Issues with compile and editing with non-scriv users

I love scrivener – as an input interface. Unfortunately, I work with other people in an academic environment, and I’m forced to send things back and forth with them, which means transferring into .rtf, .doc, or even (horror!) .docx. I suspect my boss may regret hiring me because the drafts I send him are always full of technical glitches because I refuse to use Word. So while I love using scrivener, when it comes time to compile I shudder. There are some parts of this process that are probably insurmountable (like Tracking Changes), but other parts seem to cause unnecessary problems.
My general frustration with Compile is that the settings are so numerous and complicated. I frequently have to go through several ‘test compiles’ in pdf format to see if what I want is what I’m getting. I can understand that the complexity of settings enables many different purposes for lots of users, but I think for me there’s a disconnect between a basically WYSIWYG input interface and then turning that into something OTHER PEOPLE can also see. In order to accomplish that, I usually export as PDF, but sometimes that is not possible.
As in what I’m working on for the aforementioned boss. My drafts have some images interspersed in different documents, and I attempted to export into an .rtf so he would be able to read, comment, and track changes in Word. The PDF versions of these compiles came in at about 6mb, but the .rtf or .doc, which include images in the right place, bulked up to ~110mb for no apparent reason. The images themselves are no more than 4mb. I worked around by sending images separately from a .docx compile, which removed the images automatically. But then at some point I will have to reinstate everything on the next edited version of our project… more redundant work for me, because I choose Scrivener to begin with.
This is just one example of what is a recurrent issue for me with Scrivener, that traces back to Compile - I need to be able to move things back and forth with other people. Some of this I will have to adapt myself to, given the necessity of Track Changes for editing, but this glitch in the process feels like something that could be rectified. I’m wondering if anyone else has similar frustrations with compile, or some solutions for collaborating with non-Scrivener users.

There’s no real way around this - simpler options would allow less flexibility, and then users would have to do lots more tidying up in Word. You don’t have to use all the options - you can just export everything exactly as it appears on screen (using the “Original” preset, for instance) and then tidy things up in Word, if you want. But otherwise, you have to remember that Scrivener allows for absolute flexibility in the binder, in how you structure and organise things. And because of that, it then has to allow the user to tell it just how to make sense of that when compiling - and the only way to do that is to provide the user with enough options to manage their own structure and get it looking as they want.

Well, one of the purposes of Compile is that it allows you to avoid WYSIWYG formatting. WYSIWYG is great - and necessary - if the primary purpose of the program is to format some text, to lay it out, but this isn’t Scrivener’s main purpose. Scrivener’s purpose is to let writers of long and complex documents get on with the writing, and we’ve never claimed otherwise. For that, you may wish to write using a different font, and completely different formatting, to that which is used in the compiled document. For instance, standard manuscript format usually requires a Times or Courier font, but there’s no need to use these fonts for writing; likewise, standard MS format requires double-line spacing, but that is very annoying during the actual writing phase. So Scrivener deliberately separates these things, and unapologetically so.

As I say, though, you can use WYSIWYG formatting if you so wish. Just use the “Original” compile preset (all you might need to change then is the “Separators” option).

Then presumably you have a lot of 4MB images in there. RTF only supports PNG or JPG images, so any other images will get converted into these formats. 110MB for an originally 4MB file doesn’t sound right, though - do you have any PDF images in there? If so, when exporting to RTF, these will have to be converted internally to JPG or PNG (because RTF doesn’t support PDF images); that might explain it. You can set the DPI for this conversion in the Import/Export pane of the Preferences - by default, its 600dpi, which means that the image will be enlarged by over eight times, resulting in a massive image (JPG and PNG are 72dpi, so to fake 600dpi such images need to be much bigger than the original PDF file - as I say, you can change this in the preferences).

I don’t see how; I don’t think this is a problem with Compile but rather with the particular workflow you are after. I’m not saying it’s an unusual workflow, but it’s one that Compile isn’t - and couldn’t be - built for. How could it? If you compile all your documents into one text and then send it out for editing and suchlike, there’s no way Scrivener could then bring that back in and break it apart intelligently into its original constituent parts with the original formatting etc etc - not until computers are as smart as users, at least. A better option would be to use Sync with External Folder, which allows you to share the component parts of your manuscript with other users; other users can then edit those parts and they can be brought back into Scrivener automatically, updating the original texts. But this would require those other users to work directly with the RTF format (which all versions of Word support) and not to be hung up on using .docx - obviously if it’s your boss, then you may not have much say in that!

All the best,

I understand your pain laru.

The problem with the Word document format is that it’s designed to support interoperability between different versions of Word – not between Word and other programs. I don’t think I’ve come across any piece of software that can parse all of my Office documents with 100% accuracy.

As I said in the post, I know the underlying reasons for a lot of these things… I guess the frustration is that I know why they can’t work but I end up butting my head against it anyways.

I’m getting the idea now of how you imagine the workflow - that the written text is in any format one wants, and the compile function cleans it up. True, I write in a font I prefer over Times New Roman, but in other software I can change the style definition when I’m ready to output, and the change rebounds over the entire file. In Scrivener, let’s say I’ve got text body in a level 1, level 2, and level 3 documents. If I want all that text to look the same, using Override Formatting in Compile, I have to do a lot of repetitive style formatting at different levels to get everything identical. Even if I change in the style presets before compiling, it doesn’t propagate (as far as I’m aware) through the rest of the documents. So, I’m stuck with repetitive action WITHIN Scriv. Plus, because I move between 2 Macs, I end up recreating a style preset or even a compile format because as far as I can tell, they can’t be exported or saved separately.
Then I compile (yay!) get comments, whatnot - but the titles of each section need to be reformatted the way the journal wants them. According to Word, or whatever (I’ve used Nisus, Mellel, and Pages for this too), all the text that came out of Scrivener is styled ‘Normal’, so I cannot find the titles other than by individually seeking them manually and restyling everything again. Repetitive action OUTSIDE Scriv. I would be surprised if others are not having similar frustrations. Hopefully someone will speak up here, and maybe offer other workarounds.

The Sync function - which I had never heard of before you mentioned it - will probably go a long way to solve some of my issues, because I can stay in Scrivener longer. I held on to Scrivener Gold for a long time, not updating. because the file names in the package contents were recognizable, so I could edit individual things with a text editor in a pinch. I’m just sad I didn’t know that existed a few months ago.

P.S. - I meant the total of the images is no more than 4mb, not each of them is no more than 4mb. There are some pdf images tho, so I will try what you suggest.

Well, not necessarily. If it’s just a font change you want, you can use Quick Font Override instead of bothering with “Formatting”. That changes the font for everything. If you want each level to be the same, then just set up one level (or type), then select the row in the “Formatting” pane, hit cmd-C, and select the other rows one-by-one and hit cmd-V. This will copy all the formatting and settings across.

Style presets are saved as a .plist file in the ~/Library/Application Support/Scrivener folder. Just copy the file across to your other Mac. And Compile format presets have always been exportable and importable. Just choose “Manage Compile Format Presets…” from the bottom of the “Format As” menu in the Compile pane. That opens a panel that allows you to create, update and delete formats, to choose which ones appear in the menu, and to import and export them.

True, Scrivener doesn’t support RTF styles - a limitation of the OS X RTF engine. Hopefully in version 3.0 or 4.0 we’ll find away around this.

All the best,

If you format your titles so that they use an unusual size (e.g. 17pt) and/or font face that isn’t used anywhere else, when you have opened the file in Nisus, you can select all instances of that formatting with one click and apply a different formatting to them in one go. It used to be possible to do the same in Word, but I don’t have a recent version of Word to check that it is still possible to do this. Alternatively, you should be able to Find and Exchange formatting in both Word and Nisus.

Cheers, Martin.