashes, and vertical elision for the full-line asterisks and dashes. That’s what I suspected, but good to have it confirmed.

Re the editions, I did look at the scans of several editions when I started this before. The text itself obviously is mostly the same (but see below), so it comes down to the notes. The PG one is a good balance, and they’re written by people as familiar with the source material (and languages) as Gibbon was. One of the other editions I looked at had a few more notes by a new editor (William Smith), but the existing notes were pretty much the same, although sometimes with slightly different formatting, e.g. sometimes the citations were presented a little differently. There weren’t a lot of notes by Smith, but if we wanted to add those, we could. But, really, I think this job is big enough already. :)

I also found a really nice, clean scan that talked about the difference between the first and second editions, and the small changes that improved clarity and understanding; the PG edition is the improved one. Those scans, however, doesn’t have the notes of one of the editors of the PG edition, and there are over a thousand of them and they're worth keeping.

It would appear either that the scans I found aren’t exactly the ones PG used, though. For example, there are instances of the “,)” to close a comment in the transcription, while the same one in the scans I have are the more modern “),”. I will obviously update those to match the scans.

Bottom line, I think the edition PG used is a good one.

Vince

unread,
Sep 5, 2020, 2:15:54 PM9/5/20
to Standard Ebooks
Gibbon uses roman numerals when referring to parts of other books. This can look like a typo in some places (e.g. ixth looks like it’s missing an s).
Read and feel the xxiid book of the Iliad

in the viiith Satire of Juvenal

which seems to result from the comparison of the vth and xviiith titles of the ixth book.

What do we want to do with those? Leave them as is? Change them to decimal numerals? If leaving them roman, should I add an ’n’ to xxiid like we would if it was 22d? 

Vince

unread,
Sep 5, 2020, 2:21:18 PM9/5/20
to Standard Ebooks
Sorry, hit Send too soon.

Also, he usually refers to counts with a lowercase count, e.g. the count de Buat, count de Boulainvilliers, etc. I’ve always seen the Count capitalized; leave them or change them? (Interestingly, I have seen a couple of places where he capitalizes the Count, but lowercase is the rule.)

And, he doesn’t have some silent e’s in some names, e.g. Jerom vs Jerome (with or without the St.), Shakspeare vs Shakespeare. Leave them or change them to modern spelling?

Alex Cabal

unread,
Sep 5, 2020, 6:45:33 PM9/5/20
to standar...@googlegroups.com
I would capitalize them, which will make it clear that they're Roman.
They would typically be capitalized anyway, not sure why he didn't.

On 9/5/20 1:15 PM, Vince wrote:
> Gibbon uses roman numerals when referring to parts of other books. This
> can look like a typo in some places (e.g. ixth looks like it’s missing
> an /s/).
>> Read and feel the xxiid book of the Iliad
>>
>> in the viiith Satire of Juvenal
>>
>> which seems to result from the comparison of the vth and xviiith
>> titles of the ixth book.
>
> What do we want to do with those? Leave them as is? Change them to
> decimal numerals? If leaving them roman, should I add an ’n’ to xxiid
> like we would if it was 22d? 
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/DA0D25E1-2025-4DEA-8D6E-380E03E901B3%40letterboxes.org
> <https://groups.google.com/d/msgid/standardebooks/DA0D25E1-2025-4DEA-8D6E-380E03E901B3%40letterboxes.org?utm_medium=email&utm_source=footer>.

Alex Cabal

unread,
Sep 5, 2020, 6:46:23 PM9/5/20
to standar...@googlegroups.com
Aristrocratic titles are complicated. I would leave them as-is unless
he's inconsistent. Re. spelling, Shakespeare is fixed by
modernize-spelling for sure. You can add an E to Jerome.

On 9/5/20 1:21 PM, Vince wrote:
> Sorry, hit Send too soon.
>
> Also, he usually refers to counts with a lowercase count, e.g. the count
> de Buat, count de Boulainvilliers, etc. I’ve always seen the Count
> capitalized; leave them or change them? (Interestingly, I have seen a
> couple of places where he capitalizes the Count, but lowercase is the rule.)
>
> And, he doesn’t have some silent e’s in some names, e.g. Jerom vs Jerome
> (with or without the St.), Shakspeare vs Shakespeare. Leave them or
> change them to modern spelling?
>
>
>> On Sep 5, 2020, at 1:15 PM, Vince <al...@letterboxes.org
>> <mailto:al...@letterboxes.org>> wrote:
>>
>> Gibbon uses roman numerals when referring to parts of other books.
>> This can look like a typo in some places (e.g. ixth looks like it’s
>> missing an /s/).
>>> Read and feel the xxiid book of the Iliad
>>>
>>> in the viiith Satire of Juvenal
>>>
>>> which seems to result from the comparison of the vth and xviiith
>>> titles of the ixth book.
>>
>> What do we want to do with those? Leave them as is? Change them to
>> decimal numerals? If leaving them roman, should I add an ’n’ to xxiid
>> like we would if it was 22d? 
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/0E302AD8-37C5-4349-B70A-CE11EB0B226E%40letterboxes.org
> <https://groups.google.com/d/msgid/standardebooks/0E302AD8-37C5-4349-B70A-CE11EB0B226E%40letterboxes.org?utm_medium=email&utm_source=footer>.

Vince Rice

unread,
Sep 5, 2020, 7:23:29 PM9/5/20
to Standard Ebooks
As I said (…I have seen…), he’s inconsistent. He’s definitely majority lowercase, but he also occasionally drops in a Count. I don’t see any surrounding contextual difference that would cause the occasional Count. (Wording’s not different, he’s not quoting someone else, etc.) I’m happy to leave them, but he’s not completely consistent.

Vince

unread,
Sep 5, 2020, 7:46:51 PM9/5/20
to Standard Ebooks
The only time he capitalizes roman characters is in names, e.g. James II.
All of his other roman numerals, whether they be referring to centuries (viith) or ages (xiith year) or parts of other books (as below) or whatever else, is lowercase.

This includes all of the book citations, e.g.
D'Anville, Etats de l'Europe, p. 156. Pagi, tom. iii p. 174.
Actio ii. Orat. 4
Plin. Hist. Nat. xxx 1
Histoire des Celtes, tom. vi p. 230—252
Pocock (Specimen, p. 158-161) and Casiri (Bibliot. Hispano-Arabica, tom. i p. 48, 84, etc., 119, tom. ii p. 17, etc.
(I’ve already removed the trailing periods that were on all the roman numbers.)

I’m guessing the use of lowercase in citations is some kind of a standard? The citations I’ve seen for ancient works use lowercase roman numerals in the citations. For example, the Folio Press edition of Decline and Fall I have is based on a 1910 text; they’ve turned textual romans into written decimals, e.g. “the thirteenth century" instead of “the xiiith century,” but in the above Pocock citation they kept the lowercase i and ii, and the same throughout.

Vince

unread,
Sep 5, 2020, 7:48:08 PM9/5/20
to Standard Ebooks
Oh, and what about the XIId—do you want to leave that, or change it to XIInd?

Alex Cabal

unread,
Sep 5, 2020, 10:57:07 PM9/5/20
to standar...@googlegroups.com
Use nd for ordinals

On 9/5/20 6:48 PM, Vince wrote:
> Oh, and what about the XIId—do you want to leave that, or change it to
> XIInd?
>
>
>> On Sep 5, 2020, at 6:46 PM, Vince <al...@letterboxes.org
>> <mailto:al...@letterboxes.org>> wrote:
>>
>> The only time he capitalizes roman characters is in names, e.g. James II.
>> All of his other roman numerals, whether they be referring to
>> centuries (viith) or ages (xiith year) or parts of other books (as
>> below) or whatever else, is lowercase.
>>
>> This includes all of the book citations, e.g.
>>
>> D'Anville, Etats de l'Europe, p. 156. Pagi, tom. iii p. 174.
>> Actio ii. Orat. 4
>> Plin. Hist. Nat. xxx 1
>> Histoire des Celtes, tom. vi p. 230—252
>> Pocock (Specimen, p. 158-161) and Casiri (Bibliot.
>> Hispano-Arabica, tom. i p. 48, 84, etc., 119, tom. ii p. 17, etc.
>>
>> (I’ve already removed the trailing periods that were on all the roman
>> numbers.)
>>
>> I’m guessing the use of lowercase in citations is some kind of a
>> standard? The citations I’ve seen for ancient works use lowercase
>> roman numerals in the citations. For example, the Folio Press edition
>> of /Decline and Fall/ I have is based on a 1910 text; they’ve turned
>> textual romans into written decimals, e.g. “the thirteenth century"
>> instead of “the xiiith century,” but in the above Pocock citation they
>> kept the lowercase i and ii, and the same throughout.
>>
>>> On Sep 5, 2020, at 5:45 PM, Alex Cabal <al...@standardebooks.org
>>> <mailto:al...@standardebooks.org>> wrote:
>>>
>>> I would capitalize them, which will make it clear that they're Roman.
>>> They would typically be capitalized anyway, not sure why he didn't.
>>>
>>> On 9/5/20 1:15 PM, Vince wrote:
>>>> Gibbon uses roman numerals when referring to parts of other books. This
>>>> can look like a typo in some places (e.g. ixth looks like it’s missing
>>>> an /s/).
>>>>> Read and feel the xxiid book of the Iliad
>>>>>
>>>>> in the viiith Satire of Juvenal
>>>>>
>>>>> which seems to result from the comparison of the vth and xviiith
>>>>> titles of the ixth book.
>>>>
>>>> What do we want to do with those? Leave them as is? Change them to
>>>> decimal numerals? If leaving them roman, should I add an ’n’ to xxiid
>>>> like we would if it was 22d? 
>>
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/A54AD9F1-6127-4292-A573-F1C07217E847%40letterboxes.org
> <https://groups.google.com/d/msgid/standardebooks/A54AD9F1-6127-4292-A573-F1C07217E847%40letterboxes.org?utm_medium=email&utm_source=footer>.

Alex Cabal

unread,
Sep 5, 2020, 10:59:49 PM9/5/20
to standar...@googlegroups.com
Typically in older texts like this, larger divisions like books or acts
get uppercase Roman, and smaller divisions like chapters get lowercase.
We prefer uppercase in almost all cases. I have never seen a work where
it's always all lowercase. I would say that's unusual. Then again it's
not like there was ever some kind of standard for this sort of thing
back then, and DaF went through countless editions over the years.


On 9/5/20 6:46 PM, Vince wrote:
> The only time he capitalizes roman characters is in names, e.g. James II.
> All of his other roman numerals, whether they be referring to centuries
> (viith) or ages (xiith year) or parts of other books (as below) or
> whatever else, is lowercase.
>
> This includes all of the book citations, e.g.
>
> D'Anville, Etats de l'Europe, p. 156. Pagi, tom. iii p. 174.
> Actio ii. Orat. 4
> Plin. Hist. Nat. xxx 1
> Histoire des Celtes, tom. vi p. 230—252
> Pocock (Specimen, p. 158-161) and Casiri (Bibliot. Hispano-Arabica,
> tom. i p. 48, 84, etc., 119, tom. ii p. 17, etc.
>
> (I’ve already removed the trailing periods that were on all the roman
> numbers.)
>
> I’m guessing the use of lowercase in citations is some kind of a
> standard? The citations I’ve seen for ancient works use lowercase roman
> numerals in the citations. For example, the Folio Press edition of
> /Decline and Fall/ I have is based on a 1910 text; they’ve turned
> textual romans into written decimals, e.g. “the thirteenth century"
> instead of “the xiiith century,” but in the above Pocock citation they
> kept the lowercase i and ii, and the same throughout.
>
>> On Sep 5, 2020, at 5:45 PM, Alex Cabal <al...@standardebooks.org
>> <mailto:al...@standardebooks.org>> wrote:
>>
>> I would capitalize them, which will make it clear that they're Roman.
>> They would typically be capitalized anyway, not sure why he didn't.
>>
>> On 9/5/20 1:15 PM, Vince wrote:
>>> Gibbon uses roman numerals when referring to parts of other books. This
>>> can look like a typo in some places (e.g. ixth looks like it’s missing
>>> an /s/).
>>>> Read and feel the xxiid book of the Iliad
>>>>
>>>> in the viiith Satire of Juvenal
>>>>
>>>> which seems to result from the comparison of the vth and xviiith
>>>> titles of the ixth book.
>>>
>>> What do we want to do with those? Leave them as is? Change them to
>>> decimal numerals? If leaving them roman, should I add an ’n’ to xxiid
>>> like we would if it was 22d? 
>
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/D3511743-0FEF-412E-9A54-8E12A25BD99C%40letterboxes.org
> <https://groups.google.com/d/msgid/standardebooks/D3511743-0FEF-412E-9A54-8E12A25BD99C%40letterboxes.org?utm_medium=email&utm_source=footer>.

Alex Cabal

unread,
Sep 5, 2020, 11:00:41 PM9/5/20
to standar...@googlegroups.com
OK. Let's uppercase it then. I think that that is clearer when referring
to a person without a name ("... the Count did this..."), and if we're
going to do that, we may as well be consistent and uppercase all of them.

On 9/5/20 6:23 PM, Vince Rice wrote:
> As I said (…I have seen…), he’s inconsistent. He’s definitely
> /majority/ lowercase, but he also occasionally drops in a Count. I don’t
> see any surrounding contextual difference that would cause the
> occasional Count. (Wording’s not different, he’s not quoting someone
> else, etc.) I’m happy to leave them, but he’s not completely consistent.
>
>
>> On Sep 5, 2020, at 5:46 PM, Alex Cabal <al...@standardebooks.org
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/DEF9557C-139E-469B-B7C6-BC3601AD621C%40therices.name
> <https://groups.google.com/d/msgid/standardebooks/DEF9557C-139E-469B-B7C6-BC3601AD621C%40therices.name?utm_medium=email&utm_source=footer>.

Vince Rice

unread,
Sep 6, 2020, 12:38:24 AM9/6/20
to standar...@googlegroups.com
All I can say is I’ve looked heavily at scans of six different editions, over a period of 60 years or so (mid 19th thru early 20th century), plus the printed edition I have whose text is early 20th but printing is late 20th century, and they are all completely consistent—they only use lowercase in the citations.

By “we prefer uppercase in almost all cases", you mean you want all the roman numerals uppercased, correct?

Alex Cabal

unread,
Sep 6, 2020, 11:40:06 AM9/6/20
to standar...@googlegroups.com
Yes. Should be easily scriptable. Semanticate will mark almost all
Roman, and you can write a short Python script to use the
se.formatting.EasyXhtmlTree class to select all roman and replace with
the uppercase text.

I had to look up "tom." which is short for "Tomus" i.e. Tome. I wonder
why he just didn't write "Tome" or "Vol."? Anyway, interesting to learn.

On 9/5/20 11:38 PM, Vince Rice wrote:
> All I can say is I’ve looked heavily at scans of six different editions,
> over a period of 60 years or so (mid 19th thru early 20th century), plus
> the printed edition I have whose text is early 20th but printing is late
> 20th century, and they are all completely consistent—they only use
> lowercase in the citations.
>
> By “we prefer uppercase in almost all cases", you mean you want
> /all/ the roman numerals uppercased, correct?
>
>> On Sep 5, 2020, at 9:59 PM, Alex Cabal <al...@standardebooks.org
> --
> You received this message because you are subscribed to the Google
> Groups "Standard Ebooks" group.
> To unsubscribe from this group and stop receiving emails from it, send
> an email to standardebook...@googlegroups.com
> <mailto:standardebook...@googlegroups.com>.
> To view this discussion on the web visit
> https://groups.google.com/d/msgid/standardebooks/BA372933-0FAE-43A3-BC01-2C197931C7BB%40therices.name
> <https://groups.google.com/d/msgid/standardebooks/BA372933-0FAE-43A3-BC01-2C197931C7BB%40therices.name?utm_medium=email&utm_source=footer>.

Vince Rice

unread,
Sep 6, 2020, 11:54:25 AM9/6/20
to Standard Ebooks
That’s way too much work. :) A regex will take care of it in a few seconds.

Are you sure you don’t want to modernize the textual ones, though? Talking about the XIth century or the VIIIth satire or the XXXth volume or the XVIth letter really breaks up the reading flow; modern brains have to pause to translate those to decimal numbers. Since we modernize spelling (and I’m all for that), it seems natural to modernize those as well. And by modernize I mean spell them out, i.e. twentieth century and eighth satire and sixteenth letter, not 20th and 8th and 16th.

Alex Cabal

unread,
Sep 6, 2020, 12:17:17 PM9/6/20
to standar...@googlegroups.com
Sure, that's fine

Vince

unread,
Sep 6, 2020, 12:28:15 PM9/6/20
to Standard Ebooks
Great, thanks!

Vince

unread,
Sep 7, 2020, 4:05:32 PM9/7/20
to Standard Ebooks
How exact does a reference to a book need to be to put book semantics on it?

For example:
Ockley, Hist. of Sarac. i 232.

This is referring to Simon Ockley’s The History of the Saracens, so he dropped both of the the's. And, obviously, the words are abbreviated. (Gibbon loves abbreviations, which I guess makes sense when you’re writing a million-and-a-half words by hand.) I’m assuming the abbreviated words themselves don’t matter, but I didn’t know about removing the the’s.

In other places, he refers to books in the endnotes by an abbreviated name, especially when the name of the book is long. For example, Karl Guichard’s Mémoires militaires sur les Grecs et les Romains he sometimes lists in the endnotes as
M. Guichard, Mémoires Militaires, tom. i c. 4.

Should I tag instances like that as books or no? (In other places he shortens things even further, e.g. Mémoires de Guichard; those I assume those should not be tagged.)

These are just examples, he does this all over the place (remember, 8500+ endnotes). I’m just trying to get a sense for what should be tagged and what shouldn’t. Because, after I spend the next three months going through each and every one of these endnotes, I’m not interested in doing it again. :)

Thus, in general, do you only want it tagged if it’s exactly the full name, or is some shortening OK, and if so, how much shortening?

Vince

unread,
Sep 7, 2020, 4:56:27 PM9/7/20
to standar...@googlegroups.com
As another example of shortening, a reference to Nicolas Bergier’s Histoire des Grands Chemins de l'Empire Romain looks like this:
Bergier, Hist. des Grands Chemins, l. iii c. 1, 2, 3, 4.