curl http://pear.php.net/go-pear > go-pear.php
sudo php -q go-pear.php

Press enter to use default until you get to the file layout and edit the first to be /usr/local

1. Installation prefix ($prefix) : /usr/local
2. Temporary files directory     : $prefix/temp
3. Binaries directory            : $prefix/bin
4. PHP code directory ($php_dir) : $prefix/PEAR
5. Documentation base directory  : $php_dir/docs
6. Data base directory           : $php_dir/data
7. Tests base directory          : $php_dir/tests

Press Enter to continue and Y to install

PEAR_Frontend_Web-beta, PEAR_Frontend_Gtk2, and MDB2

Hit Y to let PEAR update your include path

That’s it! You should be able to type in pear version to see PEAR installed.

yum install gettext-devel expat-devel curl-devel zlib-devel openssl-devel

Then get the source code for Git

wget http://kernel.org/pub/software/scm/git/git-1.7.2.2.tar.gz

Extract the tarball

tar -zvxf git-1.7.2.2.tar.gz

Install to usr/local

cd git-1.7.2.2.tar.gz
make prefix=/usr/local all
make prefix=/usr/local install

Test it out

cd ~/
git --version

• Understand how git actually works and what resources it uses
• Learn more advanced techniques when you do it yourself
• Save money (github and most other services charge if you want private repos)

It took me about a week to fine-tune and get everything setup the way I wanted, but the end result is:

• A privately managed git repo that looks and functions like Google Code
• Custom scripts which update my live site when the live branch is update
• My own agile development setup which allows me to manage users quickly and easily
• A visualizer that makes git easy to understand at any level

Step 1: Install Git

This step is pretty easy if you google it.

Step 2: Install Indefero

If you go to their website you can download their open source project and set it up using their instructions. Feel free to ask here if you need help with this step.

Step 3: Test your Git Repo

Setup a new project in Indefero and grab their code to clone the repo. You may need to setup a custom user with access to the indefero/ directory. For me, I created a user titled git and gave the user shell access to my /var/local/indefero directory.

Once you’ve setup the repo, clone it to your local machine drop in a test file and push to make sure indefero and git are working correctly.

Step 4: Setup Post Update Hooks

Git provides hooks for running shell scripts on predefined events. I used the post-update hook on my live repo to use my webroot (where my physical site resides) for the worktree. This allows me to track changes to the live site (which makes roll-backs or recovery a breeze) as well as run any other commands such as a mysql dump everytime the live site is updated. Here’s what you do to set this up:

Navigate to indefero/git/repositories/YOUR_WEBSITE/ in terminal, then create a script named post-update:

nano hooks/post-update

Put this into your script, along with any other commands you’d like:

#!/bin/sh
git checkout -f

Finally, edit your config file using git config or nano config:

[core]
        repositoryformatversion = 0
        filemode = true
        bare = false
        worktree = /path/to/YOUR_WEBSITE
[receive]
        denycurrentbranch = ignore

This will tell git that this repo uses your custom directory as it’s worktree (where the physical files are located) and the post-update hook does a checkout everytime you update your repo.

Hopefully it helps, if you need more hit the comment form. I will most likely come back and expand on this post with more detailed instructions if I get enough feedback or requests…

We’re quick to turnover when we really believe in something. One of the automated scripts we wanted for our company was a system to upload mockups to a directory and have them automatically publish in a web page. This system allows us to simply drag and drop images and have them populate with different versions and folders. We also wanted this functionality to be password protected and to have a way to separate public and private folders in a directory.

For security purposes I have removed the actual session control script that we use (which is a way of protecting pages via login). However, it does come standard with a simple login posting ability so even if you know nothing about programming, this script should come in handy.

Features:

• JQuery Slider for form and functionality
• PHP to read directories and manage users
• Minimal design and small pages.
• Basic frame and structure, ideal for building upon or changing functionality

Documentation:

Under the parent mockups directory

• This is the area to store your CSS, JS, and Layout Images
• The index on this page reads the directory and displays any directory starting with an _underscore
• The user name should also be the directory name that user has access to
• User names and passwords are stored in an array

Under the mockups > _user directory

• This once again reads the directory looking for folders starting with an _underscore
• This step is not necessary for the application, we simply use it for different versions/revisions

Under the mockups > _user > _version directory

• The index page in this directory reads the directory and looks for any images (you can change extensions that it looks for).
• Once it finds the images it outputs them in HTML and displays the file name, sorted alphabetically.

Check it out

Starting a Website: The Right Domain Name

It all starts with the right domain name. A good domain name is easy to spell and remember, and lets visitors know exactly what they are in for. Remember that a misspelling may take your potential traffic to a completely different site, maybe even a competitor’s site, so ease of spelling and brevity are essential. WeveGotaStupendousArrayofHinges.com is a poor domain name. Hinges.com is an ideal one.

Starting a Website: Hosting

Before you can set up your website, you need a place to put it. The process of providing a server that can get your pages on the web is called web hosting, and there are number of companies that provide this service. You’ll also find that there are a wide range of prices, depending on what you are looking for. To start, you should look at companies that offer “shared hosting,” “budget hosting” or “discount hosting.” These phrases mean that the hosting companies rent out different parts of their servers to different people or companies. Although you will share a control panel, IP address and technical support with these others, they will have no access to your website. This is a very economical option, so it’s a good one to start out with.

Starting a Website: Design

Of course, a great website needs a great design. You can design the website all by yourself, using program such as Photoshop, Illustrator or Fireworks for design and coding using DreamWeaver or similar. If you are not happy with the results, then there are plenty of web designers out there whom you can contract to help you design your website at a reasonable price. Remember, your website is your “face” in the Internet world, so you want it to be a good one. A good design is key to a successful website.

When I downloaded Safari 5 today, I noticed that apple finally added support for extensions into the browser! Safari has always been a nice app to use on mac, and I love the inspector, but lack of safari extensions for things such as grid tools, color selectors, etc have made me reluctant to switch. However, Safari also made it far easier than many other browsers to create extensions using straight up HTML5, CSS and JS! So in about an hour, I created this little extension to overlay a grid (like 960 or Blueprint) on the page. Enjoy!

Check out the RFP Application in action

Open Graph opens doors

I’ve long been a supporter and developer of microformats (formats that aim to set a standard or rule for data formatting) and RDF standards for the web, but my biggest frustration has been with the lack of adoption by other technologies. To date the only microformats that have any real weight are hCard (much like vCard, which is a data format that sets a standard for contact information) and hCal (data format that sets a standard for calendars), which allow data to be formatted with XML markup to define objects and properties. After being set as a wanted standard by the development community almost 5 years ago, it’s taken many apps up till this last year to adopt them. Now we finally see these formats used for apps like Microsoft Outlook, Mac Mail, iCal, etc.

Facebook’s claims of Open Graph imply compliance to make RDF standards and remain semantic, as well as adopt new standards. I was a bit sceptical about how open this format would really be. Facebook has long been a giant that’s bullied us developers to learn new markup and programming techniques with their proprietary frameworks like FBML, FQL, etc. So immediately following the announcement of Open Graph, I tracked down and talked with a few Facebook engineers to confirm or deny how open this standard would be. They made no claims of anything concrete, but told me that Facebook will be supportive and quick to adopt any standards demanded by the dev community. This means groups like microformats.org may be able to define how we adapt new data standards. I believe this is a HUGE step in the right direction and will allow companies like Yelp, Amazon, Pandora, etc to put new standards into practice that Facebook could quickly swallow. The facebook developers also mentioned how Facebook is beginning to open source their server side technologies and frameworks such as Hip Hop and XHP which further proves that facebook is finally giving back to the community.

An example of a new Open Graph Object

Pandora could lead the dev community by creating their own Graph Object called ‘audiotrack’ which is used to provide data about a specific track. The audiotrack object could look much like this:

{
	"type": "audiotrack"
	"name": "Without Me - Eminem",
	"metadata": {
		"trackinfo": {
			"title":	"Without Me",
			"artist":	"Eminem",
			"album":	"Curtain Call - The Hits"
		}
		"trackdata": {
			"kind":		"MPEG",
			"size":		"6.3 MB",
			"bitrate":	"179 kbps",
			"samplerate":	"44.100 khz",
			"released":	"14400",
			"formato:	"MPEG-1, Layer 3",
			"channels":	"Joint Stereo",
			"id3tag":	"v2.3",
			"encodedby":	"LAME 3.90"
		}
	}
}

This format could then be submitted to a site like microformats.org and refined by the development community. This allows developers and industry leaders to work together to make new standards that we can all enjoy.

Moving Forward

I know that I personally, am going to get back in the game to promote microformats. I really believe that Facebook could help move technology forward as a whole, but not without the help and acceptance of the community.

What do you think about Open Graph, and do you think it will be widely adopted?

Search Engines are Secondary?

This may seem controversial, but I genuinely believe that Facebook is going to dominate search with social. Think about the way you interact with a search engine. You type in a string of keywords, hit go and begin skimming the headlines which by default are fetched based upon relevancy and location (search engines can’t grab your exact location, but they can get pretty close based on IP). The problem with this, is that while search engines have very complex and amazing algorithms to compute relevancy, they can’t access personal data about the user other than the ISP and browser data. Facebook however, has a robust amount of information on each user, as well as the users’ network and friends. Facebook could easily adopt or merge with another company like Yahoo and have all the services that competitors like google search and bing have, but they can add in a new layer of complexity on top.. they can aggregate the user’s content and content among friends and provide results very specific to the user while defaulting to standard results as a backup. Let’s explore a hypothetical so that it’s easier to understand:

There is odd weather as you’re driving through Colorado. There’s been heavy hail and rain and you’re worried about the possibility of a tornado. You pull out your phone and do a quick search…

Typical search results via Google:

When you initialize this search, Google runs its algorithm on already collected data, and shows you content relative based on credibility of the site, meta data, and content. So maybe a site dedicated to tornado photography is on the top of the list, because it has hundreds of pages featuring content about Tornadoes in Colorado. Google is trying to solve this with “recent news” feed that pings other media/news sites for data that’s more current, but it’s not very robust or accurate, especially about news that isn’t popular or published much on the web. Real time results get populated fastest in Social not Search.

Facebook “hypothetical” Results:

Remember this is hypothetical, but if Facebook chooses to, this could very quickly become real. When you initialize the facebook search for the same string tornado in CO it could first aggregate across data from you and your network. Facebook would look for friends located in Colorado or part of a network/group based in CO. It could then rank your friends in order of relevancy by location and finally pull in data related to tornadoes that they user has posted. This is far more valuable than the way search engines currently work because it adds another layer of data which is in many cases far more relevant to you.

What do you think? Do you agree that Facebook may stand as a direct competitor to companies like Google?

There are tons of wordpress twitter plugins, but none of them solved a problem I was facing. I wanted to display a feed from multiple accounts, but rather than showing them in order (some users tweet more than others) I instead wanted to show 1 from each account. So I threw together my own plugin to do so. Feel free to download and submit feature requests. Get it at WordPress.org.