Intuitive image support?

I really wanted to use writer for some art I’m doing. It is story writing but consists of taking an image and then writing a very short story. I have having some difficulties which may lead to a question and also some suggestion regarding features.

Compared to Google Docs, which I don’t want to use for this for a number of reasons writer doesn’t just work. All I want is to put in an image which automatically goes into a character space and then write around it.

By default writer appears to impose too much complexity on me. I don’t want to configure the exact XY and so on. I just want to put it in a position, resize it to make it smaller and have text flow logically around it. This is almost like with HTML images.

Instead by default I find it really fiddly. There’s something quite invisible going in, it’s not like I can drag it to what is clearly a caret position. All of the images are higgledy piggledy and I just really don’t know exactly what’s going on with the lines and so on. There is a visibility issue where I’m not sure they’re just displacing stuff underneath normally.

They’re all out of alignment and it’s like by default it’s trying to simulate sticking images to a scrap book rather than naturally having a flow and placement that just works. When I right click the image and try something simply such as different anchoring then there’s complexity overload.

It’s making me have to deal with a whole bunch of complexity that by default I really shouldn’t have to. Whatever it is doing by default with things like dragging is not what people would need in most cases by default and it is something that seems like it should be an extra option.

Wrap as character is closer to what I want but then has annoying alignment issues. What people might want by default for a quick document of images and then notes is something like this…

[image] Test and paragraphs just flow normally as if only displaced
[image] to the right with the image aligned to the top of the first line
[image] it is in.
[image]
[image] With the ability to have a new line, either paragraph or pseudo
and for it to only drop off when it goes onto the line the image no longer
covers perhaps skipping a partially overlapping line.

Are there some settings beyond just as character that can produce this kind of simple default behaviour? Am I missing something? Is there a feature missing? I generally think things like this, putting new users in a position to force them down a steep learning curve and to have to spend ages going through things they shouldn’t need at this point to get something done that ought be simple will cause a lot of frustration.

It’s sort of one of those should take two minutes thing but instead by contrivance becomes an indeterminate blocker what the time required or if what they want is even possible on the current task the person usually needed to get done quickly to offload what’s on their mind before it perishes.

I would appreciate if someone happens to already know of some default settings to make it behave in a simple manner but I would also much appreciate some consideration as well of the impact of the way things might be designed with a simple advanced default dynamic and how that might impact usability as well as user experience.

Update: Anchor to paragraph with setting the position to “left” and “top” improves things a lot. I think I can make wrap to paragraph the default but I can’t find any way to do it for position. So far barring struggling to get it to be the default this configuration to eliminate unnecessary clicks and optimise workflow it seems to be closest to “just works” by default. It can still be a little confusing in some cases but is most of the way there.

ask118733.odt (49.7 KB)

Your question is lacking minimal technical information: OS name, LO version and save format.

Your issue can be solved with a frame style, but are you already familiar with paragraph and character styles?

Handling of images inside text is not trivial at all. The first question to ask is: is your image in relation with some paragraph? This will determine its anchor.

The next question is the position relative to (or independent from) this anchor.

From your terse description, you want an image attached to some paragraph (its description or comment), positioned at left, against the margin, with text flowing on its right. The image may eventually span several paragraphs.

Is this your goal?

The to-be-suggested solution will only be valid if your document is saved .odt.

This sort of experiment is useful but also highlights the complexity overload issue though I guess it has to be done at some point. I am not certain what wrap behind does here and really you have to test a lot of permutations.
:confused:
With the partial solution I have found while it might be possible to get it to look good after some fiddling there are still a lot of issues in motion. The problem is not as simple as the final result but how you arrive at it as well.
:confused:
In terms of userflow if I drag an image into a place instead of it just being a muscle memory process of right clicking and setting the position to automatic left and top which is a few extra clicks but at least predicable and repeatable.
:confused:
I suspect that if I could set that positioning to the default then it would further mitigate that issue though I suspect it may lead to a feature request as well to do something like inserting new lines to push the rest of the content down or something. That might also be a wrap issue.
:confused:
Are these settings hard coded or is there a file somewhere I can easily edit to fix the default positioning to a simpler mode where it’s automatic by default so I don’t have to specify x/y offset when I don’t need to?
:confused:
I find it awkward to navigate to but if I go to the menu Styles->Manage Styles and then look carefully for the little tab (third) above the list on the expanded right bar that leads to Frame Styles there is a Graphics option (covers Images) there where I can change the position but it doesn’t seem to want to apply to images I drop and drag.
:confused:
It looks like it applies it when I drag it in but then if I touch it at all such as the size it reverts it to “from left” which effectively leads to it going haywire. If I protect position then it over protects so I can’t move it at all.
:confused:
The images are all the same size and tend to be oversized. It is at least useful to set this style by default relative to paragraph for width and height at around 10% to 20%. I also find that turning on format markers while ugly mitigates some issues with visibility. This is a particularly useful default setting to be able to change.
:confused:
Occasionally when dragging an image in the page jumps around awkwardly. This seems to be worse with the sidebar up. I think wrap after makes sense as I want them pinned to the left and not in the middle of text but I’ve just not see where the optimal option leads to bad wrap so far. It’s a safety setting.
:confused:
While this seemingly improves things I still have a lot of awkwardness. One image seems to not move or maintain alignment. It also doesn’t automatically flow onto the next page instead overlapping the margin. I also have one part on default paragraph style and another part is on body text and I have no idea why there should be any difference or how this happened nor how to fully fix it. Setting the paragraph style to default improves it a bit.
:confused:
This is having an impact of that one of the primary points of a software tool like this is a timesaver but it induces the opposite effect while in theory having the potential to save time but being unable to achieve that. One image when hitting the boundary of the page refuses to move down. Inserting a vast amount of newlines appears to eventually push it down. It’s a bit awkward as I am afraid to touch the image as it immediately removes the settings for position and does a lot of weird things including disappearing under another image.
:confused:
It creates a situation where the word processor will be barely usable or unusable for some people for what should be a simple task. I think that I have actually gotten it streamlined to a point where it is enhancing my workflow though it also puts me in a situation of living in fear that something else is going to break with some hidden aspect to it that is going to interrupt and send me on a wild goose chase though at least it supports undo so long as weirdness is noticed quickly.
:confused:
I also have no idea why this edit isn’t showing paragraphs properly. In this reply I had to use faces to force a newline between paragraphs so it’s readable.

This is the latest version with native ODT as the format. The OS is W10 but that should not impact anything in this case.
:confused:
Even with my improvements I am still finding it awkward to just drag in images with each time requiring inserting a bunch of newlines (might as well copy and paste them) as well as more fiddling. The page still jumps about when I drag which is really awkward making for a moving target. I think this is because it’s previewing as you drag rather than just painting the target like with the point I made earlier where it’s not just showing something like a caret for where it goes though it shows the anchor thing that isn’t really helpful.
:confused:
I think at this point the usability issues are mainly in the way drop and drag works. This feels like it’s an advanced feature, more complex than the simplest approach that’s on by default and stuck on that for a lot of people does something they don’t need (active preview).
:confused:
To understand what this is like, imagine trying to walk down a sidewalk or pavement where each slab is a conveyor belt going in a random direction, left right up or down. Imagine trying to just stand still or put your bag down. Imagine putting a piece of paper on the desk and trying to write on it while someone keeps moving it around.
:confused:
That and something is still weird with the way images work in respect to text flow with things messing up for images below when inserting above so I keep having to go and fix the pages below.
:confused:
I do have some expertise in this area as someone who has developed software with a lot of experience in usability and documentation so might be able to express this issue in a fair amount of depth. Though it might be easiest to see the issue just trying out the task in different word processors. It would be very annoying if over the top software patents are making it harder to improve these things.

The correct spelling is: drag & drop.
I think you should decide:

  1. whether LibreOffice is too complex for you,
  2. or whether you want to learn how it works
    For 1. you should consider another program.
    For 2. you should aim for a steep learning curve.

Here are a few practical examples:

https://wiki.documentfoundation.org/Videos/Insert_and_edit_images_in_Writer_Part_1

https://wiki.documentfoundation.org/Videos/Insert_and_edit_images_in_Writer_Part_2

https://wiki.documentfoundation.org/Videos/InsertingImagesInWriterPart3


English documentation


Here is a possible procedure:



118733.odt (9,9 KB)

Does not mean anything because platforms evolve at their own pace. And even on designated platform, the “current” release depends on the OS version. What is installed also depends on whether you automatically or manually update. So quote exactly the release number reported bt Help>About LO.

It matters because there are difference between platforms to account for the different libraries (e.g. widget management is not the same on W$, MacOS and Linux; but other areas also differ, such as menu structuring and wording).


From your comments, you seem to prefer drag’n’drop or manual interaction with objects. This is called direct formatting. Contrary to common belief, this direct approach us not “intuitive” at all. It is the result of long conditioning by the major actor who does not provide consistent, predictable and reliable ways of formatting.

Writer is based on a very sound and well-specified principle implemented with styles. But, to “ease” switching from M$ Word to LO Writer, direct formatting is still possible and takes precedence over styles.

This means actions created by direct formatting override the corresponding formatting brought by style. It relatively innocuous on paragraph and character (though always playing nasty tricks on your back) but becomes catastrophic on frames and images. Manually resizing, cropping or even moving a frame/image immediately blocks the frame style and creates an unmanageable situation, like your images jumping in unexpected locations.

Inserting a non-text object, like an image, inside text is not a trivial operation at all. The image interacts with normal text flow and disturbs the text rendering algorithm.

You must tell Writer how the image will co-exist with text. You can, of course, do it manually (direct format) but you’ll lose auto-repositioning and “cooperation” with text when you edit it. The best way to do it is with a frame style which holds all interaction parameters. The way to use a frame style is as follows:

  • click somewhere to define the anchor (usually at start of a paragraph)
  • insert your image
    Do it as a paste operation, not drag’n’drop which will already stain it with direct formatting. Don’t care for where the image really lands. Don’t touch it. If it is not selected, click on it.
  • apply a frame style
    All parameters are transferred to the image which will be relocated to its expected position.

At any later stage, refrain from moving or resizing your frame/image manually. If you want to change something, modify the frame style.

Another “golden rule” is to prepare your images outside Writer. Writer is not an image manipulation program. It offers low-end graphics operations for quick’n’dirty experiments. Prefer a dedicated program through which you image will be cropped, sized and scaled to its final properties so that Writer has nothing to do, but take the image as is.

Now what remains to be done is to create a properly crafted frame style. This will be the object of my answer. Please, allow me some time (short I promise) to deal with my hot stuff.

While some of this is helpful, it also really only goes to show the problem. This is a word processing case but then writer is doing something or covering some niche I know not what. It’s sort of like it’s going off and doing its own thing rather than what you might expect from an office suite.
:confused:
I have to question the use case. I am trying something that should be a common and simple use case and that ought be quite simple but it is not. If there is a way to do it either I’m not searching for the right thing or it’s not really well explained.
:confused:
I think at the end of this thread I’ll go through and take the bits and pieces I have learnt to make a small guide for such a use case though it probably will still be very wanting. The steps you are giving and the information is really rather a lot for something that I don’t think need be so complex or difficult. I’m not entirely certain why it is. Perhaps it’s something going on under the hood or a conflicting design philosophy.
:confused:
At the moment I’m sticking to what works well enough just about though painfully in some cases as I need to get what I am doing done with as little overhead and distraction as possible. However, after that I will likely explore more to attempt to improve things in terms of working out how to get it to behave sensibly with content. I especially do not want to mess up what I’m working on.
:confused:
I don’t expect it to be reprogrammed and changed overnight but I think it is also useful for developers to consider use cases and user experiences including acknowledging feedback on issues such as this. I think that sometimes open source projects struggle a lot with project management due to the nature of receiving sporadic contributions. It’s sort of directionless and ends up all over the place. A common problem that can’t be fully rectified but can be mitigated. Contribution is not just about code, user feedback about higher level issues and user experience is also important.
:confused:
In a normal situation the way I would expect it to work is at least according to standard conventions such left to right and top to bottom is something like this:
:confused:
There’s a single dimensional array of content. If you insert something then it just works like a bookshelf. Items consist of images and text. You put it between the books and it doesn’t disrupt everything between. When dragging an image I would expect it to by default be positioned according to text content so it would show a cursor as the insertion point. I wouldn’t expect it to be also setting of set or secretly changing positioning mode into a kind of multimode setting position in relation to content as well as pixel offset (seems to be doing too much at once) when dragging unless either manually changed or holding a control key. Once dropped then I would expect the range of default behaviours be quite simple. Break before and for the image to go into a new line is the default behaviour with a newline starting at the top right of the image aligned and after that text flowing to the right of the vertical overflow as though a left margin just making the bounding box narrower. I might then expect some behaviour for the end as to how to handle the remainder if the image doesn’t happen to be aligned to overall line height. I would expect the bottom of the image to adhere to the bounding box of the page (perhaps that is the frame or rather within it). This bit is always a little tricky to be fair.
:confused:
It’s very crude but something like this. Instead it really keeps wanting to detach. I suspect there might be some issue with the internal model and how that reconciles different content perhaps taking two separate streams and attempting to then have them play nice in the same courtyard.
:confused:
Even with my settings so far it still does things that are awkward. At least three. One of them is like if you had two conveyor belts, one for your luggage and one to stand on. Only instead of them both going along together to keep the two aligned sometimes one stops, then the other stops, then one starts and in ways that really don’t seem intuitive. There is a detachment of things that should be more attached.
:confused:
It seems as though paste and cut/copy is being used by a lot of people to work around these issues instead which isn’t always ideal. In my use case internally I’m not using a lot of drag and drop apart from dragging images into the document.
:confused:
I am not using it for any sophisticated image editing. Resize/scaling is so basic I would expect that to be the one feature included as standard in a word processor. The use case I have is really simple. I just have three sections each with a header for the category of image. Then I just want to drag the images in and then write text next to them describing the image.
:confused:
I suspect there might be other ways to do this such as perhaps trying a table with two columns (this is not ideal as at some points I want to move them through sections ideally in one go with text and image together).
:confused:
For certain workflows people want to use tools like this for a streamlined process. I am doing something quite intellectual. Part of the process requires when inspired to write a story for an image doing so quickly while it’s still in my mind and I still remember it. There are various drawbacks to using web based tools for this, especially with drag and drop.
:confused:
This is probably not really unique but more common than people might realise since half of the time someone is using a word processor they’re offloading from their volatile memory to something more persistent. If it is constantly dragging them into having to some other exercise stealing their concentration away then it’s very often not going to be an effective tool.

Sorry, but you’ve choosen a big office suite as “your” tool. So I would expect complexity.
.
Your second “error” is not unique. “My expectations should be everybodys default.” This is simply wrong. A lot of documents are done with “floating” images and references.
.
So without some reading, and setting up your personal defaults you will hardly get a positive experience.
Because of history we have several ways to handle input (and I like the choices LibreOffice gives me), but one has to decide, wich to use…

To help improve the congeniality of your online communication skills you might want to put this prompt into ChatGPT:
:confused:
Can you explain why this response to someone raising an issue regarding usability and excessive complexity on a support forum might be insulting:
“Sorry, but you’ve chosen a big office suite as “your” tool. So I would expect complexity.”
:confused:
There is another way to interpret your logic and that is to the effect of that LibreOffice is a bad choice. In some respects it is and that’s a problem when I don’t think that it should be. Is it complicated or over complicated? There’s a big difference.
:confused:
Legacy issues are sadly a common drama though in this case I would be surprised if it were so bad that the answer is to start again.
:confused:
I suspect the area of usability has been sorely neglected. Perhaps there are other priorities. Perhaps there are other more important things to address first. I would like to hope that those things would at some point be addressed. In the meantime I think simply just accepting that it isn’t yet all that you would want it to be and being humble about it fosters a healthier forward looking atmosphere.
:confused:
At some point there should be some time to consider usability more along with a build up for that. Taking simple commonplace use cases like this to establish a testbed might be useful.
:confused:
I have described my workflow and use case in a somewhat fragmented manner but it could be concisely specified as it is not all that complex.

I’ll try to give a recipe to achieve your “use case”.

Requirement

Illustrate a paragraph with an image at left of it. Example:

+-----+ Lorem ipsum
|     | dolor sit
|     | amet,
+-----+ consectetur
adispiscing elit.

The image may be be taller than the paragraph, in which case it is allowed to move right the next paragraph.

Solution

The solution is based on a frame style.

I suppose here you have a single type of “illustration”. In other words, all your images are positioned the same. Then a single style is sufficient and I’ll customise built-in Graphics (which is the default style applied when you insert an image).

  1. right-click Graphics in the side style pane (frame style view) and Edit Style
  2. Type tab:
    • Anchor: make sure it is To paragraph
      This attached the image to the paragraph. When the paragraph moves, the images follows the paragraph.
    • Size: set some sensible width and height minima like 2cm and tick both Autosize checkboxes
    • Position:
      • Horizontal: Left to Entire paragraph area
      • Vertical: Top to Paragraph text area
        There is a subtle difference between vertical and horizontal reference. Vertical uses Paragraph text area so that the image alignes with text and enefits from paragraph spacing above. Horizontal is Entire paragraph area to align on left margin, not on paragraph left indent.
      • tick Keep inside text boundaries
        This is very important. If this image is too low in the page, it overflows inside the footer and may even be clipped by page edge. Ticking this parameter will force Writer to either send the paragraph to next page (if it is small) to avoid overflow in the footer or send the image to next page if the remaining of the paragraph is high enough on next page.
  3. Wrap tab
    • Settings: After (so that text can wrap at right of image)
    • Spacing: set Right to 0.5cm (adjust to your taste) so that there is a small gap between the image and text
    • Options
      • untick Allow overlap
        Writer will then move (rearrange) conflicting images to avoid overlap. Conflicts may arise with images attached to small consecutive paragraphs or with several images attaches to the same paragraph.
      • optionally, tick First paragraph to prohibit the image to extend into next paragraph(s)
        Enabling this option creates spacing if needed between current and next paragraph.
  4. Borders tab
    If you want a border around your image, set the parameters here.

Procedure

Type your text as usual.

Paste an image in the paragraph you want to decorate.

Don’t move, resize or touch in any way your image. Doing so adds direct formatting which cancels any effect of frame style. Always play with style settings. All images are affected at once. But, never, never, never touch them with the mouse.

In particular, prepare your images outside Writer so that you need not crop, scale or resize them which would count as direct formatting and make you lose all benefits of frame styles.

In case, your image does not look or position as prescribed by the style, apply Formula style, then immediately Graphics. This should fix the issue.

Frame styles are fragile in Writer. The feature in itself is difficult to understand and hard to master. So, don’t use Writer as an image manipulation program. This will add to the difficulty. For your peace of mind and sanity, use Writer only as a text processing application.

Remember that frame styles don’t exist in other format than .odt. If you save DOCX, you lose everything.

You may find the frame style feature very complicated. It is, but it allows for very sophisticated layout with automatic adjustment. There is a high price to pay for it. It took me years to tame the feature and even now I meet problems. Don’t be discouraged by your initial failures.