User talk:Groupuscule

— billinghurst  sDrewth  07:01, 18 September 2012 (UTC)

PDF files
If a PDF of a source is in the public domain, it should be uploaded to Commons rather than here. If it is not in the public domain, then it probably shouldn't be uploaded at all. --EncycloPetey (talk) 03:36, 28 June 2013 (UTC)
 * NYT 1866, definitely public domain. What do we do now? Groupuscule (talk) 03:38, 28 June 2013 (UTC)
 * If you can upload the same file to Commons, then I can delete the local copy. Everything will behave in exactly the same way as if the file were here, but other projects will be able to use the file as well. --EncycloPetey (talk) 03:43, 28 June 2013 (UTC)
 * Commons is acting really weird. Have uploaded there before. Will try again in a bit. Groupuscule (talk) 04:32, 28 June 2013 (UTC)
 * Yes, several MW projects started acting really weird a few minutes ago, including some aspects of Wikisource. It may be more than "a bit" before the problem is rectified, as it's not just Commons having the problem. --EncycloPetey (talk) 04:41, 28 June 2013 (UTC)
 * Things seem to be operational. commons:File:Second Freedmen's Bureau Bill.pdf Groupuscule (talk) 07:01, 28 June 2013 (UTC)
 * I've been off WS for a few days, but it looks as though the PDF situation has been handled. We do have an OCR of sorts, but it's very primitive.  Most of our works are run through the Internet Archive, where they create a text layer, Djvu file, and the other layers we typically desire.  However, this usually happens with multi-page works, and I don't often work with single-page items, much less newspaper articles, so I can't fully address that question.  I have tried applying our OCR, but go nowhere.  It's a finicky tool at the best of times, and I wouldn't hold much hope for getting it to work on a three column article in such fine print.  In your situation, I would look for OCR options in some other location, either on-line or as a dowloadable freeware package.  You can look at Help:Index pages, where near the bottom are some examples of single-page documents that has to be transcribed without full use of OCR.  You may find other information there helpful.  What I do not find is any help or suggestion page concerning OCR software for situations like this.  If I still had my previous computer, I could do it, since I had a package that came with my old scanner that allowed selection of a region of text for OCR conversion.  Unfortunately, my new computer doesn't have this, the old printer is long dead, and I haven't yet had the need to replace it with anything (nor the time to look).  You could post for Help in the Scriptorium, which is the central discussion area for all of Wikisource, and someone might be able to offer more specific advice to you. --EncycloPetey (talk) 21:22, 5 July 2013 (UTC)