Best practice anchoring images

I have a large document (100pages). It is a book with filling background images on each page. Most of them are anchored to the page, because there might be a bug: if I try to anchor such large images to anything else than the page they will somehow be anchored in the page header, thus making them appear on each page.
When the images are filling the entire background sometimes the text is not displayed even if the images are really sent to the background. If you click in the text most of the time you select the image.

I can now work around it, but for my next larger book I’d like to know the best practice how to layout a book with large backgroud-images.

When trying out to use a master document there are several issues with the styles and page numberings. E.g. using one subdocument per short-story each containing 2 or 4 pages. The sub document needs to start with an even page, but then the master document cannot start with an odd page. Also a mix of sub-documents that have a different number of columns seems not possible. I tried to use text-boxes, but then the styles are not available and I find myself manually formatting the text again.

I can also try to provide with minimal examples if that is needed.

Build ID: 1:6.4.7-0ubuntu0.20.04.1

This might not be as clear and concise as You are used to in this forum, but I’d like to work on that as well :slight_smile:

Anchorage of objects

#Background images

If you really have a different background image in every page, Writer is not the adequate tool.

Writer is flow-oriented, i.e. it manages a text flow (a run of paragraphs) and determines on its own how many pages are needed. Text is broken into pages according to various properties of the paragraphs. You can’t really predict where page breaks will occur. This means page breaks, apart from exceptions, are not important for text significance and author’s intent. Exceptions are manually forced page breaks between logical parts of the book like TOC isolation or chapter start.

In flow-oriented documents, images are preferentially anchored to paragraphs so that the image is guaranteed to appear close to the “associated” paragraph whatever the text movements caused by formatting.

Background images make sense in a page context, i.e. they must be associated with page styles. Since a page style is common to a sequence of pages, the background is also common. The image can be inserted either as an attribute of the Area tab of the style or as an image anchored in the header or footer.

If you have one background image per page, defining one page style for every page is not very convenient. Moreover there is the question of switching from one page style to the next. Forcing a manual page break is not a solution because it will inevitably result in irregular formatting and you become very sensitive to any slight change of the paragraph properties.

Page styles have an attribute to define the next page style to switch to when page overflow occurs. However as you can’t predict where page break occurs, you can’t predict the total number of needed page styles (or the last one will be used repeatedly in the series is exhausted).

When pages are the main point of your design, you should use page-oriented applications (also called “desktop publishing” DTP). Scribus is a free application (FOSS); Quark XPress, PageMaker, Adobe InDesign are commercial ones.

Here the pages are the primary objects into which you insert linked blocks (for text, images, …). You define the sequence of the pages and constrain the text into these blocks. Page breaks are known and text is clipped if it does not fit inside the set of linked blocks.

#Master document

The key to success is a consistent use of styles. It is even more critical to avoid direct formatting than in usual documents. The hard point to understand is the fact that styles in master and subs are independent (only the name is the linking factor, not the style definition). Therefore, the definition in the master is used when “gluing” the subs together.

The simplest way to achieve consistency is to base all documents (master and subs) on the same template and make style modifications exclusively in the template.

Apart from breaking the final “product” into separate file, everything you can do in a single document can be done with a master, including different number of columns. But this requires consistent definition of styles: paragraph, character, frame, page and list styles. If you only work with paragraph styles and neglect the other categories, you will fail.


Your question was in fact two-fold. There is not enough detailed information in each part to give targeted advice. You are encouraged to ask new one-topic-only questions (referencing this one to show it is a follow up) with all the needed description about your present implementation in order to receive adapted suggestions to fix your difficulties.

As always, with long and sophisticated documents, mastering styles is a prerequisite. Read the chapters about them in the Writer Guide.

To show the community your question has been answered, click the ✓ next to the correct answer, and “upvote” by clicking on the ^ arrow of any helpful answers. These are the mechanisms for communicating the quality of the Q&A on this site. Thanks!

Thank You very much for the detailed response. I will have a look at Scribus then, at least for the next project.

This is a late response. :wink: If you have the patience to wait with the insertion of the background images until the text is finished, there is a much simpler way than background images in the page style. You can simply insert each image on its corresponding page and set the Wrap to Through, and at Options select In background.

Edit for clarification: The picture should be anchored to paragraph or character, the position should be set as you would normally do, zero from top and left from top and left of text area or page.


This is effectively a good workaround provided it is applied at the very last step of production when the document is ready to print. However, do you mean you Anchor to Page? This makes sense on a print-ready document and allows for limited edition. Should the document undergo a more important revision, it is likely that the best approach is then to remove all images and to reinsert them again as there is a high risk of losing synchronisation between text and images.
The only assurance of synchronisation is Anchor to Paragraph mode but you can’t assert page sync with it. Cruel dilemma.