Wikisource talk:WikiProject Ackermann’s Repository of Arts

From Wikisource
Latest comment: 9 years ago by Laura1822 in topic Updates
Jump to navigation Jump to search

This project is in its very early stages and I have lots of questions. Please give me the benefit of your experience!

Goals[edit]

My goal is to create a complete set of Ackermann's Repository for use here and on other wiki-projects. I am very impressed with WikiProject Popular Science Monthly and will use that as a model going forward. It is a pilot project for me and I would like to recruit people from other wikiprojects to come here to help in proofreading and processing (or, indeed, to take over).

I would eventually like to do other similar projects with other late 18th/early 19th century periodicals, particuarly fashion journals, but also Gentleman's Magazine and the Annual Register. The former has had very little work done, only piecemeal, though most of the volumes are now available on Google Books.

djvu and image file quality[edit]

The underlying jp2 images of the djvu files at Internet Archive are not great, at least not for volume 1 (I think/hope they improve in later volumes), primarily because some of the images are unfocused. Here is an example of a plate where this is particularly noticeable, and here is an example of text that is not OCR-able, even from the underlying jp2 image.

The images at Bunka Gakuen (see below for link and instructions to find the Repository) are far, far superior. They are unfortunately not suitable for direct upload to Commons or to compile into djvus to upload to IA (then use here) without some image pre-processing. The most significant problems are that they are in two-page format so they need to be split, and also cropped. The latter must be done carefully for the plates because these images preserve the printer's attributions at the tops and bottoms of the plates, which provides important information that is frequently unavailable on images from other sources: for example, compare the January 1809 plate linked above to the Bunka Gakuen image.

Such pre-processing and uploading to both Commons and IA could perhaps be accomplished via automation, but Ineuw suggested that I might simply write to them and ask them to do it! What a great idea! With the new GLAM tools, they might upload not just the Repository but all of their books, some of which are very rare. So I will look into that and seek help for it at Commons.

In the meantime, we can use the Bunka Gakuen images to assist in our tasks here, while using the existing IA djvu set for master processing. Comments and suggestions welcome.

Style Guide[edit]

See working style manual at User:Laura1822/Ackermanns. (In development and subject to change; see Questions section)

  • Preserve all spelling and punctuation as on the page. Use {{sic}} and {{SIC}} as appropriate (the latter to mark actual printer errors, the former for everything else).
  • Use {{bar|2}} instead of emdashes.
  • Use “paired quotation marks” or “curly quotation marks,” which can be accessed from the Symbols dropdown on the editing toolbar. Also ‘single but paired quotation marks’ and curly apostrophes. You can add a toolbar button that converts all of the quotation marks, apostrophes, and twin-hyphens on the page to their curly or dashy forms with a single click. To use it, add the following line to your common.js: importScript('User:Samwilson/CurlyQuotes.js');
  • For ellipses, use ". . .” (periods/stops with spaces in between) instead of "...” (no spaces) or "…" (character).
  • For all plates, woodcuts, graphics, title pages, etc., leave the images in as close to an unedited state as possible. Fundamentally, the plates especially should have an "original" version on Commons, with minimal changes such as lossless rotation/straightening and minimal cropping, maintaining any printer's text (such as "Plate No. x" or the attribution along the bottom, if present in the jp2), with a jpg extension. (I feel very strongly about this: see my reasoning below.) A version with color enhancements (including an edited background or brightness adjustments) can be uploaded to Commons separately as another choice for commons users, but not for primary use.
    • When uploading images, please enter the date with month, e.g., "1809-03" for "March 1809," in the image's description; please also set the default sort the same way so that the plates will sort correctly.
    • Note on the fabric swatches: despite their appearance, they were not cover pages. They were inserted near the end of each issue and text describing them was included on a nearby page, like the other plates. The issues had no cover page, just the contents page.
    • Include text printed on plates, such as "Plate No. x," caption (usually at top right or left), and printer's attribution, following the formatting as closely as possible. For example:

{{center|{{sc|Walking Dress.}}}}

{{center|{{sc|''For N}}{{sup|o 1}} of {{sc|{{SIC|Ackrmann’s|Ackermann’s}} Repository}} of {{sc|Arts &{{sup|c}}}} Pub. Jan{{sup|y}} 1809: 101 Strand {{sc|London.''}}}}

to create:

Walking Dress.

For No 1 of Ackrmann’s Repository of Arts &c Pub. Jany 1809: 101 Strand London.

Note: this is from a plate where the printer's attribution was cut off from the plate in the version at IA, while the image at Bunka Gakuen showed it clearly.

Future development: provide a subpage with copy/paste for as many design elements such as this which repeat in every issue (noting that with the plates in particular there are often subtle variations).

Rationale for using unaltered plates[edit]

My reasoning can be summed up as follows:

  1. These plates are artwork, and should be treated as such.
  2. Removal of data is bad.

To elaborate further:

  1. Cleaning the background can inadvertently remove printer's marks such as plate numbers, which are important for people who study them. Examples: Title page (engraver's attribution removed), Plate 1a, Plate 2, and Plate 4 (which all have the plate number removed), and the page with handwriting near the beginning of the book, which had the writer's rules (lines) removed.
  2. On Commons, it is standard practice to upload edited images separately, to preserve the originals (even when uploading a new version over an old one, the original is archived in the file's history). That way the user can choose which version suits his needs.
  3. Keeping the plates as close to the originals as possible is consistent with the WS practice of preserving printers' errors in texts. We aren't supposed to be editing these works: we are supposed to be presenting them as the printer did.
  4. Minimal editing means less work!  :)
  5. I honestly don't see any improvement, aesthetic or otherwise, to editing these plates.
    1. It is not necessary for legibility, and makes legibility impossible if text is actually removed.
    2. I don't see how "the paper was originally white" (even if true) is relevant.
    3. For the woodcuts that are included within the text, I can probably be persuaded that creating a "clean" version is valuable and could be useful to future users; however, I still don't think it necessarily improves them aesthetically and don't see why anyone should go to all the trouble to do it. Perhaps engineers would disagree.  :)

Ineuw, please don't take this personally, or think that I don't appreciate the work you've done on these images. This project is a huge learning curve for me and I am sure that there must be reasons why editing the images is standard practice. Since I've refrained from stating my thinking before, we haven't had much of a chance to discuss it, and perhaps that was an error on my part. If so, I apologize. You have given freely of your time to me and to this project, and I am immensely grateful. I hope you'll continue to mentor me and to consider this project worthwhile. Laura1822 (talk) 16:37, 7 September 2014 (UTC)Reply

Questions[edit]

This project is in its very early stages and I have lots of questions. Please give me the benefit of your experience!

  • What is the consensus for best practices regarding the printer's attributions and plate numbers printed on the face of the plate? Should they be transcribed and included on the page?

Reply: We always type the caption on the page and don't use the embedded caption in the image, for clarity and consistency. You can see this in any project on WS which has images. If someone did not follow this rule, it is wrong. As mentioned before, User:Hesperian is a good guide in these matters.

I was going to ping him. I am asking because I have seen a couple of plates, including the Title page, marked as proofread without the text having been transcribed.
  • Should pages/paragraphs be formatted with {{text-indent}} or some other template?

Reply: No. We don't indent paragraphs.

Regarding the marriages and obituaries, I have some questions regarding best practices:

  • Is it okay to use {{bar}} i.e., {{bar|2}}, a bar of 2 ems, i.e., twice a long as a plain emdash? (This more closely approximates the actual printed page, and makes it much easier to read, which is perhaps why printers did it.)

Reply: You can use the template, but tapping the mdash button twice on your toolbar produces the same result. In either case you need to add the surrounding spaces for line wrap.

I don't see the same result. I see two emdashes with a small space in between, not a solid line. That is why I went looking for {{bar}}, which provides a solid line.
  • If using bars, is there any way besides adding spaces to allow wrapping before or after the bar?

Reply: No. you have to add the surrounding spaces yourself. ——

The printer generally didn't use spaces, so adding spaces changes the look.
  • Ditto emdashes: Does wrapping require usage of {{}} (renders an mdash—with hairspaces)?

Reply: —— or {{tl|—}} won't work with two consecutive mdashes. Just add the spaces to facilitate proper line wrap.

Do the hairspaces add linewrap functionality?
  • What is the practice where the printer made an error in punctuation, deviating from the style he has set forth? For example, this is the standard formatting:

Buckinghamshire.—Married.] J. Bates, esq. of Wycombe Marsh, to Mrs. Goodwin.—Mr. John Bailey, to Miss Elizabeth Starkey, of Poole.—Mr. J. Tyler, to Miss M. Foster. Died.] At Shalstone, the Rev. J. P. Javoise, LL. D.—Mr. Wm. Thorpe, of Aylesbury.

Reply: We keep all formatting errors of the original. No changes.

But what if the printer leaves out a period, e.g.,

BuckinghamshireMarried.] J. Bates, esq. . .

or

Mr. John Bailey, to Miss Elizabeth Starkey, of Poole—Mr. J. Tyler, to Miss M. Foster.

or what if he leaves out the bracket or adds or substitutes a dash:

Died.—The Rev. T. Holland, M. A. late of Thaxted.

Reply: We keep all formatting errors of the original. No changes. We leave out the period as in the original - all typos we just replicate as is..

Perfect. But I would include spaces in that, as above. Sometimes the printer did include spaces, but usually he didn't.
  • Should these errors be fixed? If they should remain, should they be tagged with something like {{SIC}}?

Reply: Personally, I only use {{SIC}} for misspelled words to aid the reader and to bring it to the attention of another editor that it has been noticed (less subsequent editing for validators). For unknown words or spelling, placing the {{sic}} (lowercase sic) in the proofread text indicates to another editor that the error was noticed.

Oh, okay. I thought {{SIC}} had superseded {{sic}}, but I was confusing it with something else.
  • For comparison, here is a set with {{bar|2}} substituted for emdashes:

Bedfordshire.——Married.] G. Rickards, esq. to Miss Frances Kettilby, of Sutton.——Mr. James Pryor, to Miss Elizabeth Gutteridge.——At Leighton-Buzzard, Henry Newland, esq. to Mrs. Mary Gurney.

Died.] At Gorhambury, in the 62d year of his age, Viscount Grimston.——At Tempsford, Richard Newman Harding Newman, esq.——The Rev. Dr. Shepherd, archdeacon of Bedford.——Miss Eliza Jane Alston, of Odell Castle.——The Rev. Edward Bunting, of Yelden.

Reply: Again when dealing with double emdashes, just surround it with space as mentioned above. Using a template like {{bar|2}} —— has no advantage over —— as both are two mdashes.

Finally, If you want a special template which generates two mdashes surrounded by hairline space let me know.

Again, as above, I see two different things (perhaps this is a reflection of font choices?), and I think there is an advantage to having a solid line with no breaks. I also think that spacing is important and should reflect exactly what the printer did, sometimes present surrounding the dashes, usually not.
Hairspaces are a reasonable compromise, but only if they add line wrapping functionality. If they do, then I would love a new template as you suggest with hairspaces.
Edited to add: I may be wrong about all of this, as it seems that the bar|2 with no spaces around it is allowing line wrapping after all. The reason I brought it up is because I thought it was preventing line wrapping! Please verify the functionality for me.

If you have any other questions, please ask. — Ineuw talk 02:25, 2 September 2014 (UTC)Reply

Thank you! This is a great start!! Laura1822 (talk) 13:06, 2 September 2014 (UTC)Reply

Conversations copied from other talk pages[edit]

Initial discussion from User talk:Ineuw re: Ackermann's Repository[edit]

Conversation moved from User talk:Ineuw

Thanks for creating those subcategories at Commons. I added to your table and copied the relevant fashion plates into their respective subcategories.

I am going to have to rename the djvu file to include the series, spell out "Volume," etc. I think. What do you suggest?

I think perhaps this periodical deserves a Project or a Portal (what's the difference?), especially since I would like to expand it to include other periodicals from the period, especially the fashion journals. I started with Ackermann's because there is a complete run of it on IA, but the images aren't all that great (some are better than others). What I'd really love is to find someone who would collect all the fashion journals from this period on Bunka Gakuen: these need to be collected, split, cropped, assembled into djvus, and then uploaded at IA and at Commons (including the cropped high-res images separately)--all of which are tasks better suited to automation than to people (or to me personally, anyway). The images are so much better than any other source, though.

Gentleman's Magazine needs a Project/Portal too, for some recruitment and organization. What do you think? Laura1822 (talk) 01:28, 31 August 2014 (UTC)Reply

@Laura1822: Each of the above are major undertakings and I advise you to focus on a single project that you consider to be the most important, to get used to the various aspects and gain an overall picture of what needs to be done just for a single project.
To better answer your questions, permit me to outline the steps needed to bring a project to fruition:
  1. DjVu file naming must be done on the commons, where the file is sourced. Once it's done there, then the change is implemented here. I can do both moves, Also, I just noticed on the Commons that the Ackerman publications repeat the volume numbers and this must be dealt with before any changes.
  2. Personally, I prefer a clear, concise, and as short as name possible. I recommend the following because there will be no collision between early and later repeat volume numbers. Ackermann’s Repository of Arts 1809 Volume 01.djvu, Ackermann’s Repository of Arts 1809 Volume 02.djvu . . . . . etc. Months are not relevant in the .djvu file name because it's one volume. We will deal with these in the main namespace.
  3. Projects deal with the organizational details, and they belong in the Wikisource: namespace as in Wikisource:WikiProject Popular Science Monthly. Sort of the planning area for the project.
  4. Portals are the final products' main namespace listings, organized according the U.S. Library of Congress classification.
  5. Images are .not djvu files, or pages, because they are very low quality. The method used is: we download the .jp2 file from IA - convert it to .jpg or .png formats, clean up the images and upload them to the Commons. I used my layout and organization from the commons:Category:Popular Science Monthly illustrations to organize the Ackerman project. Each volume category stores the book and the associated images.
  6. I already downloaded your Volume 1 .jp2 files and I can tell you that it contains 49 images. I will clean and upload these files but do not commit myself to more, as I have numerous projects of my own going on simultaneously.
  7. As the last note, I read your conversation on GO3's talk page regarding categorization, and I can help you organizing that as well. But not in this post. As I need some brain rest.

I hope this helps. — Ineuw talk 02:48, 31 August 2014 (UTC)Reply

Thank you very much, it helps a great deal. The Projects/Portals explanation makes perfect sense, but I hadn't seen it anywhere. I understand about djvu files vs. images, but the underlying jp2 images of this particular issue of the Repository are not great. Some of them appear to be unfocused. The images at Bunka Gakuen are far superior. (And they have complete runs of several periodicals not available on IA or anywhere else.) But I've been a little confused about one thing that your explanation seems to highlight: is it the practice here to always retrieve the jp2 files of images/plates and upload them separately? Is there any kind of automated system for that, a bot or something?

As for naming, I know that it is not the convention in serial cataloguing and management to include the months (sometimes not even the years). However, as a user of these publications, particularly fashion magazines and GM (regarding BMD and other event reporting), I have often been frustrated by that practice (which IMO is for the benefit of librarians, not users). Whether I'm looking up an obituary, historical meteorological data, or a fashion plate, the month is very relevant. On the fashion plates, it is frequently printed on the plate itself, e.g., "Fashions for May 1812," so if you want to locate the plate description for a plate you found somewhere else, the date is your first clue. You couldn't care less about the volume or series number; that's just a data chart you have to look up in worldcat that comes between you and what you need. With wiki and commons, those conventions are almost (but not quite) archaic. So that's why I want to include the months in the title. What's hard for me is breaking free of the habit of abbreviating everything.  :)

I have filemover status at commons too so that's not a problem, just have to decide.

I hear you about taking on too much at once. That's certainly a failing of mine. My rather nebulous far-in-the-future plans definitely included step 1 of working through at least most of volume 1 before trying to set up a project to use in recruitment, much less expanding to other periodicals. This is very much a pilot/learning project for me. But I do think that if it is well-managed and promoted, a Project could pull some new users to WS. Those discussions are for another day.

Thank you for your offer to take care of the plate uploads to commons for vol 1. And thanks for pointing me towards Popular Science: I've copied/pasted from some other magazines so far, but this looks like a better model. I appreciate all your help and suggestions. Laura1822 (talk) 12:19, 31 August 2014 (UTC)Reply

  • The naming and organization is entirely up to you. Although, naming the .djvu files in the Index namespace is not the same as naming in the final main namespace. We do have a certain amount of flexibility.
  • A portal is a subject related classification. There is no portal for Ackerman's. But there could be a classification for fashion, art etc. If there isn't already one. My work in PSM is distributed in numerous portals, a process I leave to others. Categorization for me, was far more important. Please see Wikisource:WikiProject Popular Science Monthly/Categories.
  • On the subject of categorization, I also had the same issues in Popular Science Monthly, (my "pet" project). I resolved it through organization of the project from the beginning to the end to get the whole picture, then went through the whole process - from the beginning to the end - with a single chapter.
  • I am curious as to where did you see .jp2 images? They don't exist on the Commons. Also, as far as I know, there is no method of cleaning up drawings automatically. Before you decide on image quality, let me look at the downloaded images. I will focus on those particular images that is of concern to you. Please post their link.
  • In the meantime, I looked at the Bunka Gakuen images of Volume 1. Please bear with me.Ineuw talk 18:37, 31 August 2014 (UTC)Reply
I'm talking about the underlying jp2s for the djvus from IA (the same ones you have already downloaded), sorry for the confusion. There is also a blog where someone worked on all the fashion plates (only) from the entire run from the original IA jp2s, but I don't know if her work is ideal for Commons because IIRC she reduced the sizes, and also she cropped them in such a way that the text was cut off. There is often a Plate number at the upper right or left, and a caption below, with printer's info below that which includes a date. (This is another reason why the Bunka Gakuen images are invaluable: they are uncropped.) Anyway, here is the blog post for 1828 (the last one) with a list of links to her posts for each year, for you to consult.
Here's the Bunka Gakuen image of the January 1809 fashion plates, so you can compare them to the (IMO) out-of-focus jp2 image. There are other pages where the fuzziness causes OCR havoc, such as page 125: compare the Bunka Gakuen image. (But for what it's worth, I've been trying to clean up this text image (crop, sharpen, 2-bit, etc.) and haven't had any luck with improved OCR.) I don't know if these problems are present in later volumes of the Repository at IA, though I suspect that volume 1 is the worst of the lot.
On a slightly different note, I'm not sure what to do here in WS with the image you uploaded at File:ARA 1809 V01 D066 Walking dress.jpg. How do I use it?
Your PSM work is terrific, very similar to what I want to do for Ackermann's Repository as well as other periodicals. I don't pretend to understand it yet, but will continue to study it. Thanks again for all your help. Laura1822 (talk) 12:46, 1 September 2014 (UTC)Reply
Some cross talk, I was posting on your page the very same observations as you made here.
About the image I uploaded, is only a sample and will post there additional variations. However, I will insert it for now where it belongs to demonstrate how we display it. — Ineuw talk 13:18, 1 September 2014 (UTC)Reply
Thanks for the comments about the PSM galleries, but many of my original PSM uploads are very poor and are being replaced as I come across them. I never worked on images before the PSM project, and learned here with some guidance.
Thanks so much for all your help! Everyone here has been so nice.  :) Laura1822 (talk) 13:30, 1 September 2014 (UTC)Reply

Help! You moved the page The Repository of Arts while I was working on its Talk page without even leaving a redirect. I don't mind, if you think it's better to use "Ackermann's Repository" as a more common title (rather than the official title), but not only did you misspell "Ackermann," but it looks like the page content I thought I had placed on the page, including categories, is gone. Why? Or am I misremembering what was on the page? And why no redirect? I read somewhere that using lots of redirects for title variations is proper here. Laura1822 (talk) 16:52, 1 September 2014 (UTC)Reply

I'm not really panicking, sorry if it sounds that way. I'll wait till you're finished with whatever you're doing (I made a copy of what I was working on for the talk page) and I see you've already corrected the spelling.  :) I was also about to rename the djvu file at commons, but I'll wait on that too till you're done. Laura1822 (talk) 16:57, 1 September 2014 (UTC)Reply

Additional discussion from User talk:Laura1822 re: File:ARA 1809 V01 D066 Walking dress.jpg[edit]

Copied from User talk:Laura1822

I cleaned up this first colored image. There are 30 color images in Volume 1. — Ineuw talk 03:01, 1 September 2014 (UTC)Reply

P.S: Some additional comments:

  • The Bunka Gakuen books are not from Internet Archives. I am convinced that they are originals from their library and exquisitely preserved.
  • The IA images are missing the accreditation at the bottom of the image.
  • I downloaded the same image from Bunka Gakuen, to compare and will upload it to the Commons.
  • The background should be white, the colors we see are of the acidified paper.
  • If you missed this In the Scriptorium, there is a very good chance that these images will also be there. — Ineuw talk 13:06, 1 September 2014 (UTC)Reply
@Ineuw: I'm sorry I wasn't clear. I never meant to imply that the BG images were from IA. I agree with you that they are originals. My original point was that I hope that there is an automated way to assemble the BG images into djvu files, upload them to IA, and work from them instead of the set currently at IA. But if the other volumes at IA are not as badly done as this one, perhaps it would be appropriate to get the plate images from BG and upload them to commons (hopefully with a bot), though they need some cropping (carefully, to preserve the texts at the edges of the images). I haven't figured out a way to do the cropping with a batch yet (I use Irfanview).
I saw the discussion about the 2.4 million images, and went straight to commons to look for Ackermann's plates (or any fashion plates), but couldn't find any. Doesn't mean they aren't there. Laura1822 (talk) 13:27, 1 September 2014 (UTC)Reply
  • My comment about the source of BG images was only my surprise exclamation. It was I who made the assumption because I didn't look at the BG images closely. Only after leaving the computer, it occurred to me that BG, being a fashion school in the 19th century, they own their own copies.
  • Am I to assume that your wish is to transfer the BG images from their website to IA? Why don't you consider asking BG to upload their digital copies of their books, pointing out their quality vs IA versions? IA then will spawn all kinds of formats within 24 hours.
  • I also use only Irfanview. But 49 images - some grayscale, don't require batch operations - also the images are not the same size. — Ineuw talk 14:28, 1 September 2014 (UTC)Reply
That's a good idea!
50 images may not require batch, but 50 x4 0 vols =2k images! And that's just plates!  :) Laura1822 (talk) 14:45, 1 September 2014 (UTC)Reply
I appreciate the sample with a white background, but I much prefer leaving the original color of the paper, especially for plates. I have a litany of reasons for that if you want to discuss. Just my opinion. Laura1822 (talk) 15:10, 1 September 2014 (UTC)Reply
  • Must have uploaded ~20,000 images, so 2,000 are not too scary. Also, not all images are color plates.
  • This is your project, so you make the decision. There is nothing wrong with the colored background on color images, and it looks nice. But, grayscale is not accepted with the yellow background, that I am sure of. I had to request deletions and re-upload the first 1,000 PSM images because of it.
  • As for the duplicate image captions, (on the image and then repeated below the images), please ask for clarification from User:Hesperian. He was my Rabbi in these matters. — Ineuw talk 15:31, 1 September 2014 (UTC)Reply

Comments on my moves etc.[edit]

@Laura1822: I apologize for moving the page while you were working on it. I assumed you were offline. Then, I was away from the computer and didn't see your messages. The other pages you were looking for are on this project page.

After renaming the main ns pages, I myself am not happy with it. I agree with you that you should name the main ns pages according to your wishes, I was concerned as to which part of the name is important, because I took the cue from the Index: pages. I did what I did to begin the process of organization because for me seeing results, is very important to my sense of organization.

No, no, once I realized what you were doing, I can only say that I am grateful. I didn't know how or where to create the project page, etc. I will give the names a think and change the names here and at commons if needed. I really have no objection to calling it this, though I'd tried to follow the conventions I saw here on WS.
I'll post my earlier draft of the talk page at the top. I tried to pull together some thoughts and organize it a bit. I agree with you about organization and I really appreciate your taking the time to create all these pages in various namespaces and categories, both here and at commons. Thanks again for all your help! Laura1822 (talk) 21:49, 1 September 2014 (UTC)Reply

Project component names as of now[edit]

Wikisource pages[edit]

Index

Index:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu

Project

Wikisource:WikiProject Ackermann’s Repository of Arts

Main namespace

Ackermann’s Repository of Arts
Ackermann’s Repository of Arts/Series 1/Volume 1

Commons pages[edit]

  • The unused images category is where we can store the said images, rather than constantly ask for deletions.
  • According to the publications table on the commons, the category names won't meet the needs of the project. — Ineuw talk 21:38, 1 September 2014 (UTC)Reply

Last comment of the day about naming[edit]

My last comment of the day regarding naming, consistency is the most important aspect when you plan the names for the above. Nothing else matters as much. For example: I was not the one to start the PSM project, but was quite "lost" (read "upset") that the the word "The" was subsequently omitted from the name by the person who started it. Not just because the publication name included it, but for my sense, the consistency was broken and couldn't do anything about it (changing 90+ volumes). Some project components had it and some didn't.

Avoid inconsistency and base your decisions by weighing the importance of each word you wish to include or exclude in the titles, categories, etc. Consider every word in the context of identity and content. How much of the content is actually fashion?, literature?, commerce? manufactures?, fashions?, and politics? or, Ackermann? And, what readers need to identify the work on the web, including search engines. — Ineuw talk 03:37, 2 September 2014 (UTC)Reply

IA volume set[edit]

Series 1

1809 Vol 1 (Jan-June) Vol 2 (July-Dec)
1810 Vol 3 (Jan-June) Vol 4 (July-Dec)
1811 Vol 5 (Jan-June) Vol 6 (July-Dec)
1812 Vol 7 (Jan-June) Vol 8 (July-Dec)
1813 Vol 9 (Jan-June) Vol 10 (July-Dec)
1814 Vol 11 (Jan-June) Vol 12 (July-Dec)
1815 Vol 13 (Jan-June) Vol 14 (July-Dec)

Series 2

1816 Vol 1 (Jan-June) Vol 2 (July-Dec)
1817 Vol 3 (Jan-June) Vol 4 (July-Dec)
1818 Vol 5 (Jan-June) Vol 6 (July-Dec)
1819 Vol 7 (Jan-June) Vol 8 (July-Dec)
1820 Vol 9 (Jan-June) Vol 10 (July-Dec)
1821 Vol 11 (Jan-June) Vol 12 (July-Dec)
1822 Vol 13 (Jan-June) Vol 14 (July-Dec)

Series 3

1823 Vol 1 (Jan-June) Vol 2 (July-Dec)
1824 Vol 3 (Jan-June) Vol 4 (July-Dec)
1825 Vol 5 (Jan-June) Vol 6 (July-Dec)
1826 Vol 7 (Jan-June) Vol 8 (July-Dec)
1827 Vol 9 (Jan-June) Vol 10 (July-Dec)
1828 Vol 11 (Jan-June) Vol 12 (July-Dec)


Bunka Gakuen image set available[edit]

If you have trouble deciphering the text because the image is unfocused, try using the images at Bunka Gakuen: click on Home, click on Title, click on R, click on The repository of arts (9th in list). The images are first presented in sets of 10, then you can click on the one you want, then you can right-click and save it or open it in a new tab and zoom in.

Pages and files' naming[edit]

I moved your post here because your approach is not in the right order.

  • Most important is the list of Wikisource Main namespace root page names from 1809 until the end of the publication. Sub pages as /Volume 1. . . . /Volume 2 . . . etc. was incorporated into the root name because separating them are unnecessary. Here we will have to consider the chapter names as well:

As an example I used the Index page name for the main namespace name:

Repository of Arts, Series 1, 1809 Volume 1 / January-June / Article title

  • Second most important are the Commons Category root names. They should be the same as the main namespace names you decide upon. Again, sub pages as /Volume 1. . . . /Volume 2 . . . etc. is OK. I cannot upload any images until this is done - because category name changes to the images must be changed manually. Renaming a Category does not cascade to it's contents.

A Category example base on the same concept:

Category:Repository of Arts - Series 1/Volume 1

The rest is fine. Commas are fine. Just verify for yourself that the Index pages' names are of consistent style from the first to the last volume.

I propose this for the filename:
File:Repository of Arts - Series 1 - 1809 - Volume 1 - January-June.djvu
or
On the commons - File:Repository of Arts, Series 1, 1809, Volume 1, January-June.djvu
On Wikisource - Index:Repository of Arts, Series 1, 1809, Volume 1, January-June.djvu
I prefer the one with commas but is there a good reason not to use them?
Reply: That's fine.
More importantly, when I rename this file on Commons (I have filemover status so the change will be immediate), will it break anything here?
Reply: Possibly, but not sure but if the links are broken, please do the following.

Will I need to make any changes manually, such as renaming the index page? Or will everything be taken care of automatically?

Reply:

  1. After renaming of the file on the Commons, rename the Index file here on Wikisource.
  2. Check some random pages. If the contents are OK and line up as before, then the page alignment to the original is OK.
  3. If the pages are not aligned properly, don't edit manually!!!
  4. Post a bot request for page alignment HERE.

@Ineuw: Pleast help. I'm sorry but after all my dithering, I did the rename at commons and here and yet still seem to have done it wrong. I apparently changed the filename to "File:File-Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu" (not sure where the second "File-" came from) and so had to rename it again to remove the second "File-" (I didn't think to do an "undo" until later). Now the file at Commons File:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu shows that all the Page namespace files link directly to it. On the new Index page here, it's showing the djvu under the new name, but none of the pages have been created; they're still at the old name. How do I fix this? Is this what you mean by "page realignment" for a bot request? Sorry to be a bother, not having a good day. Laura1822 (talk) 19:44, 4 September 2014 (UTC)Reply

Laura1822 (talk) 14:14, 2 September 2014 (UTC)Reply

Ineuw talk 18:46, 2 September 2014 (UTC)Reply

A heads up on the images progress[edit]

@Laura1822: There is no hurry, just wanted you to know that I have a number of images ready for upload, but won't do so until you finalize the Category names on the Commons.

While I choose the better of the two sources for now, BG's quality is poorer than our JP2 from IA. Also, searching for their work is a very slow process because of their server. The time it takes to render an image, I can make myself a cup of coffee. So, they are not a good image source. — Ineuw talk 05:13, 4 September 2014 (UTC)Reply

Can you give me an example where the BG image quality is poorer than the JP2?
I thought some kind of bot or automated scraper would be the best way to acquire the BG images.
If your holdup is BG's server speed and you don't want to use a bot, I downloaded all of the plate images to my own computer a long time ago. I have them in a directory structure with the JP2s from IA. They include a few other pages besides just the plates and woodcuts, mostly TOCs and title pages. But they all need to be cropped. If you're willing to do the image processing (I know you only committed to volume 1), I could zip them up and upload them somewhere that you could download the whole set at once. I understand though that you have other projects, and I'm in no hurry. I'm sorry I can't manage the image processing on my own, especially since I am so picky about it.  :) But I haven't the patience for it, and working with images can be hard for me with my light sensitivity. I don't know why proofreading is easier, but it is, at least in small doses (though after proofreading just half a chapter of Emma this morning, my eyes hurt). I'm not quite tech savvy enough to deal with bots and batch uploaders, or I'd consider uploading them all to Commons as-is (raw).
I will try to get to the Commons renaming soon. Finished the Commons renaming. I decided to move all the volumes into a sub-category "by volume" and not try to add months and years to their names. (I might change my mind later, but it's not too hard to move things around with cat-a-lot.) I added the last four categories for series 1, and will add more categories under the top "Ackermann's Repository of Arts" category when there are more images to play with. Thank you for all your help! Laura1822 (talk) 17:20, 4 September 2014 (UTC)Reply

Replies to User:Laura1822 concerns[edit]

@Laura1822: I think that our posts should be sequentially created rather that inserting into old posts. It's easier for others to follow and I can find your previous posts to reference.

  1. At this time, I am fixing the renaming issues on both sites. Changes will not happen on the Commons for at least 24 hours. So, don't worry about it.
  2. A bot to align the pages is a Wikisource issue. When you rename the index, that moves all your works in the Index namespace. But the pages are in the Page: namespace and still pointing to the old source. That's what you saw HERE where the Pages still refer to Ackerman. But this will change by the time you look tomorrow. The bot replaces the page links to the new Index.
  3. In case you are unaware, all files from the Commons are mirrored here (which is probably a technical requirement), as opposed to on Wikipedia, where if you click on an image, it will take you to the directly to the Commons source. It's best if you direct your references to the source as in commons:File:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu.
  4. Regarding adding the months to the date, and adding each image's description you have to do that. I only have to meet the Commons requirements for a meaningful title and description which for me is one and the same. I use the image upload wizard copy features, which fill the fields of like info automatically. I am sure that you can figure out the rest.
  5. I will upload a comparison image of the two sources to see the quality difference, as soon as I return to work on images. — Ineuw talk 23:27, 4 September 2014 (UTC)Reply
Thank you very much! I'm sorry for all the bother. Laura1822 (talk) 01:46, 5 September 2014 (UTC)Reply

Specific replies:

  1. Thank you. I was too tired to worry anyway.  :) ETA: the bot request has been marked done, but as far as I can tell, the pages haven't been moved. 14:14, 5 September 2014 (UTC)
  2. I really didn't understand your instructions. I thought "alignment" referred to the pagelist somehow, that a move might offset the pages or something. So I thought you were saying I did not need to make a bot request unless the page numbers were wrong. So thank you for making the bot request for me.
  3. I did not know that, thank you. But in what case would I need to refer to them that way? On talk pages like this one? Not on Index pages. Also, there's a setting on Wikipedia allowing a bypass of the page that says "this is a file from commons," taking you directly to the file on commons. Is there no such setting here? I couldn't find one.
  4. Okay, I'll add the dates and sorting. But please see my note above (in style guide) about the fabric swatch pages. And for the record, what is, for example, "D 067" in your filenames on commons? I recognize the raw page numbers, but what does the "D" stand for?
  5. Thank you. There is no rush. I am too tired to work very much on this today anyway. I tend to work in spurts.


  1.  
  2. . I used incorrect language - They have to rename the page source with a bot to link to the new name.
  3. . I am not sure, but I don't think it's a technical issue (as I think of it now), but rather, it is very low on their list of priorities. I remember a time when it was mirrored on Wikipedia and one had to click this Template:Commons top icon as well. Now we only have a text link.
  4. . "D" stand for the "DjVu". — Ineuw talk 03:02, 6 September 2014 (UTC)Reply


Thank you for everything. Laura1822 (talk) 12:33, 6 September 2014 (UTC)Reply

Updates[edit]

I moved the main page back to The Repository of Arts from Ackermann's Repository of Arts and fixed all the redirects, including the ones that still pointed to Ackerman's Repository of Arts (with one "n"). I hope that didn't mess anything significant up. I left the redirect as Ackermann's Repository of Arts is still a common way to refer to it. I also created a redirect from Repository of Arts. I think that should cover everything.

The WS Project is still called Ackermann's Repository of Arts and that's fine.

I am slowly working my way through proofreading the text and haven't made all decisions needed regarding formatting. I am also still pretty clueless about transcluding articles, but not quite ready to go there.

I am going to have to solicit help for the tables. Is there a way to force a page to display rotated when it's not an image from Commons? (It's easy to do in the djvu file on my computer, but I would like to display it horizontally in the proofreading editor/extension.)

Thanks for any comment/suggestions/help. Laura1822 (talk) 21:37, 14 September 2014 (UTC)Reply

To do[edit]

  • proofread
  • create pages with basic formatting: headers and footers, article titles, rules, etc.
  • check all plates for text above & below, including plate numbers (compare to Bunka Gakuen)
  • publish style guide
  • work on uploading high-res images of plates from other sites