While working on the rewrite of my simple bookmarking service, Picomarks, I needed the capability to easily embed content from video sites like YouTube and Vimeo and image sharing sites like Flickr.

Originally I hacked together some code that would generate a YouTube embed from a YouTube video url before realising that I didn’t want to do this for five or six sites. So I searched around and came across the oEmbed standard, created by Leah Culver and others while they were developing Pownce. oEmbed is a super simple way to get the embed code for all sorts of content including videos and images. For example if I found a cool image on flickr. To get the embed code (which is a ridiculous process on flickr without oEmbed) I make a simple HTTP GET request to Flickr’s oEmbed API.

http://www.flickr.com/services/oembed?url=http://www.flickr.com/photos/xdjio/226228060/&format=json

Flickr responds with an oEmbed response which we requested to be in json.

{
    "version" : "1.0",
    "type" : "photo",
    "title" : "Wasteland",
    "author_name" : "xdjio",
    "author_url" : "http://www.flickr.com/photos/xdjio/",
    "cache_age" : 3600,
    "provider_name" : "Flickr",
    "provider_url" : "http://www.flickr.com/",
    "width" : "500",
    "height" : "336",
    "url" : "http://farm1.static.flickr.com/64/226228060_c88ba6cf6b.jpg"
}

Super simple, right? So I started looking for a way to do this in ruby and I found the ruby-oembed library on GitHub, there was no documentation; so I made my own library, OhEmbedr.

To install just install the gem:

# If you don't have gemcutters in your gem sources already
gem sources -a http://gemcutter.org
 
# Then just gem install
gem install ohembedr

Then to use OhEmbedr is just as easy:

require 'rubygems'
require 'ohembedr'
 
begin
    o = OhEmbedr::OhEmbedr.new(:url => "http://vimeo.com/6382511", 
        :maxwidth => 600)
    embed_data = o.gets
rescue OhEmbedr::UnsupportedError => error
    # URL not supported
end

We wrap it in a begin rescue block because OhEmbedr will raise an OhEmbedr::UnsupportedError if it encounters a URL which isn’t in its providers list, meaning we can pass any url to OhEmbedr and if this exception is raised we know that this url can’t be embedded by oEmbed. Currently the following providers are supported:

• YouTube
• Vimeo
• Flickr
• Qik
• Revision 3
• Viddler
• Hulu

I do wish more people would adopt oEmbed, the idea of having a standard API for embedding data is really nice. Of course there are a few problems with oEmbed for example having to dig around on the provider site to find the oEmbed API URL but I don’t see anyone suggesting anything better.

So check out the OhEmbedr source on GitHub or its gemcutter page.

Wonderful image from the brilliant xkcd.

If you’re reading this, you have quite obviously found my new blog, with a new design and the same rarely updated content! So what I thought I’d do this time is put the focus on something else, make benmcredmond.com about me not my blog (keep the blog here for when I do have something to say though). Enjoy!

P.S. Been working on something pretty cool the last few months so watch here for updates.

Picture © Timothy K Hamilton

It troubles me greatly to hear that people see me as an expert or an authority, and not a fellow amateur. If I’ve learned anything in my career, it is that approaching software development as an expert, as someone who has already discovered everything there is to know about a given topic, is the one surest way to fail. Experts are, if anything, more suspect than the amateurs, because they’re less honest. You should question everything I write here, in the same way you question everything you’ve ever read online — or anywhere else for that matter. Your own research and data should trump any claims you read from anyone, no matter how much of an authority or expert you, I, Google, or the general community at large may believe them to be. Jeff Atwood

If you don’t already read Coding Horror, check it out, by far one of the best code-orientated blogs out there.

To use this script you must get an api key for Amazon Web Services and Amazon’s Product Advertising API.

To start with we do a search for “Album, Artist” (this I have found works better than searching for “song, artist” after all we are looking for “album” art), we take the ASIN for the top result. The ASIN is a unique identifier Amazon gives to every item in its catalog.

We then lookup the particular item that this ASIN refers to and ask for Amazon to give us the images for this product (lines 15-21).

Hope this explanation helps someone looking to do something similar.

From what I can tell many parents dread the idea of Transition Year, their kids will be doing “nothing” for a whole year. Of course, you have to put effort in to get something out of Transition Year but if you do what you get out of it its unbelievable.

For those who don’t know, Transition Year is the fourth year in Irish secondary school, where you transition between the Junior Certificate course and the Leaving Certificate course. In this year you do random things as will come to light in the rest of this blog post.

In transition year I have had some of the most amazing experiences which I certainly would not have experienced had I gone straight in to the Leaving Certificate course.

For a start I had my work experience in The Independent, a totally new environment. I didn’t think that I would like working in an office at all, however since I was doing what I love, it was brilliant. Now I knew I could work in an office and not be bored out of my mind.

Then I wrote a programming language. If I had been doing the Leaving Cert course there is no way that I would have been able to do this. Writing an extremely simple programming language, which I never finished, was still probably one of the most difficult things I’ve ever done but the amount I learnt was amazing.

I learnt about data structures such as Binary Search Trees, Linked Lists and Hash Tables. Things I would never have come across till college through my dabbling in web development. I learnt about the underlying structure of compilers and interpreters, all of which made me a fundamentally better programmer.

Furthermore If I hadn’t written this language, I would not have entered the Young Scientist, here I met some of the people who I would now consider some of my best friends.

Up to this point I thought I was having a pretty good year. However it was just about to get better. Next I had my week of work experience in Contrast where I learnt a huge amount about how much work 4 guys can get done by themselves, how less is more and how unbelievably awesome it would be if I could work somewhere like that when I left college.

The week after Contrast, I experienced my first “product launch”. It was nothing big but it got a great response (launching what I had built at Contrast). Followed immediately by attending Future of Web Apps conference in Dublin.

Had I been doing my Leaving Cert I might have still attended FOWA but I certainly wouldn’t have spent a week in Contrast and if I did not spend a week in Contrast, I would not have met Paul Campbell and therefore I wouldn’t have had lunch with Blaine Cook and I would not have shook DHH’s hand (hey it’s a big thing for me!).

However at this point. I thought it was over. The good part of Transition Year was over. It was the day before I went away to France for four weeks to stay in a French boarding school to learn French, I was dreading even the thought of it.

I have never been so wrong in my life. What was to follow was the best four weeks of my life, it was so good I stayed two extra weeks and if I could, I would go back immediately. In France I met just the most amazing people and had the time of my life.

When I came back, to put it lightly, I was a tad depressed. However I came back to some amazing friends who without the Young Scientist 5 months ago, I would never had been friends with.

Then today, today was my last day of “supposed to be in school” Transition Year and I was in Camara developing software to be sent to developing countries in Africa to help easily deploy Moodle, school management software.

So to all those people who skip transition year, to all those people who say it should be abolished or those who just complain about it. It was an unbelievably amazing experience, one which I think everyone should take part in. Just my two cents.

Stikked has been very slow since the latest update, there are some code fixes I’ve made which will be pushed shortly I also made some server side improvements which caused Stikked to be down yesterday:

1. I replaced Ubuntu Server with Debian on the Stikked VPS. (This shaved off quite a lot of RAM)
2. I installed and configured mod_deflate to gzip all html, css and javascript. This halved the size of the page download.
3. I also setup monitoring tools so next time Stikked goes down, I will know straight away and have more information to determine what happened.

Have a nice day, Ben

Enjoy

This week during my week of work experience I was given the task of creating an application for mapping crimes in Dublin. So here I am sharing with you what I learnt. We’ll create a simple application that loads points from a database and shows them on a map. Using Javascript, PHP, MySQL and CodeIgniter, I’ll assume you have a webserver running and understand PHP and MySQL. Note: I think I may have mixed up longitude and latitude throughout the tutorial. It’ll still work, they’re just labeled incorrectly

Lets get setup

Fantastic! We also need to get a Google Maps API key from Google Code. We’ll need this later on, so keep it safe.

Time to start real work

Open up the application directory (in the system directory) and lets get started. We’ll mostly be working in the application directory, all of our code, goes in here. We’ll start by renaming controllers/welcome.php to main.php and opening up the routes.php (in the config directory) file.

Routes.php defines the urls for our application. Find the following line

$route['default_controller'] = "welcome";

And change it to:

$route['default_controller'] = "main";

We need to do this since we renamed welcome.php to main.php. This process isn’t entirely necessary but I like to do it.

You also need to open up config.php (in config) and change the base_url setting to match the base url of your site.

Now lets open up main.php.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17

<?php
 
class Welcome extends Controller {
 
	function Welcome()
	{
		parent::Controller();
	}
 
	function index()
	{
		$this->load->view('welcome_message');
	}
}
 
/* End of file welcome.php */
/* Location: ./system/application/controllers/welcome.php */

Note: We need to replace the instances of Welcome on the lines 3 and 5 with Main because we renamed our controller.

Controllers are the link between the view, where we show the user the data and the model where we get the data from (a rough explanation of MVC, I know.). So lets start by doing a nice little hello world. Change the $this->load->view line, to load ‘main_view’ instead of ‘welcome_message’.

Open up the view directory and delete welcomemessage, we won’t be needing it anymore. And create ‘mainview.php’, note how we didn’t have to specify the .php suffix while loading the view, put the following into your view.

<p>Hello world?</p>

If you refresh in your browser right now you should see your lovely hello world message, cool huh? Well maybe you’re thinking that was a lot of effort for just that. It’ll help in the long run, I promise!

Setup the Database

Woops! I almost forget, we need to setup the database. Open up phpMyAdmin (or some mysql interface) and create a database called ‘parties’ and run the following query:

CREATE TABLE IF NOT EXISTS `locations` (
  `ID` int(10) UNSIGNED NOT NULL,
  `title` varchar(64) character SET utf8 collate utf8_unicode_ci NOT NULL,
  `summary` text NOT NULL,
  `longitude` double NOT NULL,
  `latitude` double NOT NULL,
  PRIMARY KEY  (`ID`)
) ENGINE=MyISAM DEFAULT CHARSET=latin1;

That should create all the fields we’ll need for our application. All you need to do now is fill out the relevant info in config/database.php.

Work with a beautiful model

Time to create a model. A model is how we interact with the database in CodeIgniter or any MVC application. Create a new file in the models directory called ‘locations.php’. I find it best to name models after the tables which they represent. Copy the following code into your model:

<?php
 
class Locations extends Model {
 
	/*
	*	Get the locations we want to put on our map
	*/
	function get_locations() {
		// Load the database class
		$this->load->database();
 
		// Get all the entries in the parties table.
		$query = $this->db->get('locations');
 
		// Are there any results from our query?
		if($query->num_rows() > 0)
		{
			// Set a counter to 0
			$n = 0;
 
			// Create our array
			$data = array();
 
			// Loop through the results filling our array with the locations
			foreach($query->result_array() as $row)
			{
				$data[$n]['title'] = $row['title'];
				$data[$n]['summary'] = $row['summary'];
				$data[$n]['longitude'] = $row['longitude'];
				$data[$n]['latitude'] = $row['latitude'];
				$n++;
			}
 
			// Return the array
			return $data;
		}
 
		// Returns false if we didn't get any results from our query
		return false;
	}
 
	/*
	*	Create a location for our map
	*/
	function create_location($data)
	{
		// Load the database class
		$this->load->database();
 
		// Insert the data into the table
		$this->db->insert('locations', $data);
	}
 
}

Lets walk through how each method works. get_locations will provide our map with all the data we need to generate a map. We load the database library, get all the entries from our locations table. We then check if we got any results, to avoid errors and fill an array with our data. Simple enough.

create_location couldn’t be simpler, it just inserts an array which is passed to it into the database. Which of course in real life would be validated and cleaned before being inserted into the database, but this is just a tutorial, ok?

Back to the controller!

That’s our model done. Now we need to flesh out our poor controller.

<?php
 
class Main extends Controller {
 
	function Main()
	{
		parent::Controller();
	}
 
	function index()
	{
		// Load the locations model
		$this->load->model('locations');
 
		// Populate an array with all the locations we want to put on our map.
		$data['locations'] = $this->locations->get_locations();
 
		// Load our view
		$this->load->view('main_view', $data);
	}
 
	function create()
	{
		// Load the url helper, for the view and the redirect below
		$this->load->helper('url');
 
		if(!isset($_POST['submit']))
		{
			$this->load->view('create_view');
			return;
		}
 
		// Load the locations model
		$this->load->model('locations');
 
		// Fill the array with what we want to put in the database
		$data = array(
				'title' => $this->input->post('title'),
				'summary' => $this->input->post('summary')
			);
 
		/* Get the longitude and latitude out of the stupid (x, y) string. */
		// Strip the brackets
		$location = substr(substr($_POST['location'], 0, -1), 1, -1);
 
		// Split it up into two numbers
		$location = explode(', ', $location);
 
		// Add our longitude and latitude to our array
		$data['longitude'] = $location[0];
		$data['latitude'] = $location[1];
 
		// And put it in the database
		$this->locations->create_location($data);
 
		// Redirect the user to the homepage
		redirect();
	}
}

This is also pretty simple. The index method is the first method called when this controller is called. So we want this to load our ‘main_view’. We fill an array with the data from our database and pass this array to the view through the second parameter.

Our create method will let us create a location to put on the map. It first checks if the form has been submitted and if not loads a view with the form on it (we’ll make that in a second). If the form has been submitted we fill an array with the form data, do some funky stuff to get the longitude and latitude. Finally we put it in the database and redirect the user to the homepage.

Create won’t work yet!

Well none of it works yet, but for create to work we need to tweak our routes.php just add the following line:

 
$route['create'] = 'main/create';

This tells CodeIgniter that when someone visits /create they want to load main/create. Also we need to create the view so go ahead and create create_view.php.

Time to make our create_view.

<html>
	<head>
		<script type="text/javascript" src="http://maps.google.com/maps?file=api&amp;v=2&amp;key=YOURAPIKEY"></script>
 
		<script type="text/javascript">
			function initialize()
			{
				/*** Create our Map ***/
				var map = new GMap2(document.getElementById("map_canvas"));
				var mapControl = new GMapTypeControl();
				var center = new GLatLng(53.332923,-6.239548); // Center the map on dublin.
 
				map.setCenter(center, 11);
				map.addControl(mapControl);
				map.addControl(new GLargeMapControl());
				map.enableScrollWheelZoom();
 
				main(map, center);
			}
 
			function main(map, center)
			{
				// Create our Marker
				var marker = new GMarker(center, {draggable: true});
 
				// Set the lattiude longitude hidden field to the center of the map, incase they don't move it.
				document.getElementById('latlng').value = center;
/*
				/*** Detect when the marker is moved and update the hidden field ***/
				GEvent.addListener(marker, "dragstart", function()
				{
				  map.closeInfoWindow();
				});
 
				GEvent.addListener(marker, "dragend", function()
				{
					document.getElementById('latlng').value = marker.getLatLng(); // Update the hidden form field.
				});
 
				map.addOverlay(marker);
			}
		</script>
	</head>
 
	<body onload="initialize()" onunload="GUnload()">
		<h1>Tell us about a party!</h1>
 
		<form method="post" action="<?php echo site_url('create');?>">
 
			<strong>Title: </strong><input type="text" name="title" />
			<strong>Summary: </strong><input type="text" name="summary" />
 
			<br/><br/>
 
			<div id="map_canvas" style="width: 500px; height: 300px;"></div>
			<input type="hidden" name="location" id="latlng" value=""/>
 
			<br/><br/>
 
			<input type="submit" name="submit" value="Submit"/>
		</form>
	</body>
</html>

Note: in this example and the next you need to fill in YOURAPIKEY with the api key you got earlier from google.

Looks confusing? It’s really not. When the page is loaded the onload=”initialize()” causes the initialize function to be called (which is javscript), which creates a new map on the #map_canvas div and then sets the center, etc.

The initialize function then calls the main function which puts a marker in the center of our map and adds listeners which detect when it’s been moved and if it has updates the hidden field with the new latitude and longitude.

If you know browse to example.com/index.php/create you will see the beautiful (well it’s rather ugly… but… yeah ) form we just created. If you fill in the details and click create it will insert it into the database and redirect you to our front page.

Hello World

Hi! But that’s not what we want, we want our map! Finally we just need to update our main_view.

<html>
	<head>
		<script type="text/javascript" src="http://maps.google.com/maps?file=api&amp;v=2&amp;key=YOURAPIKEY"></script>
 
		<script type="text/javascript">
			function initialize()
			{
				/*** Create our Map ***/
				var map = new GMap2(document.getElementById("map_canvas"));
				var mapControl = new GMapTypeControl();
				var center = new GLatLng(53.332923,-6.239548); // Center the map on dublin.
 
				map.setCenter(center, 11);
				map.addControl(mapControl);
				map.addControl(new GLargeMapControl());
				map.enableScrollWheelZoom();
 
				main(map, center);
			}
 
			function showPoint(location, i)
			{
				var summary = '<h1 style="margin-bottom:0">'+location['title']+'</h1>' + '<p style="width: 300px;">'+location['summary']+'</p>';
 
 
		  		marker = new GMarker(location['location']);
				this.map.addOverlay(marker);
				marker.bindInfoWindowHtml(summary);
			}
 
			function main(map, center)
			{
				this.map = map;
				this.center = center;
 
				this.locations = [];
 
				<?php
				if(!empty($locations))
				{
					$n = 0;
					foreach($locations as $location)
					{
						echo "this.locations[$n] = [];\n";
						echo "this.locations[$n]['title'] = '".$location['title']."';\n";
 
						echo "this.locations[$n]['location'] = new GLatLng(".$location['longitude'].", ".$location['latitude'].");\n";
 
						echo "this.locations[$n]['summary'] = '".$location['summary']."';\n";
						$n++;
					}
				}
				?>
 
				for (var i = 0; i < locations.length; ++i)
				{
					showPoint(locations[i], i);
				}
			}
		</script>
	</head>
 
	<body onload="initialize()" onunload="GUnload()">
		<h1>Parties!</h1>
		<div id="map_canvas" style="width: 500px; height: 300px;"></div>
	</body>
</html>

Reload, et voila.

The initialize function is the same here as before. However this time in the main function we populate a javascript array from our php array, which we then loop through and pass to our showPoint function which shows the point on the map along with our title and description, which will show up when the marker is clicked.

Wrap Up!

Well that turned out a lot longer than I expected but hopefully it covers all you need to know about getting started with codeigniter + google maps api. Of course if I were creating a real site there’re several things I’d do differently, which for length reasons weren’t discussed here. I’ve listed some here:

• I would make it less ugly
• I would split up the top and the bottom of both the views into a header and footer view, and then load them so they surround our main content on every page. This means if you want to change the layout of your site you don’t have to update every view
• I would put some of the javascript into it’s own file, for the same reasons.
• I would put the api key into a config file and load it into our view, again if we need to change our api key, we don’t have to change every file
• I would autoload commonly used models, libraries and helpers

You can download the code here.

Enjoy, Ben.

P.S. If you liked this tutorial please leave a comment!

As a software developer if I had spent hours and hours (and hours) developing an application to support myself (or a family) and I put it up for sale, how is it ok for someone to go onto bittorrent or limewire and get it for free. I can imagine that film makers, authors and artists feel the same.

Would you ever hire a builder to build a house then not pay him? No. So why is it different with software, music and movies? Would you steal a book from a shop? No.

Right now I sound like the anti-piracy ads:

However I do think it’s becoming a growing problem and that people do it just because it’s easy and that the internet generation are going to cause a revolution in how a product is financed since we/they are starting to expect to get everything for free.

When to make commits

There’re several cases when I make a commit, the more often the better in my opinion.

• When I’m finished implementing a certain feature.
• When I have achieved a particular milestone in implementing a feature
• Before I fix a bug
• When I have fixed a bug
• Before I make major changes to whatever I’m working on
• Before starting work on a new feature
• And always at the end of the day

So basically, whenever I can.

Commit Messages, Commit Messages, Commit Messages

This is all however completely useless (in my opinion) unless you provide commit messages with every commit stating what you have changed. Even if it’s a one man team, this is essential for when you need to go back later and find out what you broke, or why something is done how it is.

Registration now working, I think. The row is inserted into the user table and the number of remaneing uses on the invite code is being reduced.

Unideally the invite model is being loaded in the construct of the User model. This is the only way it would work, and is loading unnesecary stuff. So that needs attention.

Also still to do in registration.php is the setting of flashdata and redirection to the login page. the flashdata tells the login controller that this is the first login. An email also needs to be sent confirming the email address.

Example commit from hippstr

(I thought it would be more truthful to leave the spelling mistakes in, so don’t correct me on them). As you can see this commit message outlines everything that is changed in that commit and even outlines what is still to be done (which I don’t believe is 100% necessary). I do this because all Hippstr commits go into the basecamp project management system (see How I Manage Projects) so it lets the team know what’s happening.

Working with teams

When working in a team, subversion is key as it handles conflicts when two people edit the same file at the same time. I’m not going to go into details here but you can refer to the subversion book.

I also use Beanstalk for my subversion hosting. Sure you can setup a subversion server on any beige box or even some shared hosts. However I use Beanstalk as it provides fantastic integration with other services, such as Basecamp and Twitter.

I highly recommend it, it’s reasonably priced as well.

Software

I’ve played with versions, coda and other subversion guis but I always come back to the command line interface to subversion. I find it lets me get things done quicker and easier than anything else. There are only a few essential commands you need to learn.

• commit
• add
• delete
• copy
• move
• merge

Learn the above svn commands and you’ll soon be flying with subversion. However if this is too daunting I highly recommend Versions.

In conclusion

Working subversion into your workflow can often be difficult, however once you get into a habit of committing when ever you change anything even slightly significant, you will never go back. Plus don’t forget to add in those commit messages!

I hope this was at least some help to those struggling to get started with subversion.

Stikked is my ‘other’ website and this weekend under went some big updates. Stikked is a pastebin, a place for developers to collaboratively solve coding problems. Most often used to paste code in a chat environment where it is considered bad etiquette to paste code directly into the chat, instead you paste it onto a pastebin site, then paste the link into the chat. This weekend paste replies was added, a long time missing feature. Allowing you to reply to another paste for example to correct a problem in the original paste. Paste downloading was also added, still a little buggy though.

Now, if you know what a pastebin is… You’re saying why yet another pastebin? Well firstly pastebins are in general u-g-l-y with exceptions (http://pastie.org/). So I think I fixed that. Pastebins are also kinda feature lacking, sure it’s a pastebin, but I felt you could do a lot more. So Stikked has a few nice features like snip-url creation, auto-copying the link to your clipboard, remembering settings, etc. The really cool stuff is yet to come.

The final reason I made Stikked is: all open-source pastebins were awful and there was no decent open-source CodeIgniter (CodeIgniter being the framework which powers stikked) application from which others could learn. So Stikked is 100% open-source available from Google Code included with developer documentation to help you understand how everything works and get you started with CodeIgniter. On The Stikked Blog we will also be doing a series of tutorials on how we built stikked.

So check it out at www.stikked.com!

TextMate

Any of you who know me, know I never shut up about TextMate. Why? There’s nothing like it. No other editor is able to offer so many features yet stay so easy-to-use and lightweight. The power of TextMate is in the bundles. TextMate ships with bundles and language definitions for every language you could think of.

These bundles provide methods for inserting snippets of codes, tidying your code, posting it to pastebins and lots more. This is made better by the fact that it’s so easy to write a bundle, you can write a bundle in pretty much any scripting language provided with os x.

The TextMate sidebar is also one of my favourite features, making it so easy to work with projects. Especially web frameworks like CodeIgniter, rails or django. I reccomend you try TextMate today, it is availible from Macromates for 48 Euro.

Coda

text editor + file transfer + svn + css + terminal + books + more = whoah.
Panic

Coda is Panic Software’s beautifully-simple-feature-packed-all-in-one-web-development-tool. And I love it. Editor wise nothing beats TextMate, however I often find myself using coda for web development as the all in one nature makes it so easy, with a built in text editor, ftp client, subversion, css editor, terminal, books and so on.

Panic just last week released Coda 1.5, this is what got me to start using coda again. The subversion integration alone is outstanding. It has completely changed the way I work, being able to commit changes in a mouse click.

It has completely changed the way I work

In Coda 1.5 they also added custom books, allowing you to have access to any manual from your text editor. If I was to list every feature of Coda this post would never end, so check it out for yourself at panic.com, you can get the free trial or buy it for $99 (It’s worth it).

Yojimbo

Your effortless, reliable information organizer for OS X
Bare Bones Software

Those who know me also know I never shutup about Yojimbo. Yojimbo is an application for storing stuff. That’s right storing stuff. Things you don’t know where to put. I use it to store passwords, serial numbers, articles, ideas, photos, etc.

You can add an item to yojimbo from anywhere in the os using the quick input dialog. You can assign tags, give it a title and you’re done. You know it’s safe in the Yojimbo database. If you have some confidential information you want to store you can use Yojimbo’s built in encryption to encrypt it.

The usefulness of this application is endless, especially if you have lots of info you don’t know where to put. The one thing missing is syncing with iphone/touch, I wouldn’t be surprised if this was added in future versions though. Get it today from Bare Bones software for $39.

Things

I was first introduced to Things through their iPhone/iPod Touch application under the same name. Thing is an outstanding to-do list application. With a syncing mobile iPhone/iPod Touch application. Since the GTD craze first started in the Mac Community, I’ve tried countless solutions.

Things was the first to work for me. It’s not overly strict about putting this here and that there. You can input from anywhere in the os using the quick input dialog (pictured above). The interface both on the mobile version and the desktop version are beautiful.

I’ve found this application endlessly helpful in helping me get organised. It’s currently still in public beta so download it for free from Cultured Code, it will cost 50$ at launch.

Versions

Last but certainly not least is Versions. Versions is a stand-alone gui subversion client. It has in my opinion revitalised subversion, well on os x anyway. It’s beautiful easy-to-use interface combined with it’s extensive feature set is what makes this application great.

There’s not much more to say about it, but if you’re a developer and you use subversion you need to check this app out. It’s currently free as it’s still in beta. Get it today from Versions App.com.

I’m assuming you have basic developer tools installed and have a good enough concept of the command line + *nix. Start by installing subversion.

# apt-get install subversion

This gives us all the subversion admin and client tools. We’ll start by creating a central location for our repositories under /home/svn.

# mkdir /home/svn
# cd /home/svn

Now lets create a repository, replace myrepo with the name of your repository.

# svnadmin create myrepo

If you do an ls in this directory you’ll see subversion created a new directory called “myrepo”. We’re going to setup http access to our repository, so we need to ensure apache can read and write the files.

# cd /home
# chown -R www-data:www-data svn
# chmod -R 770 svn

We need to create some users to login to our subversion server, after running each htpasswd command it will prompt for a password which will be the login password for the user.

# cd /home/svn
# htpasswd -c .htpasswd firstuser
# htpasswd .htpasswd seconduser
# htpasswd .htpasswd thirduser

Now we have our repository setup we need to connect apache to it. I’m assuming you have apache setup and installed. Next we need to install apache subversion bindings and enable webdav.

# apt-get install libapache2-svn
# cd /etc/apache2
# a2enmod dav

Finally we need to configure apache.

# cd /etc/apache2/sites-available
# nano svn.mydomain.com

This will create a new file called svn.mydomain.com and open a blank file in a text editor. Copy the below into it and replace yourdomain with the correct value wherever mentioned.

<virtualHost *:80>
 
  	# Admin email, Server Name (domain name) and any aliases
  	ServerAdmin admin@yourdomain.com
  	ServerName  svn.yourdomain.com
 
	<location />
		AuthType Basic
		AuthUserFile /home/svn/.htpasswd
		AuthName "Dev Team Only"
		require valid-user
		DAV svn
		SVNPath /home/svn/hippstr/
	</location>
 
</virtualHost>

Now we need to enable and restart apache.

# cd /home/apache2
# a2ensite svn.yourdomain.com
# apache2ctl graceful

If everything went well we should now be able to access are subversion server at svn.yourdomain.com. But we need a subversion client on our desktop machine to access the repository.

Subversion clients

My favourite gui subversion client is currently an os x app called Versions. Which I’ll talk a little about below but if you’re not on os x I would recommend either Tortoise SVN on windows or the standard subversion command line client on any os. Available here.

You also need to learn how to use subversion by reading the free subversion book, there’s a small learning curve to understanding subversion and integrating it into your workflow but you’ll get used to it eventually!

To setup subversion in versions click “add new repository bookmark” and enter the following information. The user and password are what you set above when running the htpasswd commands.

There’s a small learning curve now to understanding subversion, as mentioned above and it’s all explained better than I could ever explain it in the subversion book mentioned above. Good luck! Ask any questions you still have in the comments.

“Basecamp is the smarter, easier, more elegant way to collaborate on your internal and client projects.”
37signals

Basecamp is a really simple and easy-to-use project management system. With great to-do management, a forum like “messages” section and milestones (There’s more but this is the important sutff). To-dos are with out a doubt the most useful aspect of Basecamp. They allow me to assign a task to any individual. When this task is complete it can be checked off. This will show up in the project summary page also (pictured below).

I use messages mostly for brainstorming. Where I post an idea which is blasted out to all my team via email (They can reply to the message by replying to the email). Why is this useful? When you’re working with a global team or even a local one, that are not all around at the same time, this is an indisposable tool. Reducing the amount of required meetings (which are a pain to schedule across timezones) and including all the team without the use of a messy mailing list solution.

“Basecamp is the iPod of project collaboration.”
Joshua Peterson, CEO, 43things

Both messages and to-dos can all be associated with a particular milestone. For example at the moment all hippstr to-do’s are associated with the private beta 1 milestone. Meaning we’re all working towards a set target, which is important.

In summary basecamp allows all the team to stay in touch, keep up to date with everything that’s happening and track progress (which is key and a moral booster on those bad days). I’ve used it with just myself to teams of 20 and it has always been a success. Basecamp can be described in one word, “phenomenal”. Sign up for free today at http://basecamphq.com.

Subversion. Subversion (aka. svn) is a version control system. In english, subversion keeps track of all the changes made to a code base (It can be used with anything though, word files, pictures, text files, etc. I just happen to use it for code). Allowing us to revert to a previous version when we break something, instead of wasting hours trying to find that thing we changed which broke everything.

Subversion also doubles up as a backup solution as it’s hosted on a remote server. Meaning if my house burns down all my work isn’t lost (although I think I have bigger problems if my house burns down). Subversion’s usefulness is greatly increased when working in teams as it manages conflicts. If two people change the same file, instead of one set of changes being lost, subversion presents you with a choice to merge the changes or choose which you would like to keep.

Previously I would’ve recommended trac in conjunction with subversion, but really it’s completely useless and difficult to use. I’m yet to find a bug tracking + svn interface system I like. Nevertheless subversion is an indispensable tool, get started today at http://subversion.tigris.org/. Soon there will be an article on how to configure subversion.

Coming up in part 2. I will talk about campfire, getting real, practices and what I don’t use.