19 comments

[ 8.9 ms ] story [ 72.5 ms ] thread
While the idea is quite nice, there doesn't seem to be any indication as to what version of markdown is in use. Probably not Leanpub - given that there aren't any published parsers - but is it Kramdown/MultiMarkdown/GFM/Classic?
It's GFM for now. I'd like to build it out more and have other formats and parsers.
Wow, CI for books! Cool idea :)
I'm actually writing a book now, and would love to use this. The first thing that pops into my mind is: are there any example books I can look at so I can see what this will look like?
Render in chrome in android is terrible
Should be more usably now. I'm going to be devoting more time to mobile usability.
...and it isn't open-source. Why?
Would love to and I have no objections to it. I work on some popular open source projects. It would be very difficult to work on this project while growing a community for it as well(this has been, for the most part a side project).
Nonsense. If you're hoping to earn your living on this, and it's good enough that people want to use it, they can cough up $5 a month for it, or use the free plan. And it looks like it's attracting some good attention...

People have to realize that even open source devs have to make a living, and not all of us are fortunate enough to work for an open source friendly employer. In fact, if you're like me, a single well-paying project like this one would greatly free up your time available for open source projects and increase your open source contributions.

I've seen what happens when open source devs rely solely on user donations, even ones with really popular and useful projects: they starve.

Side note: assuming the cost is indeed $5 and $19.99 a month, you need to make the "per month" explicit. Right now, if someone signed up and you dinged them for multiple months, they could technically but correctly lodge a complaint with your payment processor for illicit charges.

Good catch, its /month when actually signing up but I've updated it.
Plans for Bitbucket/3rd party git support?
Bitbucket for sure. I'd be exhausting my Git fu, if I where to guess at what it entails to implement the 3rd party support. So, I shy away from promising it.
Nice work. Similar projects are https://www.softcover.io/ and https://www.gitbook.io/.

I'm in the same space with Penflip (https://www.penflip.com/), though I'm aiming to be more non-developer friendly. The books are still backed by git repos (with full git access if desired), but it's masked with a web interface for the most part.

I've been following Penflip since you posted your Show HN(awesome damn job by the way). I was building both a front end editor along with the build system. It ended up being too much to build, so I focused on the generator portion of it.

I do know about softcover and gitbook(both great products). My hope is to work on the generator rather than providing a store front/ public facing sites.

Next thing I would like to tackle is code insertion from source files. Where the source files can be tested and Arturo can put snippets from the source files. Would be great to have community books and examples that are self tested.

Leanpub does offer GitHub integration, including private Github account access. I am using it right now.

And you can setup a commit hook to trigger the preview build automatically.

Given that common base, I am curious what the specific other difference are. Especially to justify paying for a private repo account.

Leanpub Github integration instructions are @ http://blog.leanpub.com/2014/04/github-integration.html

They are very awkward instructions. I have yet to use it, so I can't comment on the quality of it. Arturo pushes a build and a diff for each PR(attaches to the link on the orange/red/green dots). This is a start of Arturo and I would love it to be Open Source and Developer friendly, focusing more on the Books + Code aspect.

I got it working the first time, so I guess it is not terribly confusing (for a developer).

It does build all the versions of the document on each push, which I find makes it a bit slower for frequent check-in.

The diff idea is interesting as a value add, especially - I project - for per-diff discussion or so.

In any case, this was not a disparagement. At this point, the more of these ideas we have, the better it is. I just wanted to point out that at least some of the functionality was already implemented somewhere else.