Hacker Newsnew | past | comments | ask | show | jobs | submitlogin
Raneto: Markdown-powered knowledge base for Node.js (raneto.com)
62 points by gilbitron on June 2, 2014 | hide | past | favorite | 36 comments


Well, that was easy. Went ahead and took our long-standing WIP wiki which was already markdown and threw it in here. We've been meaning to do something like this for a while, you just did it.

Agreed with the point about security. We have a public part we want to make into a general KB but also a private part for internal. If anyone comes up with something, I'd love to see it.

Lastly, because what comment wouldn't be complete without some sort of "criticism" on someone else's awesome, hard work, what prompted the decision to use the /* */ commenting for page-specific config? In my experience (maybe this is just because I've used jekyll too much) it seems that Yaml frontmatter is pretty typical for .md file config.


For security, you could move this app behind an nginx proxy and handle auth (to say /admin/) via client SSL certificates. We do this for a number of internal applications that are exposed on the public web.


Thanks. To be honest I just used a /* */ comment block because it was simple, but I'll have a look at implementing YAML.


How about gray-matter[1]? It supports YAML, TOML, JSON and Coffee. It can also auto-detect the type of front-matter.

[1] https://github.com/assemble/gray-matter#lang


I played around with this last night. Woke up this morning, visited raneto.com again, noticed that the website claims that for "commercial" projects, it requires a per-domain license. Except the source code on Github is licensed under MIT. I'm not sure how to reconcile those two things. http://docs.raneto.com/commercial-licensing


I noticed that too! Why don't they sell it as a service instead of trying to sell open source code?? Anyone who knows enough to use it knows enough to make it themselves


Do you still have to style the knowledge base to match your site's design or use a layout?

We've been working on a hosted knowledge base that you drop on a page and it inherits the design you already have. Our editor supports both WYSIWYG and Markdown too. Looking to see if people are interested and get feedback. http://userdeck.com/guides


Really cool stuff! I actually wrote a [flat-file CMS](https://github.com/AntJanus/popstar-cms) in Node as well, wasn't disappointed to see that my code and Raneto's code bares a good deal of similarities!


Thanks. I actually created a flat-file CMS in PHP called Pico (http://picocms.org) so I built on some of the lessons I learned creating Pico when building Raneto.


I'm looking through it, I'm impressed by your plugins. I'm hoping to introduce that kind of structure into my application (hooks for remote posting, etc.). But gotta add one piece of functionality at a time!

I haven't looked through your code completely so let me ask you something, what kind of an object do you pass to your templates? Do you pass the entire tree of information (all contents parsed as a JS object), or do you pass the current + children + parent?

I'm just wondering. What I'm currently doing is sending the top-level content, current page, and children along with all of their information.


I had almost this exact idea recently, because I really wanted a minimal KB/wiki that I could edit in my text editor and collaborate on with others. It's encouraging to know that others feel the same way. You may have a competitor in a short while ;)


Please, let us know!


Nice work! Is there any difference between a knowledge base and a wiki?

And while on the topic of wikis, I wanted to plug dokuwiki which is an awesome feature-rich flat file wiki. I've been managing a lot of my writing as .txt files, even the ones with lots of math blocks---thank you MathJax! Also, the docutexit plugin allows me to export as .tex.


I suppose the difference between Raneto and a "wiki" is that Raneto doesn't have all the multi-author features and collaboration that a wiki offers.


Cool! I've been hoping to see more wiki/knowledgebase type static site generators.

I will admit that I fear for its longevity though. Any particular reason why you decided against building on top of something like assemble.io or wintersmith.io? Probably would have gotten a lot of functionality for free.


I did consider simply building a KB with other static site generators, but I wanted something specifically designed for Knowledgebases and easy to setup and use.


I have a question: what is the typical workflow for a user to "update" a markdown file on a static site that generates html ?

I'm okay with ftp being so 90's but I don't see regular joes git-pushing their website.


Doesn't seem to work on Windows. Links resolve to http://localhost:3000/C:/Users/... i.e. full path to content directory.


The licensing is confusing. The GitHub project is released under MIT, but the site says you need to have a commercial license to use the software?


The paid license is for commercial products. But that's still confusing.

Raneto may be used in commercial projects with the one-time purchase of a commercial license.


It just seems that the commercial license is incompatible with the MIT license. With the MIT license there is nothing stopping someone from using it in a commercial product, the MIT license doesn't have that restriction.

Even more confusing is if someone were to fork the project, would a user of that fork need to pay Raneto to use it a commercial license?


It would be nice to have some sort of security option around accessing the KB. Does anyone know of something like this with a login system?


Raneto is secure because it doesn't have any kind of interface that can be hacked (e.g. a login page). To make changes to Raneto content you need to have access to your server file system.


i think the question was asking for most of the content publicly accessible but some of the content hidden behind a login. google sites does something like that but i haven't seen any flat-file cms'es that offer it.


I believe he's not talking about authentication, he's talking about access/authorization.


Why the need for Node.js? Couldn't this simply be a static site? (I know there are tons of these, but most are "blog oriented)


You could just create a static HTML site yes, but getting it generated for you from Markdown files is super handy. To generate it on the fly it needs to be powered by some kind of server (e.g. Node, PHP, Python etc). Nodejs was just a good choice of platform for this kind of thing.


Sort of related: how do people handle images when using markdown documents for local/offline storage?


I just put it in a specific directory, like root/data/images


Good work. :) But without the cross reference module, it is just like a simple blog system.


It could easily be "hacked" into a blog if you wanted to do that. At its core it is just a static site generator.


No, I don't need another static blog. I mean (bi-directional) cross reference is very important for memorization and even better than a simple search engine. Maybe you can use Disqus comment box to get related article but that's not good enough.


Just what my colleague needed for our small KB redesign. Thanks for discovering!


Yep.. Was also looking for a markdown kb. Pref. with flat files


Any plans to integrate API blueprints?


Looks very nice




Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: