That is right, Victor. It all boils down to how complex your project is.
How do you manage to export a csv file to Obsidian and used in Obsidian? Which are the steps that you take to get this done?
1 Like
Thank you for the thesaurus link. I’ve played with Obsidian briefly, but have been flummoxed by the lack of concrete examples of what it can do. The Obsidian Publish examples help a lot.
I suspect, though, that the thesaurus specifically was machine-generated by importing an existing thesaurus. So maybe not the best demonstration for building a created world’s reference materials.
Just a quick note to say thanks to @amberv and @kewms.
I didn’t think it was appropriate (in relation to the OP and topic) when Ioa split this thread off from the original question, and then Katherine’s post about compiling out of Scrivener made me stop and think about whether I actually still needed Scrivener as a companion app in my workflow at all.
So I have spent the last three weeks working solely in Obsidian and now know for certain that it’s the only app I need for writing. These are the things that I learned that are relevant to my work:
- That Scrivener’s greatest strength is its ability to compile to different formats.
- That Scrivener’s greatest weakness is also its ability to compile to different formats, because that focus on having multiple layers of formatting in a “sometimes but not always, sort of but not quite” WYSIWYG environment really does get in the way of thinking and writing.
- That having an app that looks and functions identically across macOS and iOS is a major plus for productivity.
- That having an app that syncs between multiple devices without having to close the app on any of the other devices is another major plus for productivity.
- That having an app that syncs between multiple devices without using Dropbox is yet another major plus for productivity.
- That while Scrivener’s compile is a powerful beast, it’s also an unnecessary one. Scrivener encourages users to break works down into folders and files and parts and chapters and scenes. And a lot of the time users really do need to break their work up because as the word count builds, a single RTF file (with or without a lot of formatting) gets to be a very sluggish beast. But in apps using Markdown, it simply isn’t necessary to break projects down – although that is still an option if that’s how a user wants to work. I’ve pulled multi-file 300,000-word projects out of Scrivener (which run slowly throughout the interface and especially so when in scrivenings) and can have them in a single file in Obsidian without any performance lag. And in Obsidian, I can easily move around the document using the outliner; drag and drop sections from place to place with a trackpad or keyboard; run search and replaces in the blink of an eye that in Scrivener take a “time to make a cuppa” age; and on and on and on. So in the end, I don’t “need” to compile anything: I already have a single file to take forward.
- As is often mentioned on this forum, it is possible to create a finished product using compile (an ebook, a PDF for printing, etc) but in the main users really need to see Scrivener as a step in the production process: compile and then use other apps to do whatever it is the user needs to do with their work. So Scrivener is a “sometimes but not always, sort of but not quite” WYSIWYG production tool.
And that’s the crux of it for me—Scrivener falls between the self-referencing gap of its own logic:
- projects need to be broken up because they get slow if they get large
- they need to be broken up because there is so much focus on the different formats offered by compile and the different on-the-page formatting that that requires
- but even with all that focus and formatting, the compile process is a frustrating, time-consuming dark art for many users, and that – even after all that work – in many cases users have to do additional work anyway in Word, InDesign, NWP, Vellum, Pandoc, etc
- and all that breaking down, formatting, and compiling comes at the expense of thinking and writing, which are the only two things I want to do when using my Mac or iPhone to get work done
And thanks to Ioa and Katherine, I‘ve looked again at my workflows and found that with Obsidian I can do everything (and a lot more besides) that I can do in Scrivener, ending up with squeaky clean Markdown files that can easily carry over to other apps or be dispatched to publishers for publishing.
I think writing inputs and outputs have changed a lot since Scrivener was released in 2007, and I think I have been swept along with those changes. For the last fifteen happy years – thank you. But I’m no longer ‘there’ (nor, you’ll be glad to learn, ‘here’).
As much as I love Obsidian and Markdown (in part for some of the same reasons, e.g. the cross-platform experience)— I came to the opposite conclusion: I really don’t want to write long form fiction… or actually anything long form that way.
Of course that depends heavily on what features you actually need. For instance, as soon as any kind of meta-data comes into play, tags, YAML or whatever, the whole “I don’t have to split” euphoria crumbles to dust.
And then good luck with arranging those parts in “the outliner” (there is none at directory / file level, unless you resort to third-party addons).
3,000 words in one Obsidian file (let alone 300,000)? How does that work without a performance penalty? Unless you fold most of them away in sub-sections. I can watch Obsidian play catch-up with the rendering of markup when scrolling in way shorter documents.
It doesn’t even attempt to remember the cursor position (because there’s no need for it in “atomic” notes). Several third-party addons try do it, none of them works for me in a reliable way.
And when it comes to the output… Well, yes, since your input is your output, it will be as clean or sloppy as you write it. And from there you’re completely on your own. You can work the same way in Scrivener. Don’t use (RTF) formatting, write Markdown, compile to Plain Text and take care of the rest yourself.
Also, there’s no equvivalent to “Collections” in Obsidian. I’m sure it can be done with the heavy use of metadata (and splitting) and the DataView addon, but… No. I don’t want to.
Maybe you don’t need any of that and are perfectly happy now, which is great. For me, though — no such happy end. I managed to force either of the two to be the other, which works to some extend, but in the end it didn’t work out.
2 Likes
This is untrue, at least for me. I break projects into chunks because that aligns the tool with the way I prefer to work. I don’t think about large projects in terms of a single multi-thousand word scroll, and like Scrivener because it tells me that I don’t have to.
1 Like