Talk:Scanning old books

From Meta, a Wikimedia project coordination wiki

I agree, however...[edit]

Have the people at Wikisource been notified about this? If not, that should be your immediate priority. As for the idea, I don't think it's so radical. My main concern is that scanned books take up a huge multiple of the space needed to store the plain text, and space is something that the Wikimedia Foundation (I get the impression) does not have a lot of. When there is not a great deal to be gained by scanning rather than transcribing, I would go for transcribing. Daniel () Check out Wikiscope! 19:43, 11 July 2006 (UTC)[reply]

There is a great deal of merit to having the original printed text if you can have it. I own a great deal of antique and ancient texts, and a lot is lost when you haven't the visual "feel" of the print.
At a minimum, the graphic material should be reproduced and arranged properly within the transcription.

User:ScotMcPherson

The ease of displaying the orginal graphics inline of the wikitext in the way they were orignally formatted is one the key advantages Wikisource has over other digital libaries. The only time I have had to fudge this personally is when the image on it's own page and within the body of text. (ie not between two chapters) This is already being done where we have acces to the graphics.--BirgitteSB 13:08, 25 July 2006 (UTC)[reply]

This is the new thing in the suggestion. To have the original printed text published. I argue that this is a good thing for several reasons.

  1. Distributed proofreading.
  2. The transcriptions get more trustworthy if you can compare them to the original text.
  3. The visual feel and graphic.

The problem is the size of the image files. // Wellparp 08:20, 12 July 2006 (UTC)[reply]

Exactly what I was saying. Therefore I think that image files should only be used where there is a clear advantage from them. For example, if we wanted the NIV Bible, there is no need to scan it as millions of copies are available worldwide, exactly identical. If we wanted the Gutenberg Bible, perhaps the uniqueness of the typeface, paper etc. would merit a scanning. Daniel () Check out Wikiscope! 08:40, 12 July 2006 (UTC)[reply]
Visit Project Runeberg to see what I am envisioning. // Wellparp 09:10, 12 July 2006 (UTC)[reply]

Also[edit]

It's occured to me that scanned documents are not very wiki, are they? Wikisource stretches the boundary, although spelling corrections etc. can be done to a text file. But with an image file there can be no corrections or changes, and it is unlikely that new versions would be uploaded. You would end up with a slowly increasing, static collection which is not very consistant with the Wikimedia Foundation's goals (as I understand them). By the way, a good way to get attention to any suggestion is to post it on the mailing lists (search around to find them). Daniel () Check out Wikiscope! 20:47, 12 July 2006 (UTC)[reply]

Move[edit]

Please, move this to oldwikisource:Wikisource:Scriptorium 72.232.31.146 21:52, 12 July 2006 (UTC)[reply]

My 2 cents[edit]

I stumbled across this and must also encourage you to advertise this disscussion on the Wikisource website and the wikisource-l mailing list. Scanning of books for the addition of new texts is already a regular occurance at the english wikisource. There is currently a disscussion on commons VP on how to best collaborate in storing these files. The recent enabling of DejVu is making this a better proposition. Although I uderstand that thumbnail views are still not up and running. As for the presentation of the scans. I believe they are best used for OCR & proofreading purposes.

I dislike the idea of showing the text along side the image of the text unless there are things which cannot properly be transcribed to the wiki. Wiki text has huge advantages over other forms of presentaion currently used by digital libraries. Wikisource has no such advantage in displaying images. In fact many other digital libaries display images in a much more user-friendly way. I would like to see the Wikisource text provide link to where readers may view a scanned image (whether on Wikimedia servers, or other reliable library sites). I do not want to see Wikisource text containing 15 scan images per page nor be broken into seperate pages to mimick printed page breaks. That said there are people at Wikisource who would like to see scans displayed alongside the text. I would once again suggest you try to bring in people who are already working on Wikisource to this disscussion. I am leaving on vacation for two weeks so I unfotunately will not be able to give further feedback in the immediate future.--BirgitteSB 13:03, 25 July 2006 (UTC)[reply]