Support for Indie Authors discussion
Writing Process & Programs
>
Atom and git
date
newest »

message 1:
by
Jim
(new)
Apr 07, 2019 10:53AM

reply
|
flag

if you like the Snowflake method of writing, Novel Factory is a good using that format.
If you are needing formatting, you can write in almost any program, save as .rtf or .docx then run it through Draft2Digital to format it using one of their templates. I would highly recommend learning Scrivener and using it. It's worth the $40 for the program. If you like mind maps, Scrapple will integrate with Scrivener and is easy to use. I ended up putting out the cash to learn Scrivener and how to integrate it with other things I use. This is coming from a person who had avoided Scrivener. If you have a MAC computer, you would have the latest and easiest version to use.


I am not at all being argumentative. I would just love to find a tool that I can use on all the different platforms I have.
I have tried Google, but it does not seem to keep the formatting the way I want it.
Admittedly, I mostly write poetry, so that has some different demands on formatting than a novel or some such.

I have often considered using git for my revisions but if you are the only person doing the writing then it really is overkill. Git works really well for collaborations, you see.
That being said, when I do a re-write, I tend to either redline or simply save the edits as a different version if I think about it, that is. Mostly, I don't see the point in keeping previous versions unless I didn't already have a clear idea as to the rewrite. But chances are, if I'm returning to any given section bc I found something awkward or wanted to modify, keeping the previous version around for grins & gigs will only serve to second guess myself and I don't want to do that when I'm writing.
As for atom & markdown, unless you are publishing directly to a webpage, I don't see the point.
NB: I don't use Word bc I refuse to support Microsoft. I especially won't use Word now that MS has recently unveiled its happy regressive left sheeplethink dictionary correction module for Word.

I would be wary of using markdown and go with editors that use some sort of "stylesheet" formatting because it makes changes so much easier to do. If you suddenly want different first line indents, or line-spacing etc, the stylesheet means just change once.
I used to be a heavy MS word user, but these days I have OpenOffice, although pretty much any modern word processor uses stylesheets in some form.
The main thing with stylesheets is to keep it simple, and keep to the discipline. No "extra returns", no tabs, no one-off formatting except italics, underlines etc. Any hand-crafted layout like that just creates problems when you want to turn your manuscript into a book.

I would be wary of usin..."
Like you, I've been using styles to create my mss. With my first novel I used my old typewriter-learned carriage return - tab for paragraphs, then I learned how that affected making ePubs. I went through that novel and removed all the CR - tabs and have been using styles ever since. That along with the simple css file that I've developed gives me a very clean book.
Also like you, I've been using LibreOffice. I never used MS Word on a regular basis, so I don't know much about it.
I asked about Atom and git because I read an article by someone using it for version control and as an off-site backup. I was hoping to find out if anyone had experience with it and could tell me if it was worth the trouble of exploring it.

Version control can be a serious pain. I've used various control systems professionally for both software and document development. Software was fine, but unless there have been great improvements, document control other than plain text is a pain, and plain text for novels is also a pain.
I mostly keep on top of version control by appending an x.y.z style numbering scheme on the end of the file then taking and keeping regular snapshots, and regular (daily or even finer at times) backups of everything. I've had the occasional foul-up that has been fixed by working back through the backups.
I did try Scrivener - I liked the idea of being able to control/track each chapter separately, but I found I didn't get on with it. I've looked at using the OpenOffice/LibreOffice master document functionality to achieve the same thing, but it's very clumsy and I found that it gets confusing quickly unless the chapter headings are meaningful. I'm giving it another go for a short story anthology, but that's a much better fit in terms of structure and organisation.

I'm actually a bad one for using systems designed for the purpose. It seems as if I spend as much time learning how to use the system as I would inventing my own. More. My current "system" is a new directory for each draft. I write by hand, transcribe with LibreOffice and use Calibre to finish the ePub. Come to think of it, what am I doing thinking of trying something new.-)

For things like this, I miss my days on DEC machines and the VMS operating system which would automatically apply a version number every time a file was saved.
Upside - very hard to lose anything.
Downside - disk space usage could be horrendous unless you deleted old versions.

For things like this, I miss my days on DEC machines and the VMS operating system which would automatically apply a..."
From the days when data loss was a serious concern.

And our 1GB hard drive was a new-fangled beast, about the size of a couple of house-bricks, with everything backed up to tape drives daily.
A few years before I left that job, we had a line item in the budget for a new Sun SPARK workstation to run a high-performance modelling package. By the time we were ready to spend the money, it was "downgraded" to a new pentium PC which delivered nearly the same performance at a tenth of the price.

And our 1GB hard drive was a new-fangled beast, about the size of a couple of house-bricks, with everything backed up to tape driv..."
The mainframe guys got blindsided by the PC. Now all the supercomputers are running Linux.

Always the way. Very few people see the next big success until after it happens.
And forget all those positive things I said about OpenOffice and styles - I have just spent a day and half fixing "invisible" formatting junk in my book that looks fine in OpenOffice, but mangles the ebook. If I didn't have the tech background to take the html apart and figure out what was happening, my partner would have to shut me in a broom cupboard until the screaming and swearing stopped.

Always the way. Very few people see the next big success until after it happens.
And forg..."
Did you export it to ePub from OO? Can OO do that? I tried exporting to ePub from LO and found that it didn't create a CSS file, just putting all the formatting into the HTML. I haven't tried it since. I use Writer2Epub and then finish it in Calibre.
I do remember finding trash in the HTML once and I think I traced it back to Autocorrect, possibly word completion(?) I fooled around with the settings and it doesn't seem to be happening any more, although it might have been the chicken I sacrificed.-)

Always the way. Very few people see the next big success until after it happe..."
No, I exported from OO to hmtl and put that through Calibre. I learned the first time that the exported html is "proper" html with a stylesheet, whilst "save as" formats everything as it goes.
Then I have Calibre do all sorts of search and replace to clean things up, format chapter numbers nicely, put in a couple of explicit page breaks etc.
For some reason, a whole bundle of "border" formats appeared, so chunks of text in the final epub would be centred and about ten characters across. Took me a while to get that straight.
I ended up doing some more googling today and found various methodologies for wiping all formatting back to style defaults. The trouble is, that zaps any italic/bold/underline etc, and the real challenge is preserving those attributes. The tricks I found online worked really well, but the macro recorder has gaps in its functionality and fails to record some of the crucial elements, so the whole process has to be done manually.
As for chickens... I hit a double-barrelled chuckle there. I've just fed our chickens, and several of them need a serious talking to about not getting under foot. At the same time, the book I've been converting is an urban fantasy which has a recurring grumble from the narrator that it's chicken dinner again because his girlfriend has just done another round of "kitchen table augury."


Export. At least that way the html is already set up with a stylesheet.
I have a few search/replace actions that sort out global formatting and use relative units - change line-spacings etc, which only have to fix the stylesheet section. I change any paragraph indenting to use "emm" units instead of inch or cm, which seems to make the final epub work better.
When I did my second book, it was all incredibly smooth and straightforward. It's just this time round, so much formatting garbage crept in that it's taken literally days to fix.

Export. At least that way the html is already set up with a stylesheet.
I have..."
Okay. I'll have to give that a try on an experimental bit of text.

Export. At least that way the html is already set up with a styles..."
Good luck!
This stuff *ought* to be easy. :)

Export. At least that way the html is already set up w..."
Results:
Export HTML: Unless I'm missing something,* LO's way to export html is through the "Send" item under "File." There you can create an html document. The resulting file required lots of repairs in Calibre, mostly < spans >.
Export as ePub: LO has two ways to export as ePub, one called "Export as ePub" and the other "Export Directly as ePub." I couldn't see any difference between the products and they both produced ugly html and an empty CSS file.
Neither of these is better than my current method -- Writer2Epub extension with clean-up in Calibre.
* I was missing something. By selecting "Export" rather than "Export as ..." I was able to select html. Still needed a lot of < spans > repaired. Also needed to replace the CSS with my barebones one, but I always do that anyway.
So, no change, but fun anyway.-)
Well this is interesting. I have to write < spans > because if I don't, it doesn't appear.

Export. At least that way the html is alr..."
Yeah, that sounds like the sort of mess I was having to sort out. Amazingly, I have now got a crude macro that fixes lots of stuff in OO, basically automates wrapping italic text with < IT >....< IT >, revert everything to default format and then tracks the < IT > tag pairs and restores the italics. Even then, I see a few odd things coming out in the html.
So far as I can see, the OO internals effectively create hierarchies of nested formatting which messes with the final html. Put in italics on a chunk of text, and it creates a "leaf node" in the doc tree and formats italic. Flip some text inside that range to not italic and what seems to happen is it turns that into a sub-node of the italic and formats it as not-italic. Maybe I'm reading the html wrong, but that's my impression, so you can easily clutter your document with "hidden" formatting that isn't really doing anything except undoing an outer layer of formatting, but then gets transcribed into < span > tags on export to html.
Unless, of course, you have no geeky IT skills, in which case the formal diagnosis is "frustrating heap of ****".
(Wow. It really is annoying needing those spaces to get the < and > to appear.)