Index talk:Plomer Dictionary of the Booksellers and Printers 1907.djvu

Transcription errors

A few patterns I've noticed in the course of proofreading:

• The most common error is omitting the full stop after St, although this makes no difference to the sense.
• The most serious error is reading 3 for 2, sometimes obvious in dates such as the 39th of a month, sometimes not obvious in a year or a reference.
• There are also many examples of reading m for rn, as in 'Comhill' for Cornhill.

Addedentry (talk) 21:19, 14 August 2017 (UTC)

Yep, had picked up numbers of those though depending on the user's proofing font I can understand how they are missed. I will have a look at the dates again, to see what we can do for sanity checks. — billinghurst sDrewth 03:57, 15 August 2017 (UTC)

Noting bot tasks (as regex statements)

• St[^\.]_ to St._ though will want to visually check these for false positives
• Cormhill to Cornhill

Signature marks

After proofreading several pages, I decided to start transcribing signature marks in the footer (e.g. page 145, which is K2).

Addedentry (talk) 21:19, 14 August 2017 (UTC)

Up to you, that is usually outside of the proofreading component of the work, as they are printer's marks. — billinghurst sDrewth 03:52, 15 August 2017 (UTC)

Ellipses

Ellipses in this text are usually printed as four dots (....) which could be generated with the markup {{...|4}}: . . . . However, Template:Ellipsis is 'not recommended'. Instead, the Wikisource style guide says "Ellipses of omission should be entered as the actual character (i.e. …) without surrounding spaces". Therefore, I have followed one of the first transcribers in using … to generate three dots: …

Addedentry (talk) 21:19, 14 August 2017 (UTC)

We should use a proper ellipsis either the direct unicode or … as per the style guide. I will bot replace the remainder at completion. — billinghurst sDrewth 03:52, 15 August 2017 (UTC)

Small caps

Forenames are printed with small capitals. Should we follow this? e.g. the first entry is currently ADAMS (CHARLES), bookseller in London but could more faithfully be rendered ADAMS (charles), bookseller in London.

Addedentry (talk) 21:23, 14 August 2017 (UTC)

I don't think that we need to blindly follow the work in that component, and the reality is that it is not charles. To also note that one has to be alert around the use of {{smallcaps}} and {{smaller}}}. My reason is that if you are to do a copy and paste of the text, you want for it to make sense in reproduction, and not just in the look. So examples of abbreviations for a.m. and a.d. should be converted with small-caps, though for something like the abbreviation for Prerogative Court of Canterbury is P.C.C. so should be with smaller. I have yet to identify the other regular abbreviations and will do that later. I think we have B.M. too. Happy for others to be identified.

I will bot a tidy up to update smallcaps to smaller.

We should add some notes about the regular abbreviations as a tidy up too. — billinghurst sDrewth 03:50, 15 August 2017 (UTC)

Abbreviations are at Page:Plomer Dictionary of the Booksellers and Printers 1907.djvu/30 and most of them should be "smaller" — billinghurst sDrewth 13:39, 15 August 2017 (UTC)

Will run a bot through

@Addedentry: Thanks for your comments. I will answer each above and add my own. Once we have finished the work, I will look to run a bot through to cleanup to a consistency. — billinghurst sDrewth 03:39, 15 August 2017 (UTC)

Thank you! In particular for your reasoning about smallcaps versus smaller. Much appreciated. Addedentry (talk) 17:56, 15 August 2017 (UTC)

Rules used

• <Typo word="<enter a name>" find=" St " replace=" St. " />
• <Typo word="<enter a name>" find="Cor?mhill" replace="Cornhill" />
• <Typo word="<enter a name>" find="{{...\|\d}}" replace="&hellip;" />
• <Typo word="<enter a name>" find="(\d{3}){{[sS]frac nobar" replace="{{sfrac nobar|$1" />
• <Typo word="<enter a name>" find="{{smallcaps\|(D.N.B.|B.M.|P.C.C.|P.R.O.|T.C.)}}" replace="{{smaller|{{subst:UC:$1}}}}" />

nowrap "no bar" dates

I have seen a couple of examples where the use of Template:Sfrac nobar has the fraction separated from the first part of the date when transcluded. Need to get all those dates and put within {{nowrap}}. No need to worry about doing it ahead of time, writing a bot regex is easy.— billinghurst sDrewth 04:08, 15 August 2017 (UTC)

Regex to test

• (\d{3}{{[sS]frac nobar[^}]+?}}) -> {{nowrap|$1}}

Better solution is to add it as part of Template:Sfrac nobar which we have done, though that means we then need to bring the advanced numbers inside the template. — billinghurst sDrewth 03:45, 17 August 2017 (UTC)

• (\d){3}{{[sS]frac nobar -> {{sfrac nobar|$1