Testing Plume Creator

by , under writing

In a comment to my previous post Elizabeth (thanks!) pointed me to this post from a writer who works with the developer of Plume Creator to make it into his ideal writing machine. I’m willing to give it a spin.

It wasn’t all that straightforward to get it installed, because it came in .deb and openSUSE prefers .rpm (or even the package manager’s native .ymp), but fortunately I know people who know things, so someone recommended alien to convert the file. This was easy: alien -r filename.deb results in filename.rpm, which the package manager will happily install.

I have nine and a half chapters of novel lying around that suffer from a shortage of plot, and come to think of it structure, though there’s some pretty good writing in it and I don’t want to abandon it altogether. Ideal candidate to test Plume Creator with, in other words.

The software seems to impose structure. I was fighting it at first, “but I don’t write like that!” — I’m usually very bottom-up, just start writing somewhere and it falls into scenes and chapters all by itself. Here, it looked like I had to have chapters and scenes even before I had words. Fortunately I already had some chapters and scenes, so I could paste them into the IDE. Because that’s what it is: kind of like the HTML editors I’ve tried and abandoned in frustration because I don’t set up a page before I have content (which seems to be the done thing) but have some content and make a page around it.

I’ve only imported existing text, not yet tried to write anything new, so I can’t report on that. Also, I’ve only been pasting plain text from Kate, not tried to import .odt from LibreOffice. I wouldn’t know how to do that anyway, Plume Creator only has an “export” option, which I haven’t tried yet because I’ve discovered it only now.

The first thing I noticed was that all paragraph breaks were GONE. Plume Creator (henceforth abbreviated to PlCr) completely disregards all whitespace except one space between words. It wasn’t long until I took to typing XX between paragraphs so I could separate them again in the PlCr scene box (“sheet”, apparently the term for a scene with associated notes, synopsis, and attendance).

When I made a project called Ardyth, I got a helpful tree-stub with “Ardyth” on the first level, “Chapter 1” on the second, and “Scene 1” on the third. I could conceivably type in the Chapter sheet, or even the base sheet, but I don’t know what happens when I do that (1); I’ve only been importing scenes into scene-level sheets.

(1) I have in fact made a project for a short story I’m working on, and that makes me think I should be able to get rid of the Chapter level for things that don’t have chapters. I haven’t found a way to do that, though, perhaps it hasn’t been implemented. PlCr seems to be optimised for novels anyway.

(Screenshots follow. Click to open full size.)

Plume Creator main screen

Main screen

My laptop has a wide screen and I don’t like to run applications full-screen, but PlCr almost asks for it. All the extras left and right of the text can be detached (like dockers) or hidden, but detaching them makes them sit on top and obscure the writing, and hiding them makes them go away, which isn’t always what I want. It’s easy to get them back with the tabs on the left, though.

Let’s take the left-hand column. The docker with the system clock, the timer and the word count is one block. I don’t need the system time at all: there’s a perfectly serviceable clock on my panel, not to mention my watch and the wall clock. The timer is a countdown timer: it doesn’t keep track of how long you’ve been writing, but allots a period to write in. I haven’t used it yet, and I doubt if I ever will. Perhaps when I’m producing wordage again rather than recording characters and places and editing, so I can keep myself from writing for hours on end without stopping. Though when there are other people in the house I don’t keep going for hours on end anyway: I’m being interrupted all the time, usually in the middle of a paragraph, by someone who wants something or by wanting something myself for that matter. (Saturday, when I was alone in the house, I did keep going for five hours without stopping, and only then realised that I needed to answer a call of nature and that I was hungry.)

The wordcount is interesting: when you click All you also get the wordcount of the project, the current book (the same in my case, there’s only one book in this project) and the current chapter. I tried to take a snapshot of it, but either I fumbled or KSnapshot can’t do that.

(Now I’m discovering that you can put any docker anywhere; even with the tree next to the tools. It makes the tools docker look very ugly because it will flow to the height of the window. Hmm, may experiment with that some more.)

So, the tree. It started out with Ardyth – Chapter 1 – Scene 1 but it isn’t willing to call further chapters Chapter 2 etcetera, I have to do that by hand– not that I mind. New scenes get called New Scene until I call them something. My scene names are all over the map, from a bare pointer (“vigil”) to description (“Sedi reads bestiary”) to frivolous commentary (“Proposals of Marriage”). I think this is a feature rather than a bug, it’s only for me anyway. I think the export function can export the scene titles, but that’s a checkbox that doesn’t have to be checked. There’s room in the tree for additional levels, but book – chapter – scene works for me at this stage so I haven’t explored it further. And it’s easy to move whole scenes and chapters around: all branches and twigs are draggable.

Now we get to the middle column: the actual text. It took me a while to discover that I could have PlCr in Plastique (KDE) style, and that I could specify the width of the text column, so I got a bit impatient with a text column as narrow as the scene tab with lots of unused space –this seemed to occur only in the default Gnome style– or so wide that I had to make PlCr fullscreen and scroll. The word-processing component is (still?) a bit primitive, but as it’s an application for writing and first-stage editing (see my previous post) and not for polishing that doesn’t matter much. I can have italics, though for submission I’d probably have to change that to underline (or don’t publishers want that any more in this age of word processors?) Can’t have styles but if I have each scene in its own self-contained sheet thingy I don’t miss them. I may change the font to Sans, though, writing rough draft in a serif font makes it look far too finished.

The tabs at the top are all the open sheets. I find myself periodically closing a lot of them because every time I click on something (to check a name or something) it stays open. Even the top level is open sometimes, though I don’t remember ever consciously clicking on that. I haven’t found out the logic behind the order of the tabs: they are sort of in order of opening, but sometimes one creeps between the others, not necessarily in the order in which the scenes are in the book or even in the chapter. I find myself not using the tabs at all, clicking on a scene in the tree when I want one, but that may change as the monster grows ever larger and I don’t have all the scenes in my head any more.

The buttons at the bottom right are a mixed lot. Show Prev. Scene gives the previous scene above the current one, so it’s easy to see if a transition works. (I’d like a Next Scene button, though, for editing.) Fullscreen Edit gives the text, and only the text, white on a black background. It’s possible to get synopsis and notes boxes on this screen. I don’t think the white-on-black is configurable (yet?), the options only have “Show scrollbar”. The Outliner button doesn’t do anything yet, except say that it’s in heavy development when clicked on for the first time. (The same goes for search and replace: it has great promise, but it’s still a mockup.)

Notes and synopsis dockers go in the right-hand column:

Plume Creator with synopsis and note dockers

Expanded view with synopsis and notes

I sort of hope that this will teach me to keep a running synopsis as I write, because I’m exceedingly bad at synopsising after the fact. I might even keep the synopsis docker open at all times when I’m adding wordage again.

I’ve kept the best for last:

This is TEH AWESOME. Keeping track of who is in a scene, where the scene takes place, and even objects important in the scene, and all of that stored with the scene. Better still, the expanded version where one can add people, places and things, and give a person, place or thing a page of notes:

Expanded attendance manager

By the way, I love the term “attendance manager” for this, as if it’s a roll call. The arrows between the full list (left in the first screenshot, centre in the second) and the scene list (right) put a person, a place or a thing in the scene and keep that information with the scene.

I can think of some improvements. The dropdown list on the left now only has “Main”, “Secondary” or “None” for categories of characters, and the one next to that “None”, “Protagonist” (causing bold type in the list), “Supporting”, “Neutral” and “Antagonist”. These are global, so you can’t have the protagonist in one chapter being the antagonist in another. The categories aren’t editable either. Also, I’d like a way to indicate who is the viewpoint character of a scene. At the moment that has to go in the notes or even the synopsis if I want to indicate it at all. I’d also like to have a way to distinguish between characters present in the scene and characters mentioned in the scene (who may be present in another scene, of course), and to indicate characters who are important in the scene but happen to be dead. (I do this in the character details now, but haven’t found a way to have them alive at the beginning but dead later– not in the attendance manager anyway.) It would also be nice to split up the list into groups of people/places/things that belong together, like “Ardyth’s family” and “Order of the Sworn”. Very much a database issue.

And I would like to click on a name or a place or an object (or a group) and get a list of all the scenes that entity is in. When adding a name, you get “First name” and “Last name” fields, and this is where the Frenchness of the developer shows: suppose I’d enter my own name, first name Irina and last name Rempt, I’d show up in the list as Rempt Irina. People in Valdyas don’t usually have a last name, but they do have a matronymic, and to get Rythei Leva in the list I have to enter her as Leva Rythei. Disconcerting, but it works, sort of. This could perhaps do with some improvement too, because people’s names are not consistent.


I can work with this, I think. It takes some getting used to, but I like it a lot now that I know some (not all) of its niceties. It may force me into a more structured way of working, which might be a good thing. I’ll see when I’m done importing and labelling if I can write new wordage, and if the structure stifles me or inspires me to think both inside and outside the box.

  1. Cyril Jacquet

    Hello Valdyas,

    Thanks to you, I now know Plume runs on Fedora !

    I listed your reported bugs and comments. In the next release of Plume, v 0.53, you will be able to paste plain text. Sorry for that. 🙂

    A quick note : to rename scenes “scene1”, “scene 2”, etc… right-click the parent chapter in the “tree”, go to “advanced” and click on “auto rename children”

    The Attendance manager isn’t finished and you gave me a lot to think about in your other post “Attendance”.

    Cyril Jacquet


Leave a Reply

  • This site uses Akismet to reduce spam. Learn how your comment data is processed.