Skip to content
Aug 31 10

Writers Block

by Chris Wiegman

So it seems I’ve hit a bit of writers block lately. I don’t know if this is a symptom of being overworked or if I’ve just hit a bit of a lull. Hopefully as the fall brings better weather and plenty to look forward to I’ll find a way around it.

Jul 18 10

Higher-Ed vs The Private Sector: It’s the mission stupid!

by Chris Wiegman

A short exchange on twitter the other night left me thinking. A comment from a friend who is a web designer in a private sector firm made two points: 1.) Anyone who uses the title “webmaster” doesn’t belong working on the web and, 2.) many places, including higher-ed, have no place running their own websites. As someone who works on the web for higher-ed I find these statements interesting, not because I believe them, not because they struck a personal nerve, but because I’ve heard similar comments about higher-ed from many private sector developers and I’ve heard the exact opposite about the private sector from a number of higher-ed developers.

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.

Jul 1 10

iPhone 4…yawn

by Chris Wiegman

With all the buzz around iPnone 4 one would be inclined to think that not only has the product been the single best phone to ever see the light of day, but it perhaps has also brought about world peace, stopped the oil leaking into the gulf and maybe even cured cancer. I can’t turn on my TV, open up my browser, or even look at a newspaper without some fan-boy proclaiming its wonders. The question I have to ask is why?

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.

Jun 30 10

After 15 Months The Redesign is Complete

by Chris Wiegman
SIUC-Aviation

Back in March of last year I started the redesign of the SIUC Aviation homepage. Right from the beginning I knew it would be a rather ambitious project I just slightly underestimated exactly how ambitious it would be.

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.

Jun 9 10

Running Apache+FastCGI+Suexec in Ubuntu 10.04 without /var/www

by Chris Wiegman

For the last few weeks getting fastcgi and suexec to run on apache without having our sites in /var/www has been something of my holy grail. I even went as far as writing a blog post asking for opinions on the situation here. I have finally made it work. Here’s how I did it.

  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.