Felix Kasza

That must have been the easiest and quickest workaround I ever put in place – move the pin to line 4, leave it there, and it’s off to the races. Running the risk of counding like a broken record, I am once again deeply obliged to you! Cheers and a million thanks, Felix.

This was a post in the "Desktop Questions" forum and had no takers. After further experiments, I am fairly certain that this is a bug. My Publisher version is 2.5.5.2636: Hi, it's me again! And I think I need to set up some fake accounts, this is getting to be embarrassing. :)) (The file with the repro is attached below, as usual.) I added a new picture frame and text box, grouped them (group "14" below), pasted them approximately where they belong, pinned them there. Here is the layer stack and what the page looks like: Here are the pin and text wrap settings: So far, so good. When I move the pin where I want it, that is, to the first letter of the first line of the text paragraph, this happens: "Inside right relative to column" and "inside top relative to line + 1 mm" is not where the groiup has ended up. Also, I wonder why my group "14" now shows up under "Body-left"; the pin certainly still is in the Body-right frame. But wait, I thought, maybe this is something with a master applied to each page of the spread separately? In the Pages panel, I selected the last two spreads, right-clicked, "Apply master…", and Publisher silently died. "Oh well, never mind!", to quote SNL’s Emily Litella. So let me move the pin a little lower, to the words "vertical lines" in the second line of the text para after heading 3.4: So we are back on the correct side of the spread, but the vertical position has been rounded up. Way up. The pin settings are still as they should be (float, inside top, line, 1 mm offset). I wonder: what happens if I were to move the pin another line or two down? Ah. That works! The entire thing is highly dependent of where on the page we are. If I change the "Body" style leading to, say, 125%. the text is flowed differently, the pin is no longer close to the top of the page, and it all works. The same if I delete or insert a paragraph such that my test text and test pin slide farther down the page. At this point I doubt that it is an error on my part. I would love for it to be my fault, but right now, I can see no logical reason why the formatter/layouter in APub would behave this way. Cheers, Felix. Statistics with the TI-Nspire CX-II.afpub Statistics with the TI-Nspire CX-II.afpub

I have no problem with Lora at all, but if you wish to typeset just this one paragraph so that it may seamlessly replace the original paragraph, then differences will stand out, more so than whether the line breaks and hyphenation are perfectly replicated. As long as your paragraph has the same length and width overall, the eye is much more likely to pick up on subtle font differences than on hyphenation choices. Here is one thing you can (and maybe should) do: Find an original paragraph with a reasonably long last line (because that line is not justified and has unmodified spacing). Scan it, zoom in, and measure the length of the entire line. That should tell you how close your Lora comes to the letter widths of the original font and how easy it will be to replicate paragraph lengths. Cheers, Felix.

Hi Ryan, could you share the title page and, if there is one, the colophon of the book, please? Assuming that the book font is indeed Lora (it looks like Lora, but I fail to see why a textbook would use Lora), you seem to be using weight 400 (Regular). Try weight 500 (Medium); there is a reason why many fonts with this weight are named “(Something) Book.” They are called “rivers,” and back when I first learned the basics of the trade, they were to be avoided like the plague; if necessary, hyphenation was manually changed or in the worst case, the text was altered. In your example, it's not so bad, and as soon as you switch do a different font weight, that may all change anyway. Rule #1: Leave the letterspacing (second row) alone. Modifying that is something that used to be done in US typesetting, and it may be necessary when justifying narrow columns with very few characters per line, but changing the spacing is fugly; just think about the pretty kerning that would be all messed up. The only exception I am willing to grant is increased letterspacing to emphasize Fraktur text („Sperrsatz“). The word spacing is conceptually easy: a line is filled, hyphenation is done if possible, and the word spaces are increased until the end of the line reaches the right margin. If there is not enough text, word spacing may be decreased to pull more in from the following text. This may be done in stages (like, try up to 105% word spacing, then try down to 95%, then up to 110%, then down to 90%, until it works). Also, hyphenation may be modified: secondary hyphenation locations may be chosen, or a rule of “max. N hyphenated lines following one another” may be relaxed. The bad part is that the weight assigned to each method or stage varies, as may the sequence in which they are tried. That gives you next to no control over the whole justification process. I still remember my teacher and mentor, it was the 1980s, and how he hated those crappy DTP programs with a passion. Aldus Pagemaker and Ventura Publisher were the bane of his existence. “Look, I designed my own business cards! Please print them!” – “No, no, no, no, nooooo …” Anyway. Don't waste too much time on playing with the settings; if you get a river to go away here, another will open up right over there. “Proper spacing” as in “the one and only correct spacing” does not exist. The letter spacing will generally be OK as long as you allow kerning to happen, permit at least discretionary ligatures, and use the proper optical size of your font: You may be able to enlarge Palatino nova out the wazoo, but at larger sizes it won't look nearly as good as the Titling font style of Palatino nova. Word spacing depends on your line length, font size and leading as far as body text is concerned. For titles, at least in high-quality typesetting, word spacing should be adjusted by hand for perfect looks (and don't forget to use a display or even a titling font). Short answer: The spacing is correct when you don't notice it while reading; if you do, the typesetting got in the way of reading, and for body copy at least, that is a major sin. ⁂ I get the impression that you are trying to reproduce the original such that the work could be reprinted with just one plate changed, and nobody would notice. For that you either have to typeset manually (manual line-endings, allowing these partial lines to be justified, or you have to have the exact parameters that the original software worked with. Good luck with that. Cheers, Felix.

What does your View Menu look like? Here (Windows), the first item is "Zoom", the submenu has "Zoom to Fit": Cheers, Felix.

Hi, it's me again! And I think I need to set up some fake accounts, this is getting to be embarrassing. :)) (The file with the repro is attached below, as usual.) I added a new picture frame and text box, grouped them (group "14" below), pasted them approximately where they belong, pinned them there. Here is the later stack and what the page looks like: Here are the pin and text wrap settings: So far, so good. When I move the pin where I want it, that is, to the first letter of the first line of the text paragraph, this happens: Having consulted my dictionary, I am led to believe that "inside right relative to column" and "inside top relative to line + 1 mm" refer to a different part of the spread. Also, I wonder why my group "14" now shows up under "Body-left"; the pin certainly still is in the Body-right frame. But wait, I thought, maybe this is something with a master applied to each page of the spread separately? In the Pages panel, I selected the last two spreads, right-clicked, "Apply master…", and Publisher silently died. So let me move the pin a little lower, to the words "vertical lines" in the second line of the text para after heading 3.4: So we are back on the correct side of the spread, but the vertical position has been rounde up. Way up. The pin settings are still as they should be (float, inside top, line, 1 mm offset). I wonder: what happens if I were to move the pin another line or two down? The entire thing is highly dependent of where on the page we are. If I change the "Body" style leading to, say, 125%. the text is flowed differently, and it all works. The same if I delete or insert a paragraph such that my test text and test pin slide farther down the page. At this point I doubt that it is an error on my part. I would love for it to be my fault, but right now, I can see no logical reason why the formatter/layouter in APub would behave this way. As always I am grateful for and open to any ideas! Cheers, Felix. Statistics with the TI-Nspire CX-II.afpub

And this is the key to understanding Publisher and its layers. Thank you! Your excellent APub manual comes close to articulating this thought, and it does show how to work with layers and how to caption and pin a picture frame. The documentation never mentions anything about layers beyond clipping, blending, masking (essentially Photo/Designer stuff). A note where it talks about moving objects in the z-order by moving their layers would have been useful: "You cannot move objects into a text frame, you can only pin them there. Pinned objects are z-ordered in the order in which they were pinned. This order can be changed with the "Move Back One" / "Move Foward One" and related commands from the "Arrange" group, but not by dragging the layers around in the "Layers" panel. And finally there is the book "Affinity Publisher 2" (Christian Denzler, Rheinwerk Verlag) which manages to separate picture frames and images (the former go into a vague category of "objects"), never mind combining them and pinning them and so on. It has good parts, but also nothing on how layers, images, picture frames, z-order interact or why some frames are more equal than others. So, thanks again! Cheers, Felix.

[\x{0080}-\x{ffff}] will find both æ and œ. The regex engine in APub makes me fairly happy (it does not yet do Unicode scripts or Unicode blocks, but the other \p{…} stuff seems to be there); the regrettable part is that there is nop Unicode class that would let you select letters minus ASCII letters. Also, the && syntax does not work as I had hoped: [[:graph:]&&[^a-zA-Z0-9‘’”“;.,-—–()!]] should find your ligatures, not to mention the code point á or it'a "a" + combining accent version. Regrettably, the engine seems to dislike "&&" in a bracket expression, dooming also attempts like [\p{Letter}&&[\x{007f}-\x{ffff}]] and [[:graph:]&&[:^ascii:]]. Set subtraction syntax also fails: [[:graph:]-[:ascii:]] O Serif: It would be nice to know which regex engine you use. Thanks!

Hi Ryan, Is there a possibility that your imported text has direct ("local") formatting that initially just looks like the style you want? Because if you then change the style, the direct formatting will override that change – until you reapply the style, overwriting the local formatting. Anyway, here, text that is styled but has no direct formatting changes just fine when the style changes, both in Word and Publisher. So, my suggestion: In Publisher, import some text from Word. Verify that it does _not_ follow style changes. Then, apply the intended style (with "Apply <style> to Paragraphs and Clear Character Styles") and verify that the text now _does_ follow style changes. If this works as I described, take a _very_ close look at your Word document: Make the Styles panel visible (en-US: Alt, H, F, Y); at the bottom, click the magnifier button to enable the Style inspector panel (screenshot below left). On the Style Inspector panel, click the A-with-eye button to show the "Reveal Formatting" panel (screenshot below, centre). Once there, tick the two boxes at the bottom (screenshot below, right). Here is a word that I just "manually" boldfaced: Or, as Walt suggested, just post both files and let us have a look. I may be a dumb n00b with Publisher but I can make Word sing and dance. Cheers, Felix.

Thanks again: I owe you. This works for picture frames, too, which is how I like to do things. I very, very much appreciate your patience with me and your help! Cheers, Felix.

Just so I am sure I understand: I create a picture frame. I pin it to a position in my text frame. Whoosh, its position in the layer stack fixes itself. I create a text frame. It goes wherever it wants, but let us assume that it's my lucky day and it goes into the Body-left part of the layer stack. I can move the two frames (pic and text) together but I cannot group them: If I have pinned the pic frame, it is because of the pinning, and if I have not pinned the pic frame, it is because the pic frame is in the layer stack's Outer Mongolia. The copy/paste you mentioned is something I fail to understand. Sure I can copy an existing group of ( group ( textframe pic frame ( picture ) ) ) and edit it, but that makes me think that APub needs to be shipped with a file in which a dev has manually put a group like that, and a Readme file stating "Do not delete this file: Cloning it, and copy/pasting from it are the only ways you'll ever get pinned and captioned picture frames into your documents." I mean, really? What big thing am I missing? Thanks, Felix. P.S.: Do you use APub professionally? I can take it or leave it, even though I would not like to do so; but what do people do who try to make a living with it?

Thanks for the tip on pasting pictures Let me try this: Yay! Thanks again! Anyway, I just placed the caret next to the last character on the left-hand page; there is not much else to select, as the current layer stack shows: Now I create a picture frame. New layer stack: Curses – foiled again! You wrote: "Select the master layer and the object will be created above it", but there is nothing else but master layer items to select. How am I supposed to select a non-master-layer item when I cannot insert one in the proper layer stack position to begin with? I'll say this, cuneiform looks better and better. Give me a stack of clay tablets, and I'll give you any ordering you want. Cheers. Felix.

Hi! I am trying to write (re-write, as most of the stuff exists ina Word document already) a short guide to the TI-Nspire's descriptive and inferential statistics. I figured something short and simple would be an ideal place to try things, learn more about Publisher &c. I am having *nothing* but trouble, and here is the next installment of this episodic tragedy: If you look at the attached document (you can ignore the missing images, they make no difference), e.g. at the spread sequentially numbered as pages 6–7 (pagination is 2–3) and examine the layer stack, it looks just fine: Gee, I hope the pic will be shown smaller in the finished posting … Now we add a picture frame anywhere, doesn't matter where: Why is this thing not part of the hierarchy under "Body - 2 Pages"? As a consequence, the usual flow of "place picture into frame, draw a textbox for the caption, group them, pin the group" fails utterly; as an informed guess I would say that pinning is only possible from a placed object to somewhere in the same hierarchy (with a shared parent, that is). Things I have tried: Reapply the master spread "Body" to individual spreads and to everything from p. 5 to end; select random page elements (including the main text fraoime that comes from the master page) before adding the pic frame; manually rebuild the whole document (now five times) by copying all text into a text editor, stripping off anything that is not text; creating a new document; redoing the master pages and the styles; paste the unformatted text back in; reformat everything; place the pictures. And after a while it happens again. I am at a decision point: Stick with Word (strong dislike) or give Adobe even more money (great, massive dislike) or cut a reed stylus and write cuneiform on clay tablets (better than feeding Adobe) or find a way to work with APub that lets me make some progress instead of producing the same two pages over and over and over. I am pretty much at the end of my wits here and will be extremely grateful for any help you can offer. Thank you, Felix. Statistics with the TI-Nspire CX-II.afpub

Hi, @MikeTO – once @Oufti pointed out what was wrong and you root-caused it, I understood my error (even though I do not consider it an error). Mu work begins with an odd (right) page and ends with and even (left-hand) page. Every time I add pages, I do so in multiples of two pages, so at the insertion site, there will be an old even (left) page that is now followed by a freshly inserted odd page, so this spread is a patchwork. Well, at least I have a work-around now, thanks! Felix.

@MikeTO, thank you! It is amazing how many people spent hours of their time to help me out, and I am very, very grateful. I am at a loss trying to explain how I managed to achieve this split on pages 4–5. I did not apply master pages except automatically while adding pages to the document, and I certainly did not aply a master to a single half of a spread. Colour me puzzled! As for the crash: It is not clear to me how to notify Serif of a crashing bug, except through a forum post. I do have a separate thread open for that, but no official response so far. Thank you (and all of you!) again for your help, your time, and your efforts on my behalf: I appreciate them highly! Cheers, Felix.