Scribus 1.3.3.8 on Ubuntu Feisty amd64 I recently completed a flyer which uses seven fonts, three of the Bitstream Vera Sans face and all four of the Junicode face (regular, bold, italic, bold-italic). The flyer is four letter pages, later imposed to two landscape tabloid pages (4-1 and 2-3). I spent hours trying to get Frank Cox's script to work (posted a week or so ago here under another thread). When the frustration level got too high I just said "to hell with it," exported from Scribus as PDF, then manually placed the PDF pages into a new InDesign CS document for the imposition. I'm trying to get away from InDesign, but in this case I was running out of time.
When I exported from Scribus as a PDF the export dialog box showed the Vera Sans fonts as to be embedded. In the Window below the Junicode fonts were listed to be converted to outlines. Being new to Scribus, the first time I exported to PDF I was unaware of this dialog box, so I just hit the defaults. Then I did the imposition in InDesign and exported to a new PDF. I noticed the screen display was a bit rough, but I printed it anyway. The imposed tabloid pages took 35 minutes to print on my Laserjet 8000. The problem, of course, was rendering all those outlines. I went back to Scribus to figure out why the printing was taking so long, and that's when I discovered the dialog box showing the Junicode fonts to be converted to outlines. Junicode is a free font and is listed in the Ubuntu main repositories. I can't believe it has a "do not embed" flag in it. I tried adding the Junicode fonts to the upper "to embed" window in the dialog box, but the left-right arrows wouldn't move them there. However, there is another arrow between the "to embed" and "to convert to outlines" windows. When I selected the Junicode fonts in the "to convert to outlines" window and clicked on this arrow I expected them to move up to the "to embed" window. They did disappear from the "to convert to outlines" window, but did not appear in the "to embed" window. I exported to PDF anyway, and this time Scribus did seem to embed the fonts. I have InDesign CS and Adobe Reader 8.1 running under Windows 2000, which is running as a guest OS under VirtualBox. First, I opened the four-page PDF from Scribus in Reader to see if it looked OK, whereupon Reader announced that it couldn't find the Junicode font. I fixed that by installing the Junicode fonts under Windows 2000. Then I proceeded to do the imposition in InDesign and export to a new PDF the same as before. This time imaging the file took only six minutes on the Laserjet 8000. (Normally first page out on this printer takes only 15-20 seconds, but these are tabloid pages, plus there are three photographs in the flyer, so six minutes is reasonable.) I took the copy and sat down to give it a final proofreading. I quickly noticed that there were four instances where I had used "am" and "pm" in the original Scribus document. Consistent with Chicago and most other style books, I had set these as small caps. To do so I had inserted them as special glyphs rather than use the small caps attribute. I did so because Junicode, although a TrueType font, is very complete and contains true small caps. They are in Unicode table 247, that is the hex codes are A=F761, M=F76D and P=F770. In the Scribus document they appeared correctly as small caps, but in the printout they were just "am" and "pm," as regular lowercase letters. I traced back through the steps I took to figure out where things went wrong. I still had the PDF originally exported from Scribus where the Junicode fonts had been converted to outlines. I opened this file in Kpdf, and also the new PDF file where the Junicode fonts had been "embedded" (except evidently Scribus embedded only the name of the font to be used, and not the actual font). In the PDF-outlines file the small caps appeared correctly. In the PDF-embedded file the small caps had been convered to regular lowercase letters. I'm not sure if this is a bug in Scribus, a bug in the Junicode font, something I'm doing wrong, a dliberate design issue in Scribus or Junicode, or what is going on. Any suggestions and discussion?
