I’ve been looking for a database for quite some time now and I have found some sources [1][2]. I decided to go with number 2, which is a similar type of program for Python. Number 1 is focused on Japanese characters (Kanji) where in essence are the same as Chinese characters, but some characters are different, thus character decomposition will be different. Number 1′s decomposition however is more thorough. Hopefully I’ll combine in the future. I took the second database and wrote a module for Node.js, called Hanzi. I know, how original. For the non-Chinese learners, Hanzi means Chinese characters in Chinese.

It was interesting to write my own module for Node.js, as it is something I have not done before. I always Node Package Manager for many awesome utilities in my Node.js projects. This time I filled a gap and published my own.

It’s a bit cumbersome at the moment and definitely has a few errors, but this is just the start. So head on over to HanziJS.com to check the module in action.

I’m going to assume that you have already programmed your app and can run it locally. This is not a HOW TO program Node.js post, but rather a tutorial on how to get the app online on Heroku.

Heroku Prerequisites

1) Switching to Node.js v0.4.7

Heroku says that you need to use Node.js v0.4.7 as this is what their servers use. It’s also a stable version. I was coding however in > v0.5. How do you change your Node.js version? Node version manager is your candidate! This simple script allows you to have numerous Node.js versions running on your computer. The github page has easy instructions to get this going. Warning: if you reboot your computer, Node.js defaults back to your originally installed version. Just activate Node Version Manager again, as this might create errors when you try to run your app again. Something like this:

FATAL ERROR: v8::HandleScope::Close() Local scope has already been closed

2) Using variable ports

Before hosting on Heroku I used fixed ports for my apps. You need to change this as Heroku does not like this. Where you define on what port your app should listen to, just add this:

var port = process.env.PORT || 3000;
	app.listen(port);

This is expressjs documentation. But the “process.env.PORT || 3000;” is the important part. You also set this up later on Heroku where you need the variable port.

3) Adding the correct files

Now this is where I struggled a lot, because Heroku does not really explain these things to the n00bs like me. You need a package.json and Procfile in the root of your Node.js package. These two files help Heroku determine how to startup your service when you deploy it. Think of them as files that act on your behalf to start the server. They are instruction files.

3.1 package.json

First create a package.json file that sits in your root folder where your node app is running from. Here is an example that I used for dotJunky.

{
  	"name": "DotJunky",
	"author": "Niel de la Rouviere",
	"description": "A Static File Site that sells domain names",
  	"version": "0.0.1",
  	"dependencies": {
    	        "express": "2.5.x",
		"now": "0.8.x",
		"mailer": "0.6.x"
  }
}

Those are the bare minimum details that you need. The main importance of the package.json file is to manage your Node.js module dependencies. If you need Heroku to install modules using NPM (Node Package Manager), then you need to add the dependencies. In the case above, you can declare the version numbers. The ‘x’ is a wildcard.

To test if your package.json file works correctly, run “npm install” in your command line in your root directory, where the package.json file is, and it will, if configured correctly, install the dependencies locally under the folder “node_modules”. Just add a .gitignore file also to your root directory that ignores “node_modules” as this does not need to be sent to Heroku, because Heroku creates this itself by using your package.json file.

3.2 Procfile

The Procfile tells the Heroku server which file (and command) needs to executed to run your app. Think of this as the command line “node app.js”, but in a file. A general Procfile looks like this:

web: node index.js

Replace the “index.js” with whatever your first file is to run, “server.js”, “app.js” whatever your preference is. The “web” prefix tells Heroku’s stack which Dyno/Process needs to run which command. As you can get workers and more web processes in Heroku, the Procfile will get more complex. You’ll need Foreman to start your Procfile on your local machine to test it. I now use Foreman to run all my Node.js projects. Its as easy as:

foreman start

This triggers your Procfile.

Just a note of warning, the Procfile needs to be capitalized, otherwise it won’t be found by Heroku.

Getting it online

4) Create a local git repo

Now that you have your node app, your package.json file and Procfile ready, you can start uploading it. However, first init a git repo in your folder. If you have a git repo already, then skip this step.

git init
git add .
git commit -m "First commit"

5) Install Heroku

Now that you have a local repo, you need to install Heroku on your local machine. It’s fairly simple. Just choose one of these options, in my case MAC OSX. This adds “heroku” to your command line. Everything on your local machine is now in order.

Now we are heading towards Heroku. First login:

heroku login

You’ll be prompted for your details and to add a ssh key. Enter them and add one.

6) Creating a Stack

Now you need to give heroku a command to create a stack for you. This is where your app will reside.

heroku create --stack cedar

If everything runs smoothly, you’ll see this type of response:

Creating sharp-rain-871... done, stack is cedar
http://sharp-rain-871.herokuapp.com/ | git@heroku.com:sharp-rain-871.git
Git remote heroku added

7) Push to Heroku

Heroku automatically adds a remote repository address for your local git repo under the name “heroku”. Now push your git file to heroku.

git push heroku master

This takes a bit more time as it installs the dependencies and starts the app. If everything runs smoothly, you’ll get a response like this:

Counting objects: 9, done.
Delta compression using up to 4 threads.
Compressing objects: 100% (7/7), done.
Writing objects: 100% (9/9), 923 bytes, done.
Total 9 (delta 2), reused 0 (delta 0)

-----> Heroku receiving push
-----> Updating alpha language packs... done
-----> Node.js app detected
-----> Vendoring node 0.4.7
-----> Installing dependencies with npm 1.0.8
       express@2.1.0 ./node_modules/express
       â??â??â?? mime@1.2.2
       â??â??â?? qs@0.3.1
       â??â??â?? connect@1.6.2
       Dependencies installed
-----> Discovering process types
       Procfile declares types -> web
-----> Compiled slug size is 3.2MB
-----> Launching... done, v2
       http://sharp-rain-871.herokuapp.com deployed to Heroku

To git@heroku.com:sharp-rain-871.git
 * [new branch]      master -> master

8) Final Heroku Configurations

Before you run your app, you need to send Heroku some configurations.

heroku ps:scale web=1

This scales how many dynos you use for your “web” process. 1 is enough for small apps and sites. It’s free too!

heroku config:add NODE_ENV=production

This is the variable port, but you need to add this so that expressjs can use it in some instances like caching (don’t ask me!).

Now, if you have done this all correctly, your node app will be online and ready to kick some ass. The emphasis on this tutorial is to first make sure everything locally is in order BEFORE you send it to Heroku. I don’t know how many times I pushed to heroku, got errors, went on troubleshooting, pushed again and repeat. In summary, make sure you have Node.js version 0.4.7, Procfile and package.json correctly configured as well as variable ports. If all these are in place, then pushing to Heroku will be easy. Good luck running those apps!

I’ve acquired some cool domains since I started domain trading. Some of them could work really well for startups, apps or brands. This is where dotJunky fits in. I sell these awesome domains straight to the entrepreneurs. It’s a tiny marketplace for these domains.

It is written in Node.js using Masonry, Bootstrap and jQuery. I learned quite a bit creating this, especially email handling. Quite easy with the right modules, but I haven’t done this before.

The front-end is written manually at this point, but if dotJunky proves to be popular among startups, I’ll think of creating an admin backend for myself to handle things better.

So, if you are an entrepreneur and need inspiration and/or a name for your new startup, dotJunky is for you!

A current project is called Rouviere. This is purely an electronic music project at the moment. Mostly dance genres, ranging from dubstep to electro house. However, Rouviere, is an umbrella project for music Simon (my twin brother) and I create. Sometimes, I do a whole song on my own. Other times we collab together. It’s mainly just a release for electronic music that doesn’t have a place for its own.

I just finished a song a few weeks ago. Have a listen!

The Souncloud Page has more tunes. Happy listening!

The response to it was just crazy. I thought some people might get a kick out of it, but not to the extent that it got the top link on reddit for 13 Feb with over 1900 net upvotes. The video now has about 430,000 views. I made this contraption, because I was bored with beating Quake on the hardest difficulty and wanted something different. I only uploaded it now, because I discovered my old phone’s memory card at that time.

That night when it hit the top spot on reddit’s frontpage I was blown away. This is what redditors and web loving geek dreams of. We celebrated at my flat with my flatmates with a few shots just for kicks. I had to go to sleep though. The next morning I was hit with a truckload of emails from Youtube comments. The video truly went viral. It got almost 200, 000 views in two days. It got posted on Crunchgear and a host of other sites as well.

I always wanted to see how it feels to be on the other end of going viral. And now I experienced it first hand. It is interesting. The Youtube comments as always are a blast to read. A few suggested ad deals for my video, but I declined. I was not in the position to make money of the vid for a bit of ad views/clicks. I got my few minutes of Internet fame and it was interesting. Maybe in the future some might re-post this and it will get a few more views again, but the height of the viral spread has almost completely halted.

The strangest thing was trying to explain to friends that I went viral by taping my mouse to my keyboard with masking tape. I remember my girlfriend trying to do the same, trying to explain to her friends what I had done. In the end, linking to the video was the best thing to do. Another odd thing was a realization that two different worlds were aware of my brief Internet fame. I would walk to class and it hit me that hundreds of people were probably watching my video as I walked.

Now it has over 400, 000 views and that is an immense number to comprehend. Yet, here I still sit, no-one really knows me. No-one actually really cares. Yet I was seen playing Quake by thousands of people.

I have been on Sina Weibo for a short while now. Here is my account. Sometimes it gets a but much on my Chinese senses, but I have found Weibo to be a very interesting experience. I found an article which explains the data structures really nicely. It makes for an appealing user experience.

On Sina Weibo’s official interfaces (Web and mobile), the equivalent of a Twitter retweet is instead shown as two amalgamated entries: the original entry and the current user’s actual entry — which is a commentary on the original entry, often with a mention of his sources, if the original entry was obtained from an intermediary.

At first it was daunting, but later it made more sense to me. Furthermore, Sina Weibo actually feels like how a proper microblog should be, with the addition of comments on updates. This makes it feel much more relevant when you are discussing an update. For instance, when communicating via Twitter, your stream is filled with @ replies, when these could easily have been contained within a comment system. It just makes it feel much more cleaner.

The communication shouldn’t be on the channel itself, but rather separated into it’s own system. It’s like the equivalent of sitting at a bar where people shout at each other. Eventually you’ll be so battered with cross-talk that it becomes too much and irrelevant.

However, it must be added that cross-talk allows for interesting eavesdropping.

Sina Weibo is like an interesting mix of Tumblr, meets Facebook updates, meets Twitter. It works for me. Besides being filled with requests by people to help them with their English, Weibo has been interesting experience. It feels as if people comment a lot more on updates. This is probably as opposed to Twitter, where sending @replies would later start to feel like clogging the system.

I’m interested to see how Weibo develops. It also hosts it’s own pictures. Which is nice too.

In other news, Simon and I also reached a milestone in Google Adsense. Our first cheque will be on its way very soon. :)

Apart from Social Mandarin, I’ve got another Chinese project called Confused Laowai. Laowai is a term for a foreigner in China. I started this personal blog in January 2010 as a way to post about my progress as I go through my studies in Mandarin and my Mandarin journey. Confused Laowai has been a very interesting project experience.

Response

To me, Confused Laowai, is the closest I’ve come to a proper blog/reader relationship. Yes, Twindie.net has more readers and more hits, but Confused Laowai has more engagement. I feel like I have a proper community that follows my blog. This has been really good.

Growth

The blog has grown slowly in the start. I posted comments on numerous similar blogs by Mandarin learners and created a Twitter account that focuses on my Mandarin learning missions. I have also posted most of my blog posts on Social Mandarin and have posted one or two posts on Hao Hao Report which got me nice hits and readers in the process.

The Future

As long as I study Mandarin, this blog will be alive. Which will probably be quite a long time. It will only grow and I feel that I can create an even better blog/reader relationship, learning a lot in the process. I intend to do a bit more video blog posts too. It has gained PR2 in a very short while and is getting hits from Google from a wide range of search terms, probably the most of all my web projects. It doesn’t get the most search hits, but definitely the widest!

So, if you are interested in reading more on my Mandarin learning expeditions head on over or grab the RSS feed.

I’m finishing my Mandarin studies this year as one of my majors. When I started learning Chinese I primarily got formal instruction from University, however after class I would get back and still be interested in the language. I started following blogs and other learners, where up to this point my Google Reader is almost entirely made up of Chinese blogs and China interest websites. Thus the idea came for Social Mandarin.

The Idea

I felt that there was a disconnect between Mandarin learners on the web. I thus created Social Mandarin as a social bookmarking site where people can share online Mandarin resources with each other, find the resources that I took almost two years to amass. Eventually Social Mandarin should be the first site you visit when you are looking online Mandarin learners and online resources.

Development

I haven’t got much web development knowledge, but I’ve always been a tinkerer: using existing platforms and scripts and reverse engineering and adjusting them to suit my needs. Thus, I used Drupal with the Drigg module, that turns Drupal into a Digg-style site. It took me quite some time get everything up and working. I made some minimal changes to the design as well.

I had some problems with spam early on, but I’ve managed to get two moderators to help me and I’ve also since then installed captcha forms and various other protection systems. I’m however still sitting with a trackback problem, which needs to send a trackback/pingback to blogs if a story is posted. Alas, Drigg support has been dropped and most of the solutions I tried have not been working. Will have to look into it later.

Progress and Lurkers

Social Mandarin has been received quite well since its inception in March 2010. It has grown steadily since then. However, although there seems to be interest in the site, most of them are lurkers, only utilizing the content and not posting their own or other links. This, has lead me to post the majority of the almost 200 stories/links currently on the site. There are some other active users, but they are only a handful. I need to figure out how to provide a better incentive for users to post links, besides self-promotion. Alas, I don’t mind posting most of the content for now.

Marketing and Monetization

I have experimented a lot with social media marketing via Social Mandarin. I currently have a Twitter account for it, posting the latest articles. I got most of the followers via blogs posting my service and me following users who tweet about Chinese and Mandarin. I set up a search stream in my preferred Twitter app, Hootsuite to track people talking about it, and if someone says something relevant as I go through it I follow them. This is not an auto-follow script.

I have also put up two Facebook ad runs for the Facebook page. In the end I got about 130 new fans/likes within two weeks altogether. I opted for social media ads, because this gives me recurring traffic due to the users seeing new articles being posted to the page every time. Also, like the site suggests, this gives me the ability to interact with the users.

I’ve also experimented a lot with ad positions on the site. Currently, Social Mandarin has the highest CTR of all my web projects. It unfortunately is still a small amount of money and still needs to gain a pagerank, but once it does, it will surely pick up a lot more traffic.

The Future

Currently, the site is gaining momentum and I’m waiting for it to pick up on some Google juice. I’m also looking into redesigning the site. I think this site, as long as people post links, me included, will grow no matter what happens as it works on sharing. I’ve also got some ideas to enhance the social aspects of the site, to get users to interact with each other. One of these were the blog features, where users can have their own blogs on the site. However, more incentive needs to be provided and that’s where my brainstorming needs to go.

The Site

Simon wrote the site from scratch. I designed the site, but did not end up coding the design. Since the start of the site, we were always trying to improve the design, but we never came round to it. At first the site was popular. It got a few mentions on forums. It was during the FML popularity last year. The site is basically a clone, however I think we got into it just a little bit too late.

Marketing the Site

We did almost no marketing for the site whatsoever. It basically got known via word-of-mouth from our Twitter accounts (@nieldlr & @shotbeak). I did experiment with some Adwords in January of this year to test it: never used it before. It got some hits and submissions from it. I think if Simon and I put some more effort into marketing the site, it can speed up again and gain some momentum.

What now?

The site is still there. It does get a few thousand pageviews a month, from Google and just general linkage. Submissions are however irregular. We need more submissions and quality ones to make it work. Thus, it gets sort of like a club, if the dancefloor is empty, no-one wants to dance. We need to reach a good level of readers and submitters. We have played with the idea of selling the site, so someone else can take over. However, I think we wan’t to hold onto it and try and juice it up in the future. The first thing on the agenda? A redesign!

So if you are a nerd and want to share your adventures in the real world. Submit one right now!