IMSLP Forums

1. I have some files that have been scanned from reprint editions (like Dover). Do I need to remove any reference to those publishers?
2. is there a program that will easily decompose b&w pdfs into TIFF files (or other image format)? Thanks

1. I don't think so... do you mean the reprint publishers or the original ones? Though I don't think removing either would be necessary. Though, what exactly do you mean by "references"? Copyright labels?
2. There are several ways. The best I found is to use pdfimages from the xpdf command suite (http://ftp.foolabs.com/xpdf/download.html). You can also go the more circular route by first splitting the PDF file into individual pages via PDFtk, and then converting each individual page to another image format via ImageMagick, but be warned that this process is *lossy*, which means that the quality will decrease. On the other hand, pdfimages is lossless; there will be no decrease in quality. The other difference is that pdfimages only *extracts* the images, while ImageMagick actually converts the entire PDF page into an image (including the text and stuff), which is also why there's quality loss. For scans (like the ones on IMSLP), I'd definitely use pdfimages.

I mean simply the name of the publisher; in this case, there is a big MASTERS MUSIC on the bottom of the first page of music. There are no copyright marks or anything, so I'm assuming then that it's OK? (What about the publisher's original title page?)

Well... If by original title page you mean the title page of the reprint publisher (the publisher who reprinted the score), then it's best to not include that page, since the page *is* under copyright. Otherwise (i.e. you mean original as in the original publisher) the title page is also in the public domain.

On the other hand, the two words by themselves should not pose a problem.

It would be a difficult case to argue that a mere title page is protectable by copyright unless the title page incorporated a design that is copyrightable in its own right as artwork like the Picasso drawing that graced the cover of Stravinsky's Ragtime when first published. However, corporate names and logos are protected under trademark law. Thus. it's not such a good idea to include a current company logo because trademarks - much like diamonds - last forever. Better to get rid of the logo and insert a line such as [Reprint edition issued Masters Music Publications, Inc., 1990] before generating the PDF.

Masters Music is very much in business as I write this, as is Dover. It's fine to mention that the PD score you're scanning is a Masters or Dover reprint of Durand or other older publisher. It also likely OK to reproduce the original title page of the PD work in question since company names and logos have all generally changed substantially since 1922. Just try to avoid reproducing the current logo of the publisher, especially if you see the symbol ® or ™ anywhere near it.

All the above applies to the USA, where yours truly has worked for various music publishers for about three decades now, including numerous consultations with copyright attorneys. (That's where I picked up my copyright background.) I can't really say very much about the copyright status vs. trademark status of title pages in Canada, the EU, etc., but my overall impression is that it's not drastically different when it comes to trademarks.

As for extracting TIFFs from a PDF, it's a snap if you happen to own the full ($$$) version of Adobe Acrobat, provided the file if not locked. All you do is pull down the file menu and select "Extract images as TIFF." It extracts in a choice of color, grayscale or monochrome at resolutions up to 600 dpi.

BTW, if anyone is interested in trolling this site to see if there's anything they wish to put together as a PDF, Gutenberg used to have a page ( http://web.archive.org/web/200502152312 ... available/ ) where you can download zip files of TIFFs taken from various PD works, usually scanned from Dover scores. I've gotten some of the zip files to open, but others appear to be corrupted. Perhaps the fact that I work on an antiquated Mac has something to do with it also, who knows.

Wow! Nice find. The files originally appeared corrupted to me too... but then I realized it was a bug (perhaps deliberate?) on the part of archive.org; when you download a file through their site, you are never able to download the last byte of it, and so it may seem corrupted. However, when I went to the main Gutenburg site and download from it (http://www.gutenberg.org/music/available), this problem seems to disappear, and I can open the ZIP files fine

I also just noticed that the link to the downloadable scans is working at Gutenberg again. It automatically bounced to the main sheet music page for about the last year, which is why I looked up the archive on the way-back machine. They appear to have scanned a number of Dover scores for some works, and to have (ahem) extracted TIFFs the normally locked PDFs avalable from CD Sheet Music in others. I don't see why CD Sheet music is so paranoid about the scans they have been using, as most of the ones I've run across appear to be about 150 dpi. Elibron's files, which are even more leavily locked, also have very low resolution scans.

Kalmus, and their subsidiary Masters Music are currently scanning all of their PD source scores at a full 1200 x 1200 dpi monochrome, which (as you can imagine) is requiring them to invest in many terabytes of storage space.

Carolus wrote:Kalmus, and their subsidiary Masters Music are currently scanning all of their PD source scores at a full 1200 x 1200 dpi monochrome, which (as you can imagine) is requiring them to invest in many terabytes of storage space.

Wow... is there even an advantage to that over, say, 600 dpi? Except for making the files really slow to open...

Carolus wrote:I also just noticed that the link to the downloadable scans is working at Gutenberg again. It automatically bounced to the main sheet music page for about the last year, which is why I looked up the archive on the way-back machine. They appear to have scanned a number of Dover scores for some works, and to have (ahem) extracted TIFFs the normally locked PDFs avalable from CD Sheet Music in others. I don't see why CD Sheet music is so paranoid about the scans they have been using, as most of the ones I've run across appear to be about 150 dpi. Elibron's files, which are even more leavily locked, also have very low resolution scans.

Kalmus, and their subsidiary Masters Music are currently scanning all of their PD source scores at a full 1200 x 1200 dpi monochrome, which (as you can imagine) is requiring them to invest in many terabytes of storage space.

CD Sheet Music uses the old RC4 encryption which is cracked since years. What do you think about removing the security of their pdf's and removing the logo's? They CAN'T copyright the sheet music as it's public domain. The only thing they claim to copyiright is the commercial "niche" of putting it all on a cdrom, so I think the cd can indeed not be copied legally, but the pdf's are free of copyright.

There is nothing copyrightable about CDSM's PDF files. The potential problem we face is when someone posts a CDSM file complete with their logo, which just happened with the Paganini: Violin Concerto No. 1. That's a trademark violation which they could sue for.

The bh2000 scans appear to be CDSM files with the logo removed. This is easy enough to do once the files are unlocked. Simply use Acrobat's crop tool to remove.