Before I voice my opinion on the matter lets look at the facts. First, neither side is developing the same type of sites. From the audience to the message to the types of content the only two things are similar, the desire to sell a product and the basic media that is used to sell that product. Beyond these two similarities the two worlds are really very far apart.

So how different are they? Well, lets break down the vast majority of what each camp does. First there is the private sector developers. Often they’re working for small firms (less than 100 employees) and make their living based on the quantity of work they turn out. That’s not to say they don’t do quality work, on the contrary some of the most cutting edge design comes from the private sector. However they are concerned with providing an often singular message armed with only an intermediate level knowledge (sometimes less) of the sender of the message they are trying to convey. Whether building a site for a small mom-and-pop or the largest of the fortune 500 by nature they cannot by design know the finer details of the product their message is trying to convey. Because of this their expertise tends to rely on the message itself and while effective it usually does not provide much real in depth information about the sender of the message. This works great for individual products, small companies, and many other situations where background information isn’t important however for higher ed this model simply doesn’t work.

Higher-ed is a different beast. The mission of the higher-ed developer often isn’t so much the medium in which the message is sent, but is in fact the message itself. Cutting edge graphics must be replaced by content that meets various accessibility laws, brand new photos must be passed through various levels of bureaucracy before they can be released to the public, and, worst of all, the idea of academic freedom means there are thousands of individuals providing content for and demanding their equal place throughout the site. This isn’t to say there isn’t a message to be conveyed. On the contrary higher ed has at least 3 messages to convey. First and firmost a good higher ed site must recruit students. Second, a higher ed site must engage alumni in an effort to generate donations. Finally, higher ed websites must serve a plethora of audiences with numerous tasks from test taking to class registration to housing information all from the umbrella of a single site. To due all this successfully a web department is not only nice, but required to both make sure the messages are conveyed successfully and to ensure that that the college’s brand is not lost in the numerous agendas and missions that make up the average college web presence. This is often done not with cutting edge graphic and design technologies, but instead with cutting edge code and algorithms that can handle the raw data that comprises the college’s message. From the beginning of the internet colleges and universities have worked to advance web technologies from web browsers to cloud computing to semantic web.

So who is right? Well, they both are. In the case of the webmaster title it is true that most of the private sector, particularly design firms, have turned away from it’s use and consider it antiquated. That doesn’t mean it’s wrong however. In both higher ed and other organizations the official title of webmaster has been built into the very system that must employ the people in question. In many cases this is the result of a lot of work in systems to which change isn’t on the daily menu. As a consequence some very capable web professionals still use and will continue to use the webmaster title. While the private sector my shun the word for whatever buzzword is popular at the moment, the public sector and in particular higher ed, will continue to embrace it for the foreseeable future.

Now for the second point. Who does or doesn’t have any business developing web sites? The answer is they both are necessary for the web in their industries. Private sector developers private thousands of site for brands that would otherwise not be on the web and continue to advance the technologies that present these messages on the web. On the other hand, higher ed requires their web developers for both providing the numerous applications that can make up their sites and to help provide the content that makes up the average collegiate website. For even those college that outsource the design of their site still need people who understand the web and current web technologies and can use these to merge the numerous messages with the most current and accessible technologies.

In the end, the web has room for both private sector designers and higher ed developers. Who knows, they might even learn a thing or two from each other.

OK, so it has a few things last year’s model didn’t. Does that really make it worth disposing of a perfectly good phone? Will any of these new features do anything to really help anyone’s daily life? To both I have to say it’s doubtful. I myself will once again avoid the mass hysteria. Even as an OSX fan I just can’t justify a switch to iPhone and here’s why.

1. AT&T
   This alone accounts for numerous reasons why I won’t touch the product. They don’t have nearly the coverage of Sprint, and I have yet to meet a single person around here who has ever had anything to say about their service other than “it’s ok.” In addition, I pay $35/month/phone for unlimited everything on Sprint and I can use it everywhere I go. Why would I want to pay more for a data cap?
2. The Antenna
   Does this really even need to be addressed?
3. Android
   With Android ever increasing it’s market and market share iPhone really is yesterday’s news. Not only have I been able to find thousands of apps I don’t really need for my HTC Hero, I’ve been to install custom ROMs and do all sorts of tricks iPhone users only dream about. In addition, Android Market is growing so fast that it quiet possible will soon be bigger than Apple’s App Store anyway and doesn’t enforce ridiculous restrictions on it’s developers.
4. Phone Cost
   I paid $99 for my HTC Hero. Why would I want to drop double that for a phone that will be laughed at in a year?
5. Google Apps
   I use Google Apps both at work and for personal use. The integration of Email/Calendar/Tasks/etc with Android is something that iPhone can’t currently match and doesn’t seem to have on the drawing board either. Google definitely has a leg up by already offering the services most people are using their smartphones to try to connect anyway.

Apple makes some good products and I in fact own 4 iPods and a Macbook Pro. These past experiences however simply don’t overcome the downsides of both iOS and AT&T. Now should Sprint ever offer iPhone and Apple perhaps start allowing people to actually develop what they want I might be convinced to change, but for now I really don’t see a point in paying extra for an inferior service and closed product.

Initially the idea was to update the homemade CMS we were using to make better use of technologies such as RSS and mod-rewrite. As the core of the code was 12 years old it seemed like it should be a long, but relatively straightforward project. In addition, we hoped to streamline the site content to better serve prospective and current students by eliminating obsolete features and re-writing a lot of the content to both bring it up to date and to address the needs we’ve seen arise over the last couple of years.

Like so many simple ideas this one quickly grew to an entirely new monster.

First, I decided that the department would be better served by established CMS products than by my own code. This would serve two purposes. First, it would allow me to concentrate on content rather than the functionality of the CMS. Second, it would allow for easier maintenance should I leave here (which we plan to do in a few years at most). I didn’t want something that would require constant coding and bandaids by someone whom I wasn’t familiar with the code. The only way I could see to overcome this was to use existing software.

Next, while concentrating on content, two things became very obvious. First, we were trying to do too much with one site. Our homepage was trying to be everything to everyone and winding up being less then useful to all. Second, we have been selling ourselves simply as Aviation for years. The catch is we, as flight and management, are only one piece of the total Aviation offerings. There is also an aviation technologies department which had absolutely no presence on the aviation website. This division was bad on many levels for both departments.

Finally, up until this point we have been operating on a rather obsolete mindset. Like many groups our administration viewed the website as a nice bulletin board but not much more. While students were searching Google for where they want to go to college our budget was being spent on ads in magazines that barely touch our target audience as well as give-aways and other programs geared toward a sometimes apathetic alumni base. Nothing was going to attract new blood in places where that new blood was lurking.

To address these problems required a lot more legwork than originally planned on. First, we decided to replace our site with not one single site, but two sites and a dynamic system capable of providing sites quickly for almost any aspect of our community. Next, we tackled a long standing political feud to combine the web presence of two departments into a single online presence. Finally, we spent almost 4 months on initial training and, as my colleagues call it, brain washing our people into working with us online in providing content and working to increase our presence in many areas where our competition already had a great lead.

Today our online presence is divided into 3 pieces. Our mane website, www.aviation.siuc.edu, is a recruiting and retention tool designed to provide information to students, their families, and our alumni. Next, for all our internal needs we built a wiki to allow for quick collaboration and easy editing of volumes of information that only really pertains to our faculty and staff. Finally, to provide a quick, easy, and uniform publishing platform for all of our community we created blogs.aviation.siuc.edu which allows any individual or group in the SIUC Aviation community to quickly and easily build a website specific to their own needs while still maintaining a level of control of both security and brand at the department level.

As each of these sites is designed to fulfill a different goal it follows that they each then use different software to accomplish those goals. To do this we picked Drupal for our main website, MediaWiki for our intranet/wiki, and WordPress MU(Now WordPress 3 with multisite) for the “Blogs and sites” system. Together these software packages have allowed us to expand our online presence in a manner that makes sense for everyone while preserving all that data that previous was smashed together on a single site. In addition, through the use of such common products, we have been able to train a host of users who maintain content on their own sites or projects throughout all 3 sites.

Once we had the software picked and installed and the initial users trained the cultural shift in our marketing began to take shape. For the first time we saw some of our marketing budget going to online resources instead of dying magazines and newspapers. Data from sources such as Google Analytics was finally views as something other than a bunch of lines on a piece of paper.

In the end, I’m happy to report that we have come up with a not just a website, but a solution that brings the department into the modern era. While there is still and always will be work to be done it is nice to see that we have finally shifted into drive and are proceeding ahead.

1. I downgraded PHP to 5.2 using this procedure. While this might not be necessary I cannot say for certain that the rest will work with PHP 5.3
2. Install the required packages
   sudo apt-get install apache2-suexec-common libapache2-mod-fcgid php5-cgi
3. Enable the necessary modules
   sudo a2enmod fcgid suexec alias
4. Disable the old php module
   sudo a2dismod php5
5. Change the suexec configuration file to correspond to your sites
   cd /etc/apache2/suexec
sudo nano www-data
   change the first line to the root of your sites (i.e. /home)
6. navigate to the user’s home directory and create the php-fastcgi directory
   cd /home/[username]
sudo -u [username] mkdir php-fastcgi
7. Create the wrapper file
   cd php-fastcgi
sudo -u [username] nano wrapper
   Enter the following lines:
   #!/bin/sh
#PHPRC="/usr/local/etc"
#export PHPRC
#PHP_FCGI_CHILDREN=8
#export PHP_FCGI_CHILDREN
#PHP_FCGI_MAX_REQUESTS=5000
#export PHP_FCGI_MAX_REQUESTS
exec /usr/lib/cgi-bin/php5
8. Make sure the file is executable
   sudo chmod +x wrapper
9. Edit the sites’ configuration file
   cd /etc/apache2/sites-available
sudo nano [sitefile]
   Add the following lines
   SuexecUserGroup [username] [username]

   ScriptAlias /php-fastcgi/ /home/[username]/php-fastcgi/
   FCGIWrapper /home/[username]/php-fastcgi/wrapper .php
   AddHandler fcgid-script .php
   Options ExecCGI Indexes

10. Restart Apache
    sudo /etc/init.d/apache2 restart

This should take care of it. You should now have fastcgi and suexec processing your php files for both better speed and better security than the alternatives. In addition, you can repeat this process for any sites you have located on the server.

1. mod-php
   Fast, but all scripts execute under the standard apache user creating problems with file permissions in applications such as WordPress. The fact that all scripts, regardless of user, are on the same user can also lead to security problems.
2. mod-cgi with suphp
   This is slower that mod-php because the php threads have to be restarted preventing cache and creating extra overhead. It does allow for better security due to allowing all php scripts to execute under the user that created them.
3. fast-cgi with suexec
   This is the best of both worlds with the speed of mod-php and the security of suphp. The catch is the setup.

Now judging by the title of this thread I think we all know where I’m trying to go. Currently we’re running suphp. It works well, but as traffic to the server increases I would like to increase it’s performance. Although there are numerous tutorials out there, getting them to work on Ubuntu server isn’t easy.

I have numerous virtual-hosts all based in /home/[username]/public_html, a rather common setup. I’ve installed apache2-suexec-custom on the ubuntu box and pointed it to /home to overcome the home directory problem in the standard apache2/suexec package.

Now the problem is I can’t get any further than this. I got fastcgi to work by adding

FCGIWrapper /usr/lib/cgi-bin/php5 .php
AddHandler fcgid-script .php
Options ExecCGI Indexes

But no matter what I try I cannot get it to work with suexec. Any suggestions? I’m really pulling out what hair I have left on this one!

Here’s how I did it:

1. Remove PHP 5.3sudo aptitude purge `dpkg -l | grep php| awk '{print $2}' |tr "\n" " "`
2. Create a directory in /etc/apt call sources.list.d (I do believe it may be there on some installs, but I wasn’t so lucky)
   cd /etc/apt
3. In /etc/apt/sources.list.d create a file called karmic.list.
   sudo mkdir sources.list.d
4. Create the karmic.list file and add the appropriate entries.
   sudo sed s/lucid/karmic/g /etc/apt/sources.list | sudo tee /etc/apt/sources.list.d/karmic.list

5. Next, create the folder /etc/apt/preferences.d
   cd /etc/apt
sudo mkdir preferences.d
6. In preferences.d create the file php and add the following: (note I’ve just included the file to download)
   1. Download php
7. Install PHP 5.2
   sudo apt-get clean
sudo apt-get update
sudo apt-get install libapache2-mod-php5
8. Install any additional php packages you might need (php5-cli, gd, etc)

I’ve compiled this method from a couple of sources and successfully implemented it on our servers. If you’re running a number of php packages I would recommend running sudo dpkg -l | grep php| awk '{print $2}' first to build a list of the extra packages you’ll need.

What’s your cutoff? How do you balance the stress and time of side jobs?

Thankfully, the fix was a lot easier than a new host. I installed the WP Super Cache plugin and for the last few weeks I haven’t been throttled for more than a few seconds a day. What a lifesaver!

or

Although these may sound like trivial items to many groups, unfortunately the powers that be where I work have not, until recently, been convinced of the power of online marketing and started to take it seriously. Now in many ways this shift does make my work a little harder with added oversight and more pressure, but it is a challenge I have looked forward to for years.

It has been a good day!

You can download the plugin here replace-target.zip

For the latest download, please check out my project page at http://www.chriswiegman.com/wordpress-replace-target-plugin/

My solution to this problem has been to use my own wp-signup.php page within my sign-up theme. Here’s how it’s done:

Before you start, make sure your server supports Apache mod-rewrite. An easy way to do this is to simply go to a post on one of your blogs. If you don’t see ?= anywhere in the address you should be OK. If not, go to http://www.yoursite.com/wp-admin/options-permalink.php and see if permalinks are available on your server. If they are not you will need to get this enabled by your server administrator before going any further.

Next, the most important part, open up the .htaccess file in the root of your WordPress MU installation in your text editor. You may need to create the file if you haven’t been using any redirects previously.

Find the line:

RewriteBase /

Right below it (before anything else) add the line

RewriteRule ^wp-signup\.php(.*)$ wp-content/themes/[your theme name]/wp-signup.php$1

Save the file.

Finally, copy (don’t move) wp-signup.php from the root of your WordPress MU installation to your theme directory (usually wp-content/themes/default). Make sure you leave the original file in place. Moving the file alters the WordPress MU core installation which is never a good practice.

Now you may edit the wp-signup.php file to your heart’s content. Redirects to the new file should be seamless to your users.

If you would like to see how I’ve used this take a look at http://blogs.aviation.siuc.edu/wp-signup.php. Now my needs were simple, we just wanted to clarify some of the text to reduce confusion for our users. There is a lot more that can be done to make this form yours.

* edited Feb 27th, 2010: added a couple of important lines *
Never write on a deadline when you don’t have to. I forgot to change 2 important line.

In your new wp-signup.php page (the one in your themes folder) find the following 2 lines:

require( dirname(__FILE__) . ‘/wp-load.php’ );
require( ‘wp-blog-header.php’ );

and replace them with the following:

require( ‘../../../wp-load.php’ );
require( ‘../../../wp-blog-header.php’ );

This will ensure that WordPress MU can find the files it needs to process your data.

In the web the idea of an expert is often a misnomer. It seems that to web people the expert title is often given freely to any person who knows more about a specific topic than you do. For example, I was once labeled a Joomla! expert after a single installation of the software without even launching a site on the platform. Why? because I had done it and the other person hadn’t I must be an expert.

On the flip side, after coding CMS systems for 11 years I was told the other day that I was a rookie as I didn’t code in this particular person’s language of choice. To this I had to say really? Simply because I don’t know your language I am a rookie at all things web? How strange.

A single install of Joomla! does not make you an expert in Joomla!. However, the ability to look at the code of Joomla! and be able to understand it and modify it could very well make you an expert in PHP. On the other hand, the ability to code a million line PHP application may make you an expert in your application, but if you can’t read and understand anything else you might never truly grasp the language and will never be recognized as an expert. This last example I see all the time as a CS student in code which is never written, but merely cut and pasted together to perform some function. After all, if you can’t explain or understand what you’ve done than you really haven’t done much at all.

So then, what really makes an expert. In my humble opinion and expert is someone who through their merits and experience is recognized by their peers as a leader in their field. Someone who can not only do the work, but actually understands what they are doing and can communicate that idea to others. So, web people, when throwing around titles such as expert make sure they’re going to a deserving soul and not simply someone who can follow a readme file. Anything else is just insulting to those who really have mastered their craft.

1. Akismet
   The one plugin that is now included with a base install of WordPress, Akismet is the ultimate in WordPress Spam protection. It creates a “Spam” box in your WordPress comments and has the unmatched distinction of being the only spam blocker in any app I use to not miss a bad comment.
2. WP Security Scan
   Every new WordPress installation has it’s problems. Whether it’s weak passwords, file permission problems, or some other glitch it doesn’t take much for an attacker to exploit an insecure WordPress installation. WP Security Scan will find the most pressing problems and offers instructions and scripts for repairing them before they bring down your site.
3. Login LockDown
   Another security plugin, login lockdown prevents brute-force attacks on your WordPress administration pages by locking any account after a specified number of bad login attempts. After using WordPress MU in a public University I can tell you first hand that Login Lockdown not only improves the security of your site, but can also help reduce server load by reducing the number of allowed bad login attempts.
4. Header and Footer
   Header and Footer is a handy little plugin allowing for inclusion of custom code in the header and footer of all your pages. Items such as Google Analytics code, Yahoo Site Explorer meta tags, and numerous others can be added, removed, and changed at ease with this simple little plugin.
5. Clean Options
   I’ve added Clean Options to the list somewhat reluctantly as it may be a little much for some novice users. However it is an incredibly handy tool to have on any blog or site that has been out there for more than a little while. As other plugins are updated, added, or removed many of their options stay in place in your database. Do this enough times and it can start to impact the performance of your MySQL database. Clean Options fixes this by removing unused and obsolete options from the database allowing for a much more streamlined options table.

Of course these aren’t the only good plugins out there, and in fact putting this list together wasn’t as easy as I originally thought. A few of the plugins that didn’t make the cut include: All In One SEO Pack, WordPress Mobile Pack, and WordPress.com Stats among others. Have something else that I missed? Let me know so I can give it a try.

Looking back there are a number of goals completed, a number still to go, and a few that I never even anticipated. In the case of our redesign we started in April, the main site is still to be done. This is due primarily to the uncertain outcome of our university’s web standards committee. We don’t want to over-invest in a product that will not meet whatever standards they decide to come up with. While waiting however I’ve had time to really refocus and expand our web presence while [hopefully] being able to provide a far more useful product to all of our customers. We’ve installed and launched WordPress MU for individual, group, and class sites. We’re nearing the launch of a MediaWiki installation to be used as our local intranet. Finally, we’ve greatly expanded our social media presence and finally convinced the powers that be that our marketing dollars are better spent on the web than limited magazines and other media.

Stepping away from the web I’ve made a lot of headway in a number of projects. We launched (and then decommissioned) a Squid proxy to restrict internet access during the day due to very limited bandwidth. It worked almost too well and was quickly removed for mostly political considerations. We’ve also begun the rollout of non Microsoft desktops to our faculty and staff in the form of Ubuntu as the primary OS. Only 3 desktops and our webserver have been migrated so far but considering that we had never looked away from MS before that it is definitely a good start. Finally, we added a windows update server to provide more control and better use of our bandwidth while still maintaining the ability to make sure all systems are completely patched.

Now on to the future. My first priority is, as always, our web presence. Pending the outcome of the campus web standards committee I hope to get a new main site in place very quickly. In addition, depending on current content contributions, I would like to see our new Wiki launch at the beginning of the Spring semester. It has been in the making for a little while now and promises to provide a much more complete and, more importantly, standardized method of intra-department communication. Beyond that I would like to see our file services migrated to Samba and Linux away from MS Active Directory as well as continue the gradual migration of departmental desktops to the Linux platform.

With all that I’m sure it will be another busy semester. However as with every semester I look forward to the challenges and can’t wait to see the outcomes!