Spinoff from this thread to save hijacking it.
You have something to write - it could be a report, a manual, a novel, a memoir or anything. How do you approach your writing? What tools and methods do you use? What do you consciously NOT do?
Frustrated novelist here…
I ease into my comfort zone, each evening, by going back about four pages and re-reading and re-editing them. That builds up a kind of momentum, so I can go forward from the end, adding four (or more) new pages.
I’m not a fast writer… Four pages a day makes a novel in a few months…
Music in the background helps.
I turn MS Word’s automatic spell checker OFF. I don’t want to be distracted by the little red squiggly underline that comes with errors. I’ll fix those later. Distractions are death to the writing trance.
A nice cuppa tea or coffee helps.
My bete noir is using the same word too closely together. “The base was protected by razor-wire and coils of concertina wire.” The redundancy of “wire” makes my teeth ache. A hell of a lot of these slip past me unconsciously, and I have to edit them out in re-writing.
(Hey, Jack Vance let one of those redundancies slip through in “The Last Castle,” so even Homer nods!)
My own writing (aside from posts on this board, which I generally just make up as I go along) tends to consist of:
Blog posts/articles on my website. I write these inside the CMS and they tend to be pretty linear and self-defining. I take a series of photos as I make something, then the blog post is just a sequential list of photos with commentary alongside, interspersed with a few blocks of ‘theory’ - especially for articles that play out over longer periods of time.
User Manuals for technical/business processes in IT, and reports on incidents, procurement recommendations, changes, etc.
For all of these business documents, I now use Outline Mode in Microsoft Word (don’t hate me!) - I learned this on a report-writing training course a few years back - and it has significantly improved my output.
The thing I like about Outline mode is that it imposes a hierarchical model on my thoughts and words - and this is how I like to do my thinking - you write the headings, then the subheadings, then the paragraph headings, then the content.
This, I find, is a great way to structure a document so that there is nothing repeated or left out.
For example, If I need to write a report on a service outage, I start with the main headings:
[ul]
[li]Executive Summary[/li][li]Technical Description of Failure[/li][li]Timeline[/li][li]Preliminary Analysis[/li][li]Recommendations[/li][/ul]
Then this gets broken down into (in this case, for an event spanning more than a day):
[ul]
[li]Executive Summary[/li][li]Technical Description Of Failure[/li][li]Timeline[/li][LIST]
[li]Monday[/li][li]Tuesday[/li][li]Wednesday[/li][/ul]
[li]Preliminary Analysis[/li][ul]
[li]What Went Wrong[/li][li]What Went Right[/li][/ul]
[li]Recommendations[/li][ul]
[li]What We Can Do Now[/li][li]What We Should Do Soon[/li][li]What Cannot Be Done[/li][/ul]
[/LIST]
Only then (when I at least have all of the main and sub headings from start to finish) do I actually start to fill in the content in each section. In some cases, as that content flows out, it turns out to require further division, so it might end up looking like:
[ul]
[li]Executive Summary[/li][li]Technical Description Of Failure[/li][LIST]
[li]Hardware[/li][li]Software[/li][li]Monitoring[/li][li]Standards[/li][/ul]
[li]Timeline[/li][ul]
[li]Monday[/li][li]Tuesday[/li][li]Wednesday[/li][/ul]
[li]Preliminary Analysis[/li]
[ul]
[li]What Went Wrong[/li][li]What Went Right[/li][LIST]
[li]Internal Response Times[/li][li]External Supplier Support[/li][/ul]
[/LIST]
[li]Recommendations[/li][ul]
[li]What We Can Do Now[/li][LIST]
[li]Remedies Already In Progress[/li][li]Costs And Effort[/li][/ul]
[li]What We Should Do Soon[/li][ul]
[li]Costs[/li][li]Timeline To Deliver[/li][/ul]
[li]What Cannot Be Done[/li][/LIST]
[/LIST]
I find this approach works for pretty much any technical context. I guess it’s a bit like a simple template at the top level - but I just find it to be the best way (for me) to impose clarity and ensure that I don’t miss anything.
If I just sit down and try to write the content from top to bottom, it ends up with gaps, inconsistencies and repetition, requiring a lot of laborious editing (the end result of which is always the structured document I should have written in the first place).
I think of what I want to write and then I write it. No full structuring beforehand: the detailed structure comes during the writing itself.
There have been times that a Very Frustrated Boss has told me I was not “writing right” courses and manuals because according to them I had to begin by defining what the index would be, and only after that was done should I write the text, and only after that was done should I add any screenshots.
My approach is kind of the opposite and I’m very, very grateful to whomever came up with the “automatic index” features in document processors:
I do the process I want to explain. As I do it, I take screenshots, edit them and paste the edited pictures into word or OO (never “paste and edit”, this is a leftover from when this second approach led to much-bigger files), writing around them and inserting headings as appropriate.
Once that is done, I add any appendices (Glossary, repeat of any Tables which are included in the text).
Then I go back to the beginning and add an Intro.
And then I click Insert Index and the Index appears! It’s magical! (Hey, back when I wrote my “undergrad thesis” I had to write the index manually, and update the numbers manually)
Then I review the document looking at it as if I was translating it. I discovered this helps me see both typos and places lacking clarity.
If there have been any updates, I refresh the Index. And the titles and the page numbers change by themselves! It’s magical!
A lot of my colleagues, for the same types of documents:
Do the process, collecting unedited screenshots into the document as they go.
Do the writing in plain text.
Edit pictures.
Add headings.
Add Index.
Different from mine, but still upside-down from that “right process”. I suspect that “right process” is taught by the same people who teach how to describe people by going over a checklist of physical features in a given order.
I imagine this approach might not work very well for a novel (at the very least I’d need to have a notion of who’s going to be alive at the end, shouldn’t I? Same for training materials I do know what the end result will be), but for courses, manuals, emails or posts it works ok. And by the book can kiss my ass.
I work on paper initially - particularly for a new book or major area of interest which may end up written about in any number of formats. Basically, any topic which I think might become sustained gets an exercise book.
I get a 5 mm grid exercise book and cover it with wrapping paper and clear contact so I can easily identify it. I have nearly a hundred of these now. Same colours are same topic - all labelled on the back.
I number the pages and then put all my main points in the contents at the front. I put scribbles, paste things I have printed, lists - everything in place with little squares (hence the 5 mm grid) next to each idea or item. Lots of bits and pieces tend to get pasted in so the exercise books often end up bulky. And the pages make a very satisfying crinkly sound. Each item gets ticked when they have graduated into the written piece. Sometime I note where they were used, cross-reference and so on.
I will then start writing on the computer, but the exercise book comes everywhere with me for random scribbles and thinking. I have a pen hooked over the cover.
I have a new book coming out next week. I will now scour the relevant exercise books for anything I didn’t use, or want to use again, and blog them. The system means that no ideas are lost in editing, but I am not wasting time refining the writing by hand.
Oh, and I use voice recognition software, so I often just waffle on about my notes and it becomes the basis for the writing. It works for me.
That’s repetition, but not redundancy. Each of the two “wires” in your example refers to different objects: in redundancy, the repetitive word refers to the same object or entity and has no additional functionality: “a roll of razor-wire wire”.
For most articles and blog posts I usually just start at the beginning and continue to the end. I find that getting it right the first time makes for the best results. Of course if I don’t get something right the first time I go back and improve it. And sometimes it’s hard to get started, so after a couple of false starts I don’t let that bother me and go on even if the beginning isn’t good and needs fixing later. There’s nothing worse than being stuck staring at a blank screen.
For long articles or books I collect topics I want to cover, put them in the order that makes sense and then write those sections. For my technical books I would first start creating configuration examples and/or screenshots and then explain those in text.
For articles sometimes I write first and find images later, but then the text and images tend to gel to a lesser degree.
For a long time I resisted spell checking as you type, but I use it now and it no longer distracts me. There’s a certain level of attention to spelling/grammar/layout detail that I can afford before it starts detracting from the content; both completely ignoring those details so there’s a lot to be fixed later or obsessing over everything so it’s right in the first draft are counterproductive for me.
I usually work in Pages on the Mac or Word but I’m not exactly a fan of either. The advantage is that they support styles, which is a necessity for longer documents. For shorter stuff I often write HTML by hand which is annoying but I haven’t found anything that’s easier.
I compose nearly everything in a text editor, including any verbose multi-paragraph posts to SDMB or Facebook. I do my best writing when it’s just me and plain text.
For LONG projects like books I keep a “notes” file in which I"ve outlined what I want to write about moving forward from the point I’ve written to, and I edit that when the ideas come to me then later work on the manuscript itself, using the notes as a guideline.
I don’t write too much, I’m not very good at it.
For fiction I start out in broad strokes, writing out plot points, then adding more in between to bridge them, then more in between them, etc. Occasionally I’ll see the need to rearrange. Somewhere in the middle I’ll write out a list of characters by name and a brief description.
Basically it starts out simple and gradually gets more and more detailed, and then it reaches a point where it’s time to write the actual story, and I’ll always start from the beginning.
For non-fiction I just wing it, and edit once it’s done.
For technical writing I use an outline of the type you outlined. That’s if it amounts to more than a paragraph or two. A dozen times a day I have to communicate with others with very brief descriptions of issues and more detailed follow up occurs when everything is resolved. When it comes to fiction I start writing and something comes out less than 10% of the time or I get nowhere. I asked about the other writers about this in a CS writing contest where I realized I just don’t conceive of plots, I just know how to write descriptively but I’m not a plot person.
For my own amusement: Vim. I don’t need fancy stuff. Just plain text. This is by far (and I mean by far) most of it.
For “export” to others: LibreOffice. A handful of times/year.
For munging other people’s code: Notepad++. Very bursty. None for months then I have it open nonstop for a week or two. At some point, I might get so used to it that I’ll start doing general writing in it.
I learned a technique called “effective writing” back in the 70s. The concept is this:
- Sit down and write whatever you want. Keep writing until you get to the end.
- Go back and edit.
I’ve used this since I learned about it. Works fine with fiction; many of my ideas came from just writing the dialog and seeing what the characters say. Or thinking about the worst thing that can happen to them at that point.
I prefer Word, but would switch if I needed to. Google Docs is adequate, but misses a lot of the features I use often.
Yes, my approach has always been to simply write, then go back and edit. This has worked for all of my writing thus far, but granted I’m not writing 50 page dissertations or anything. It did mean I could write papers in high school and college quickly and just do a once-over for typos and a poor sentence here and there, pretty much.
These days I only really do writing on forums and on outlining the story for my webcomic(s). I have one notepad file that keeps all my random ideas for a comic, one notepad file with the summary of the story, and one notepad file for the entire script. Well, currently, sometimes I start a new one for the script simply because a huge file gets tedious to look through, and after the script has already been drawn and set in stone by being published online, it’s not like I can go back and change it anyway. So I just periodically cut off the old script into old files and keep going with new stuff. With the next comic though, since it’s not publishing for a while yet I’ll probably just keep all the script together as one file so I can edit the whole thing as needed.
I find very little use for advanced word processing these days since all the text I write isn’t published through it. I haven’t opened Word in years.
Use notepad++ for all my code though.
For pretty much everything I write, including SDMB posts, I first have to spend some time thinking and figuring out what I want to write, before I start actually writing it.
I just realised that my structured thing in Word is really just a variation on mind mapping, which is something I used to really love, and still tend toward if pen and paper is the starting point.
I write lots of different things.
My column has to be about 550 words. I read the abstracts of the papers in the issue the column is going in, and try to find a hook. Tricky since I usually don’t know that much about the subject. For instance, for the special issue on 3D packaging I wrote a version of He Built a Crooked House on 4D packaging in the future. I write some jokes in my head, and after 20 years of this I can pretty much hit 550 words the first try.
For papers I sketch it out, and then write a section at a time.
When I was writing my novel (which turned out to be two) I wrote everything in my head, and was able to write the chapters out of order - first chapter, then last chapter, then ones in between. This was useful since I know what hints to put in.
Now I’m revising, I do a read through where I read aloud, since I find that infelicities show up this way. Then I search for words I use to much or may represent something I should revise - like “it” to make me be more specific or was/were to get rid of passive phrases or “ing” ditto.
My wife has taken to creating a pdf of a section and sending it to her Kindle, since she finds she can pick up typos better that way than using paper or a screen.
I also find that reading on a screen, reading on paper and reading aloud all pick up different things.
It is good to make up a checklist of stuff you want to look for. Searching for words like “were” in Word is good since it kind of forces you to deal with them, where you might be able to skip over them while reading.
This is pretty much what I do—however, I generally use some TeX-editor, since most of my stuff is technical and will need formulas, which are basically impossible to get to look right in Word or similar programs. I tend to think of it as ‘roadmapping’: I start out with a globe, a really rough map of the territory I want to cover; then I break it down into maps of the individual regions I want to visit, which have more detail, so to speak upping the resolution one notch; then I look at the detailed towns and villages with their precise street names and so on.
I find that I approach not just writing, but also learning this way: get something that gives me an overview of a topic, then delve into individual sub-topics to get a clearer view, as interest and time dictates.
In terms of “literature,” it’s a redundancy. It’s a bad repetition. One of the two has to go.
I know a chap who wrote (and published) a novel. He wrote chapter one, went back and revised it, then wrote chapter two… And went back and revised both chapters one and two. Then he wrote chapter three… And went back and revised all three chapters…
This would have worked out okay, except he ran up against the publisher’s deadline, and had to hasten through to the end. Thus, the first half of his book is VERY finely polished…and the second half is rough and kind of coarse in comparison.
(But what an elegant first chapter he ended up with!)
I write scientific papers which can be split into the structure, or story, of the results - critically important; and the actual writing of the text, which is less important. There is obviously some interplay, as well-crafted prose will re-inforce a good structure, but it’s very important not to get bogged down in prose issues when writing a paper. Finding the right narrative, though, really occupies my mind.
A brilliantly written paper at the sentence by sentence level, having the wrong context and / or weak structure, isn’t going anywhere good. But Science and Nature are full of blandly written pieces that have the right structure.
That being said, crafted prose is something I care about, personally, but the scientific community is so broad and international, and journal production editors so basic, that scientific prose gravitates to beige. It is a fun challenge to try and write in an original way in this environment - you have to be quite subtle as big rhetorical flourishes just aren’t the name of the game. Like yelling in a cathedral, no one wants to hear it.