I'm Building a GitHub for Writers | Hacker News Reader

I'm Building a GitHub for Writers

madebyloren.com

311 pointsguynamedloren143 comments

A lot of programmers already find it awfully hard to wrap their heads around how git works. I can only imagine how hard it will be for non-programmers to (it doesn't suprise me that the successful example used is written by mathematicians).

HOWEVER, that doesn't mean it can't be done. In fact, if there are ways to visually simplify git and make it more intuitive for non-programmers, those techniques could wind up making git even better for programmers too.

I could also imagine a convergence of the git model and the wiki model someday -- where anyone can edit (like Wiki), but where there are branches, merging, etc. Obviously, a lot of internal wiki's don't need such complicated version control, but for things like Wikipedia, it could be amazing.

I think there are a lot of areas for working on writing collaboration -- group projects in school, business proposals, technical manuals, all sorts of things.

And the main attraction for users over, say, Google Docs, is that your changes don't overwrite others'. The fact that your edits create a "branch", that then others can accept/reject/modify/merge, is a vast improvement in creative collaboration.

Learn Git Branching did a really good job of making git easier to understand:

http://pcottle.github.io/learnGitBranching/

If you take that approach and update with precise but plain english vernacular (redundant?) and domain specific examples, then you are halfway there to making it understandable by regular people.

The most important thing to help people really get it is to promote small commits to their work. The real value of git only makes sense when you start making small commits. Having worked as an editor in a past life, having something like cherry-pick and interactive rebase would have been most dope. In a way, maybe editors are the best audience for driving adoptions among writers in general. For professional editors, version control is a painkiller. For writers, it is a painkiller if the writer is an obsessive editor. For those who write and commit once, but don't go back and review commits and refactor, it doesn't provide much value.

Think about how Apple Time Machine made a stab at operating system backups and snapshotting. Was it perfect? No. But it was a step in the right direction.

Loren and others may not solve all the problems right out the gate, but they will lay the groundwork for new approaches to teaching version control to writers and other audiences.

Learn Git Branching is best learning tool I ever tried! With this tool and a little practice on the side I was able to learn the basics of git within a couple of hours so that I can confidently branch, merge, push, pull and rebase and I know what I'm doing! Programmers should invert more learning tool like this!

Love this site. Hoping people contribute a few more lessons to it in the near future.

Thanks Alex! I'm working on some lessons around push / pull / rebase right now that I think will be super helpful for learning the next steps.

Progress has slowed a bit since work has ramped up, but now that intern season is over I'm expecting to make some progress

learnGitBranching is a fantastic resource that I didn't know about. Thanks!

Whoa! Now that's how you teach anything command-line to non-programmers. Someone should do something like this for unix commands and the like.

> In fact, if there are ways to visually simplify git and make it more intuitive for non-programmers, those techniques could wind up making git even better for programmers too.

That's the goal. I want git to handle everything behind the scenes, and everything that the user touches just 'makes sense' - ie its not all that different from what they're already doing. A 'save' button makes a git commit, editing somebody else's content automatically creates a new branch, 'submitting' that content on the new branch creates a pull request .. etc.

You could combine the advantages of git versioning with wiki style collaboration and make a git-wiki like the parent said, but for writers.

Wikipedia is largely prose, but of a nonfiction variety. This would be a good intersection for git, which was born out of code.

You could call it..."geeky"...too bad it wouldn't be for geeks.

This is a really fantastic idea. I can also imagine writers will get a great deal of use out of git cherry pick.

The intersection of git and wikis is interesting, it's one of those things that sounds compelling but then the details usually end up making it hugely complicated. Worth thinking about.

For general writing collaboration I would seriously consider not allowing arbitrary branches. Most people just get confused managing multiple independent outstanding branches that potentially branched from different points in the history.

One restriction could be that each user gets a single branch (their suggested change). As changes get merged into HEAD each user's branch must be rebased (with a click) before they can add more changes to it. From a programming point of view this is restrictive; why can't I suggest two independent features? But keeping track of dependencies is precisely what's hard for people.

> For general writing collaboration I would seriously consider not allowing arbitrary branches. Most people just get confused managing multiple independent outstanding branches that potentially branched from different points in the history.

Yep, absolutely true.

> One restriction could be that each user gets a single branch (their suggested change). As changes get merged into HEAD each user's branch must be rebased (with a click) before they can add more changes to it.

Heavily considering this approach.

Honestly, I think writers could benefit more from the collaborative editing model than the git branching model.

When do writers need to branch their stories/articles into two forks? If they do that, why not just copy the document and work on it there? The large bodies of text don't have the filesystem dependence that programs do; it doesn't matter what the manuscript is called if nothing #include s it or require()s it.

Instead, it can be quite valuable to have a collaborative text editor: your colleagues' changes are shown in real time as they make them. If you go on a plane ride, they're synced when you get more internet access, with conflicts clearly shown and deliminated so you can resolve them simply.

My wife is a novelist.

The collaborative editing would be useful in the later phases of a book -- i.e., working with an editor and copyeditor, it's be great to have a really efficient way to review changes made to the actual text (instead of reviewing a hard-copy with hand-written corrections, and then hoping those corrections are accurately hand-merged into the text...).

But for most of the life of a manuscript, it's just the author working completely solo (from what I've seen). In the standard flow, there's no editor until the book (in a fairly complete and polished draft) is sold to a publisher. A literary agent may read drafts along the way, but normally wouldn't touch the text, just write up higher-level feedback.

On the other hand, the task of designing a narrative that flows well at the scope of a novel is really hard, and can take a lot of large-scale trial and error to improve. She usually does have several "branches" of her current project, as well as large chunks that she has cut from the current draft but doesn't discard (and sometimes may be re-added).

Her first published novel shed more than 300 pages that were never re-added on its way to the final version... it's hard to keep track of all that. It wasn't like "drop this chapter", generally, more like "remove this subplot from the hundred or so places it shows up".

I'm not sure how well the git model would address the problems -- the big problem is scale, and visualizing large structures made up of lots and lots of tiny little words. They are plenty of pain points worth attacking, though, except that most writers don't have much cash to spend on tools.

For something like that you would need a way to mark parts of text, that deal with certain subplots/plots. Like different colors for different plots, but what happens, when they intersect? Or you could use flags, or something like that.

An easy way to trace a plot throughout the novel. Interestingly, a tool, that enables this could even be used by literature students, when analyzing texts for an interpretation (been there, studied that). The only problem with this is copyright and the inability to get your hands on a digital os-/reader-independent format, to work with in the first place.

I worked mostly with texts from medieval times and it was a pain to get hold of (good) digital versions. May be, that this is because language processing approaches are not that widely used in my former field of study.

So to wrap it up - a great tool, enabling authors with this could enable also a better understanding of novels for literature students alike.

"And the main attraction for users over, say, Google Docs, is that your changes don't overwrite others'. The fact that your edits create a "branch", that then others can accept/reject/modify/merge, is a vast improvement in creative collaboration."

This.

Version control is a capital-P Pain In The Ass for group writing projects. Anything that improves the version control problem in collaborative writing solves a very real need. Perhaps not a huge TAM at first glance, but a real need nonetheless.

Haha! Agreed, but like lots of software things--I think that's mostly because the instructions are written by the programmers. I'm management at my company (don't code), but have never had trouble grasping the conceptual issues and intricacies of code, and that makes me really good at teaching it. You just have to contextualize it properly for their perspective. Writers, or anyone that's not a programmer, doesn't care what a git is or how it work technically. They want to know how it will fulfill their wants/needs, and they want big buttons labeled with those solutions. I could teach my grandma to use to the github client if there was a shortcut to it on her desktop.

I love git, but some of the metaphors and inconsistencies it makes are definitely not good and make it much harder to pick up than it should be. I think with the right set of metaphors and ui it can work.

I was looking for something like this. I'm planning to write an epic sci-fi novel and make it available under some open license (Creative Commons BY-SA probably). Since I'm not a native English speaker, I'm hoping those who are would jump in and help me perfect the text.

My final hope is to see a Hollywood movie made out of it one day, because many recent SF movies lack good story.

My other idea is to make the story branch into different directions. I would write the main storyline, but another writer could come and fork it at some point. The reader would be left with a choice at some point: would you like this character to take that decision or the other - and after choosing, the reader could read the forked variant.

This is the first time I'm publicly writing about this idea. I'm still undecided what technology to use to create it.

I'm cheering you on, babuskov!

I've used git with my sci-fi novel, Planet Oz, since I first drafted it back in 2008. Quite glad I did since it makes the editing process much easier to track.

Can't wait to begin the Kickstarter in October.

> I'm still undecided what technology to use to create it.

I've got friends working on projects a bit like this. Markdown served from github pages is nice, as you edit and compare diffs in markdown, but you can get it automatically rendered into a nice readable form too.

My own view though is that the most likely way to fail is to lose motivation, and having interminable discussions over minutia early on while writing is likely to suck away your energy. I'd probably get at least the skeleton of the whole story into your repos before encouraging others to contribute.

Yep, the skeleton for the whole story is ready. I have been pondering about all those main story parts in the past 2-3 years. My next step would be breaking each chapter into smaller parts, without writing down any specific sentence. The last part of the process would be actually writing it.

Have you looked at Z-code? There's lots of reasonably modern interpreters for it and there seems to be a fairly active community.

No. I just googled now, but there seems to be a lot of stuff named Z-code and I'm not sure what exacly am I looking for.

Can you give me some URL?

I think Max is talking about the Z-Machine interpreter that Infocom created to make games like Zork and Hitchhiker's Guide to the Galaxy. There are many of modern ways to use it, like Inform (I recommend Inform 6 over 7), and even a web interpreter, Parchment.

That's if you want someone to PLAY the choice in the story. Which is slightly different from what you're talking about.

Yeah, the Z-Machine interpreter. But I think it is reasonable to look at choose your own adventure as a fairly straightforward game, so Z-code presumably provides the necessary features for it.

Git branching a storyline is an awesome idea.

I'd love to know when this book gets off the ground - I enjoy both sci-fi and correcting other people's English.

Thanks, I'll take you up on that. I assume that the e-mail in you HN profile is valid?

Yup - look forward to your email.

Cool, I know whom to call for reviews :D

This is sort of what Nate Kontny is doing with Draft, which I like kind of a lot, much more than I expected to.

Draft's editing/version control is rather awesome and easy to use. It would be neat if Loren and Nate could work together to get some of the github-like "social" features into Draft--following updates and broadcasting progress.

My feeling would be that having one platform with both features would be better than having two very similar things here.

I've been loving draft to the point that I haven't really engaged with any of the other new platforms out there, like editorially. I sign up to beta just about any of them that come along, but Draft just feels like home already. That said, I can't imagine I wouldn't at least play around on the platform the author describes.

My biggest issue with Draft is consistency in behavior. Personal anecdotes are frowned upon here, but for most of last week the ability to export files wasn't working at all for me. When I'm dealing with Markdown, this is the last thing I want to happen.

I much prefer a system that's geared to projects and not specific types of document too. For example, if this supported housing a collection of documents and an image or two - with changes able to be made on each file, it would be amazing. Google Docs I edit with others quickly tends to devolve into three different types of sharing/permissions issues, and I'm left frustrated or wondering why some people can access one thing or another.

There's definitely room to innovate here.

> I much prefer a system that's geared to projects and not specific types of document too. For example, if this supported housing a collection of documents and an image or two - with changes able to be made on each file, it would be amazing.

Nailed it. That's what I'm building :)

Agreed. Draft tackles versioning and editing in markdown very well. The incremental save feature makes all of this transparent to the writer.

This is interesting, but I think it kind of fails to take its audience into consideration.

1) It assumes writers write like programmers code. This isn't true. At least with scripts, what typically happens is the writer will write an entire script, send it to their boss for revisions, and then rewrite it. They would consider it a pain in the ass if they had to make "commits" for every page they wanted to add, and especially if their bosses had to review each change (mostly because bosses are executives who just want to leave their mark by making unnecessary changes on things).

2) In either book or script writing (or whatever) small grammatical changes are unimportant, and there will be way more of these than we would care to look at. Who cares about reverting a change of "its" to "it's"?

I see this kind of thing as being useful for writing an informal, informational book in the form of an extended blog post (see http://gettingreal.37signals.com/) but otherwise, I can't really see it being that useful for professional writers.

Also, something that I think would be REALLY useful (that I'm honestly surprised I haven't seen is a Final Draft - meets - Google Docs (screenplays editable by multiple people via online, all stored online). Dunno if anyone's seen anything like this.

Sources: My dad is a TV writer.

Scientific writers do write like programmers code. There are as many as ~15 authors on a paper. The scientific paper has intensive markup, often in Latex. And collaborative writing like GoogleDocs doesn't work b/c 3 different authors will make changes to the same sentence and you need to be able to see all three changes next to each other. Not to mention figures, equations, etc...

MS Word and Libre office aren't powerful enough for scientific writing and aren't universally embraced or consistent across operating systems.

Github for writers is desperately needed. Draft doesn't do nearly enough

Source: physical science graduate student

Hey, I'm working on a similar project for scientific writing. You can find the link in my profile along with my email.

We have had success with large collaborations working on it, 15 authors should not be a problem.

Check it out and feel free to contact me.

Exactly. I don't care about the document progression or the changes of words. I just want "versions" and "notes". I actually did a writing application for myself only and I had two windows, one on the left with the actual text, one on the right with the notes, also a sidebar with the chapters and progression of the book.

i try to get my SO to write academic essays in applications that have history or diff viewing... but never had any success. but that only means the current offerings are bad. even for programmers.

regarding your points, when i set up `meld` to show the text she sent vs what her professor or coleague sent back, she is amazed. So there is usefulness.

Second, have you ever used Kwrite in KDE? you don't have to commit or anything. just type. and it shows the diffs marks on the left. pretty hand. and avoid commits and all.

about point #2, just make branches and tags more practical. Also, try to merge back those 'word fixes' in other branches ' heads. would be awesome even for programmers.

source: optimist :)

Is it just me, or are there a number of fairly high-profile writing tools that have launched recently? The ones I'm thinking about:

  * Draft http://draftin.com
  * Editorially http://editorially.com
  * Quip http://quip.com
  * Loren's Penflip http://www.penflip.com
  * Google Docs (obviously not new)

A lot of thoughts running through my head, including that finally products are being built from the ground up both for collaboration and for mobile/tablets. (And that this appears to be an early sign of the end of Microsoft's dominance in the office productivity software world.)

I'm on your 'high-profile' list!

Made my day, thank you.

Are any of these projects aiming to interface well with MS Word or OpenOffice?

Well you could use pandoc [1] for creating MS Word or other file-formats. As far as I know MS Word is still the choice to use for novelists, when it comes to giving test to editors and a like.

So pandoc could be used for transforming a text into many different formats (even ebook formats).

Might be an interessting thing to look at for the creator of "git for writers" ;-)

[1]: http://johnmacfarlane.net/pandoc/

I'm currently using GitHub to collaborate on writing a book with a non-technical coauthor. Generally it works great, but there are two key issues:

* Changes are tracked by line which is equivalent to a paragraph in a book. If I go in and add a comma to a paragraph and my coauthor simultaneously changes a word in that paragraph that can create issues.

* Errors are very difficult to solve for my coauthor. When Github Window's app encounters an error, it basically says "Just open up the command line and you should be able to figure out how to solve this". Of course this isn't feasible for a non-technical audience.

If your product can fix these two issues (which it looks like it is trying to) it could be very valuable.

Disclaimer. I'm the creator of GitSense which is working on building advanced power tools for Git and GitHub.

The diffing problem is solved quite easily with the Google Diff, Match, Patch algorithm and our Smart View technology. When GitSense was created, it was obvious that we all work differently and we all have different needs which was why we made it insanely customizable. If you are not familiar with the Google Diff, Match, Patch algorithm, you can learn more about it here:

https://code.google.com/p/google-diff-match-patch/

I've also put some screen shots of what the diff results for two markdown files would look like at:

http://screenshots.gitsense.com/google-diff-patch-match.html

The context aware source code that is used in GitSense to display the diff can be found here:

http://screenshots.gitsense.com/google-diff-match-patch-inte...

> Changes are tracked by line which is equivalent to a paragraph in a book.

When using Markdown, I usually split paragraphs over multiple lines (usually between sentences) which makes diffs much easier to read, while still rendering as a single paragraph.

I know people who break lines per phrase/clause, rather than per sentence. You wind up with lots of very short lines, but that doesn't matter after Markdown is done processing. The benefit is some _very_ granular diffs that are easy to read, as you work out rewording things in the editing process.

Same here. "Single line paragraphs" are a pet hate of mine - they have no place in plain text files even if those files are intended to be post-processed.

Absolutely, but you still have the issue that a minor wording change can cause an entire paragraph to be marked as changed if your editor re-wraps the paragraph. Or, if the editor doesn't rewrap the paragraph, then they slowly get more and more ragged over time, making it annoying to read through.

A diff tool that understands paragraphs is an essential tool for this project.

I disagree with this. In a plain text file that consists of prose, line breaks are presentation, not content. Hardcoding them into the file doesn't appeal to me. It makes editing a pain in the neck. It also means the line breaks have to be removed again if you want to move between formats.

That sounds like an editor issue, though. You should be able to reflow a paragraph, indent it or whatever it as you like.

A compromise might be to break lines on sentence boundaries: edits are often constrained within sentence or paragraph boundaries, and then diff would give more meaningful output for most sentences.

^-- This one sentence is two lines long, though, so perhaps my idea is not that good. Diffs would still be easier to read than if each paragraph were one line.

I'm still undecided on this. Say you have a long paragraph and have to add one word at the beginning. You have two options - either rewrap and introduce a silly diff, or don't rewrap and exceed your chosen text width.

Single line paragraphs don't put such a difficult choice in front of you. If your diff software is any useful, it'll highlight your change within the line.

Perhaps you could have a special kind of commit, a "text-rewrap" commit (similar to merge commits), which might let your diff presentation tool alternate between diffing lines, pararaphs, or the like. I agree, the presentation layer really needs to be as granular as your editing team needs to be. I __really__ like how Github shows the differences within a line that is different.

That's why I don't use fixed-width. I break lines up logically, usually at the end of a phrase/sentence, regardless of their length.

i write quite a bit with git, and recently brought an old collab project with a friend onto it.

i do all of the git work, and i just have my friend add my edits to his document, then he sends it back and i diff the two for potential errors... he requested it be this way because he is stuck in his ways, but we've agreed to go full git on book two in the series...

that said, when i first started sending him diff files we both saw this as a problem... my solution? i just wrote a script that i run on the diff to just highlight the changes made, then send him the new edited diff.

    -PARAGRAPH
    +PARAGRAPH

becomes :

    -SENTENCE
      -word...word
      +word...word
    +SENTENCE

... i place git's diff at the bottom of the file in case he wants to look that over as well, but this cleans up the edits real nice

What I mostly want, when writing collaborative documents, isn't a git equivalent - it's a code-review equivalent.

I want to leave comments on individual words, sentences, paragraphs, etc., with suggestions for changes. And that's something that git doesn't allow, because the smallest change you can comment on is a file, with no ability to comment on specific parts of it.

Git doesn't support it, but GitHub does. You can comment on individual lines via GitHub. I'm going to do something similar - maybe on a more granular level (words/sentences).

It sounds like you're looking for Microsoft Word, which has offered this feature for more than a decade (albeit though not online until this past year).

Latex has line comments and since it is compiled into a document you can place line breaks anywhere. It does break the flow of the sentence though.

I've used Latex+vim+git to write scientific papers recently. It is unfortunately much too complicated for my colleagues to adopt. They're holding onto MS Word which is far inferior in terms of markup, citing, and VCS.

I've been converting my Latex documents to ugly word documents for collaborating and editing. The MS Word editing isn't great either. The strikethrough method breaks up flow. I've also had problems deleting side comments in Libre office. And the MS Word experience isn't always the same on Macs.

You can usually do it with shared Google docs. I know a few writers who share their content with a few trusted betas. It provides the necessary features, I think. Highlights, commenting etc.

I think there is also a "real-time" feature that it provides (not sure about that), so there is usually more than one eye on the doc as the chapter is being typed.

I asked how this could be implemented online 2 days ago: https://news.ycombinator.com/item?id=6284610

I believe Operational Transforms are the way to go, but I have no basis for that.

If you're using google docs for your collaboration this sort of commenting is built in. It's a great feature. Almost better than collaborative editing

I think the guys at Github do want to help non-technical people learn to collaborate with git in areas like education, government, and literature.

Check out this interview where one of the Github trainers, Brent Beer, talks about these ideas. You might want to get in contact with him.

http://blog.begriffs.com/2013/08/an-interview-with-brent-bee...

> I think the guys at Github do want to help non-technical people learn to collaborate with git in areas like education, government, and literature

Yep, they're doing it, just not quickly enough :)

Neat, I'll check out the interview. Thanks!

Take a look at "Why use version control systems for writing a paper" (http://academia.stackexchange.com/questions/5277/).

When it comes to diffs for text, just add "--color-words":

git diff HEAD~1 --color-words my_file.md

But I agree with other stuff, that for non-techies Git + GitHub may be to hard to start.

I know that Michael Hartl of RailsTutorial is working on something very much like this, but will be a full end-to-end product including things like sales.

I can't wait to see what he comes up with.

How are you solving the diff problem? Word based diffs tend to be jumpy and hard to follow. On the other hand line based diffs do not work well for text (prose).

Yeah, prose.io uses word diffs and it's definitely hard to follow.

I'm doing 'fragment diffs' (doesn't really have a name) - but basically, all the text is displayed as you would read it (sentences and paragraphs), with diffs displayed inline and colored. Added content in green, removed content in red - doesn't matter if it's characters/words/sentences/paragraphs - it all works.

Screenshot is not the best, but something like this:

http://www.flickr.com/photos/madebyloren/9620674181/

Don't know if you've already thought of / planned this, but please please implement a side-by-side diff view.

Code is easy enough to read in a unified diff because lines are usually short (unless java). Prose flows a lot more fluidly, and there is no formal grammar for human language. I have had the displeasure of having to read "redlines" (prose with inline diffs marked in red strikeout) and I find them completely impenetrable.

I'm using latex+vim+git for this purpose currently. I put two spaces after a period to signal the end of a sentence in Latex. Which helps me break the document up by sentences with vim.

If you're using markup then encouraging the period-double-space could be a harmless way of easily breaking the document up by sentences

"Sentence-based" diffs would work well for prose, but the problem is recognizing sentences, given the other uses for full stops. If you use hard line breaks at, and only at, sentence ends, you could use line-based diffs to do this (but this may run into editing convenience issues.)

We've been working on a similar problem. Writing in markdown, pre-processing the saved markdown to truncate to 80 column (best fit) paragraphs and then diffing on the markdown.

I think for writing, sentence-based diffs would be the most useful.

From @kmatzen's comments on Authorea.com: " What do you get when you combine authors with gonorrhea? Authorea!

I like it! There are two things I would still like though. One would be page rendering in the browser. We often have to iterate several times on a paper adjusting figures, text, padding, etc. to get everything to fit within the page limit. I could write the paper in this and then export it to tex, but that defeats the purpose. The second thing is support of popular LaTeX packages. It looks like I can upload arbitrary tex documents, but that's going to get messy if I have to upload the same packages for every paper I'm writing.

We use github for our papers right now. Less than ideal. "

What can I say? Naming is hard. We initially thought to ourselves, "Spreading science should be fun and intuitive, like an STD. And it should be hard to get rid of." Thus Authorea was born.

Have you tried our pdf exporter to check page length? We are trying to build it out to where you, as a user, don't have to worry about packages. If there is a journal you would like to submit to and we don't currently have the journal setup as an export template, please contact us and we will be happy to add it.

It's funny how authors worry so much about page length and figure placement, and then the first thing a journal does is remove all of the formatting the author has put in place.

hi @kmatzen/@mallyvai! Adding briefly to Nate's comment (my co-founder at Authorea), I would like to point out that we do support many popular LaTeX packages. One big reason why we do not support custom LaTeX packages is that some of them are really intended for the printed page and cannot be easily ported to the web- and the point of Authorea is to create web-native documents, which render in the browser, first, and can then be ported to print formats. In other words, we support a safe subset of LaTeX which we can easily convert back and forth into other web formats (such as Markdown). Do let us know if you have a LaTeX package in mind and we will see if we can add it for you.

What about draftin.com? I use that already and I like it a lot.

Big fan of Draft (and Nate). It's got the version control part down, and I think it works well for gathering feedback on blog posts and other short form content, but it isn't designed for longer form content as far as I can tell. It's also missing the community aspect that GitHub has, which is huge, as well as features that would make writing an ebook or textbook feasible. Definitely a great tool though.

Any reasons not to work with Nate? It seems like the social stuff could be built in tandem with his strong editor.

https://draftin.com/ - Agreed, great.

http://www.roughdraft.io/ - Also, I like it.

I don't think it's a bad idea to proceed with it, I would just find a way to differentiate it from what's out there.

ditto. simple and easy to use.

I'm a little late in the party but I wanted to express my feelings anyway. I wrote a book with 2 other authors 2 years ago, and would have LOVED a tool like this. Going over Word documents in revision mode was a colorful nightmare. In general, having to work with Word was not the ideal stuff anyway.

To my point: to be totally useful along the whole typical writing process with traditional publishers (authors -> technical review -> authors -> editor review -> authors -> punctuation/misc review -> authors -> editor) you'll need at some point to export as Word documents. Ideally, collaborate in Git like environment, export as Word, receive a new Word with all those colored revision and import into Git like stuff.

I think this is a really interesting idea; however, I don't think it's a very good method to build it as 'Github for Writers' since coding and writing are two very different processes. For instance, writers often create their work on their own instead of with multiple contributions. Contributors is more of a code project attribute, for anyone who knows how to program well can add/modify code while not anyone who knows write well can add/modify to a story.

It is true that writers need editors but editors are certainly not writers. If there was a system where all of the contributors acted more like 'editors' rather than writers than I personally think it would be an awesome version control program.

I think this may be true for fiction (. . .writen often create their work on their own. . .not anyone. . .can add/modify a story), but I think it's a great idea for non-fiction, such as the textbook and research paper examples that the OP cited. Other examples include user or technical manuals, requirements documentation. . .The team members are often more akin to contributors (code or otherwise) than to editors. Individual but interlinked units have to come together to form a whole.

Absolutely. My wife's job as a professional editor involved editing hundred-page documents, often with more than one author. A tool which made version control simple, yet writing easy, would be a godsend.

The down side is, many such organizations (e.g., military) have standardized on Word, which is terrible for collaborative review, editing, and version control of long documents. (Of any documents?) She would have been extremely happy had her employer allowed them to use LaTeX or similar.

> Absolutely. My wife's job as a professional editor involved editing hundred-page documents, often with more than one author. A tool which made version control simple, yet writing easy, would be a godsend.

I hope you signed up - I want to that editing process easier!

> I want to that editing process easier!

I believe that sentence no verb.

Ha, woops.

One question is, do writers create their own work because of the nature of writing, or because they lacked the tools for collaboration. I have seen several examples of collaborative writing. The first example that comes to my mind is the SCP Foundation [1].

[1] http://www.scp-wiki.net/

EDIT: Actually posting the link might help. (Although redacting it has a nice sense of irony too).

As someone who writes software and fiction (heck, even software-fiction in the form of visual novels), I actually keep finding more and more parallels between the craft.

I like to explain the fiction creation process as "writing code that targets the Human BIOS."

I'm familiar with similar workflows. Last week, I edited via a (private) Github repo with a scientist client who use Emacs/ org-mode/ ediff, authoring .org files, editing using Git staging to create change sets and inline \NB[]{} commands to annotate changes, and ediff/git-merge to step through and resolve changes.

Some observations:

1.Ediff is a nice tool to use for resolving Git conflicts;

2. Org mode works very well with this workflow, provided you use one paragraph per Unix line;

3. Latex generation can be very nice from org-mode sources, and the org-mode -> Latex exporter allows a nicer separation of content from presentation than you get with vanilla Latex.

I'd love to hear of anyone who has success with a similar, but vi-based workflow.

I use a similar setup with vim+latex+git for chemistry articles. It works amazing for me. I feel like it is the most powerful presentation setup and most efficient editing setup currently available.

I break sentences by ending with two spaces. (this doesn't matter since Latex is markup) The only issue arises when ideas span across multiple sentences, but I believe there is a flag for controlling the number of lines around a diff. (sorry, I'm new to git) But this sort of proves the point that an easy wrapper for git should be created for scientific writing.

The two main problems I have are: 1. adoption by collaborators (they can't surmount the learning curve) 2. diffing sentences is very different from lines of code

Typical Org-mode code diffs better than typical Latex code. This is a serious argument in favour of authoring in Org-mode and using its Latex export facility.

The key to this one will be figuring out what features of git (and in particular github) will make it successful in other verticals.

I wonder if some key CS concepts are so fundamentally ingrained, but that writers think about their work flow differently.

Great idea. Yes please!

But Markdown as UI... is less than ideal. It's great for hackers, but there's already a GitHub for hackers, and it's called GitHub. No matter how much prose.io prettifies it, you still get those damn asterisks all over the place.

Personally, I write in TextEdit, Helvetica 12pt. That's not a universal answer, but it works for me. (Previously I used MacWrite Pro, ClarisWorks and Word 5.1 at different times, but the basic appearance was the same.) Anything that makes a bold word look like "* * this * *" breaks 20 years of habit, and I won't do it.

Markdown as backend storage, fine. But not as the primary editing interface.

At work we are trialling Lyx + Subversion for collaborative creation of developer documentation. The stored data is in a text + markup format which allows for easy diffing. Lyx being a layer over Latex means we can easily transform into a wide variety of formats, but mostly we output to PDF and HTML. Subversion was chosen as we already had a server available and there isn't really the same need to branch/merge documents compared to code. Writing in Lyx took a little time to get used but I definitely feel more in control that when trying to wrangle complex docs in Word.

I used to write my documents in Lyx too but spreading the tool amongst my colleagues involved a lot of evangelism. Markdown is a backend storage easier to adopt as developers are used to it (stackoverflow, github) but it produces output for the web : no pagination, header/footer, etc. Throw pandoc+tex template into the mix, and it starts to get really interesting : a quickly hacked markdown file transforms into nice pdf will all needed attributes. But why stop there? Now that you have a build process, move it to a server and add some hooks to injects traceability infos (author, last edit date, commit hash...) into the generated document, etc...

I went down this road, and it works surprisingly well. If you're trialling different solution for writing technical developer documentation, see a more precise description of the workflow that I pushed on my blog yesterday: http://kray.me/pro/doclegit-git-documentation-server/ (sorry for the copy quality, english not my first langage, corrections by mail are welcomed)

Interesting. When we set this up we looked at maybe adding an automated build (transform) step on commit via our CI (Jenkins) server.

While the software developers are comfortable learning and using Markdown or other markup syntaxes, I eventually would like to spread this approach to other technical people in the company. In previous attempts at teaching basic scripting (Awk, Ruby, Python) to engineers who are comfortable writing complex logic operations in Excel and who can dabble in VBA - I found a lot of resistance to the idea of learning a new syntax. So I expect the evangelism required to get these colleagues to write in Markdown will be greater than showing them Lyx, which at least looks like one of the editing modes in Word.

I thought the main technical hurdle (for not dev crowd) in my current workflow was git as there exist a lot of good Markdown editors on all platforms ;)

> Markdown as backend storage, fine. But not as the primary editing interface.

That's interesting - hadn't thought about storing in markdown and just displaying as html. Hmmm.

Is this going to be focused on those who use latex and markdown, or the the broader audience of 'people who write'?

If the latter, you run into a problem pretty quickly - MS Word. It's still one of the most popular document editing tools out there. More recent versions do store content in zipped XML, but are you willing to put the work into parsing that content at a level sufficient to let it integrate into the kind of flow you discuss?

(EDIT: That said: if your target is those creating text-based documents, then I think this is an excellent idea. )

How many of these options penflip, draft, etc are actually open source?

I'm looking for very specific features on the level of Latex libraries that would never make it into a for profit application.

This is a fantastic idea, beautiful and beautifully simple. Of course, one can think of a million objections to its success, but I really really hope it succeeds!

An idea is to think about rendering/typesetting as a feature. For many, getting an environment set up to turn the project files into final files of whatever form is a huge barrier. It would be great to have a bunch of back-end rendering be just a click: Markdown->HTML, Markdown->PDF, Markdown->epub, Latex->PDF, Latex->HTML.

especially when you throw in citations for scientific articles. Which often require compilation of the citation database, citation order, and citation insertion. Adding 3 more steps to the process.

I've recently switched to papers to help with this. It seems to have a strong community and supports both MS word, libre office, and vim. Jury's still out though.

http://www.papersapp.com/papers/

Already done

Have you asked a representative sample of writers whether they write this way, and would like to use your site to write this way? Outside of business, collaborative writing seems to be pretty minimal (I say this as a novelist)...

Also, if there could be some cross-compatibility with MS Word's Track Changes (I know, I can dream...) that would be great.

As a writer I'd be glad to use something like this. I'm working on a book with the founder of a non-profit in DC, and it involves a lot of sharing articles, trading thoughts, and sending documents. We've been using Honey.is which is nice for small businesses but just not as collaborative as we need it to be.

Ironically, I posted a call to action to build a community for writing together. Very curious to see how Loren's project does and how it compares to Draft and others.

https://medium.com/writers-on-writing/920214571e3d

This might be useful for contracts, as well. Editing points and having history, being able to comment on specific parts etc.

Not 100% sure on how to solve the "legally binding" part if the other party declines digital signatures, but it'd be a good way to get the contract written and agreed on.

I wonder if you might prefer something like Darcs to Git. I've heard a lot of amnbivalence about Darcs as a practical software project (just seems to be losing momentum, and it has never matched Git for speed), but I wonder if the model is more apt for writing than Git.

Can't emphasise enough how important offline editing would be for something like this. As a professional writer (whose co-written a book), you can't rely on always having internet; but you always need to be able to work on your current project.

This sounds great.

I am also working on something similar called Papyrus. It lets authors create and edit ebooks simultaneously.

http://papyruseditor.com/en/home/collaborate/new

This might be naive, but don't authors use Latex to write? So, in a sense why can't they use github as it is to collaborate on writing?

(I guess it goes without saying, I do not have any experience or knowledge about the writing process authors follow)

It's sad for me to see what you are doing because I had the same idea like 5 months ago. But I'm not that good on putting ideas into actions so it didn't go that far. :(

That'd be an awesome tool for witers, bloggers, editors, etc.

Keep up the good work! =)

I wonder if Github is working on something like this.

I'm sure they're familiar with the idea and they seem like the best team to bring it online. They're still a startup looking for growth opportunities: this one seems obvious.

Do authors really collaborate as much as developers do? Wikipedia is an example of people collaborating on writing something, but the subject matter is quite different from the kinds of writing that gets done.

> What's so awesome about GitHub?

It isn't all reasons mentioned in this article ... the most awesome about github are the developers. This is the biggest reasons why a "github for writers" could fail

I am trying to do the same. I never even thought about sharing this concept till I could launch. I know for sure this a great area to work on and it will make a huge impact for collaborative story telling.

Just a heads up: I just tried to signup for the newsletter and got a nasty Rails error page (500). It turns out Iwas browsing with JS turned off (using NoScript). It worked fine after activating JS.

As I was reading all these words about project collaboration I started getting more and more excited about the thought of myself collaborating in your project. Until I saw it was not open sourced...

Wooki* is a smart tool to write documents. It let you publish them and get reviews

https://github.com/spreadthesource/wooki

Github for lawyers, laws, court records, etc. I wonder how the lobbying industry would change if there were highly-public records of everyone's additions to a bill.

Github works on source code, which is text files.

Writers work on English text (or other languages) which is text files.

What does this do that Github itself (or equivalent - bitbucket, etc) does not?

Just today, I was checking out the JS API wrappers for Github API v3 and I found Github.js. There, I discovered this site called prose.io

Check it out! It's awesome!

This site doesn't work very well on mobile http://imgur.com/opmjUGn

Ah I know.. I've been meaning to fix that.

Fixed now btw.

Could this be used for other text-based documents?

I am thinking of accounting systems such as ledger-cli which let you keep your accounts in plain text files.

So I get the vibe that this will never be open sourced but is a side-project of yours that (should it come to fruition) will be run as for-profit.

Am I right?

I thought https://poetica.com/ was doing just this