'Rescuing' websites from themselves: another library skill?
Times have changed a lot since then, with object-oriented programming languages (such as PHP) rising in importance together with server-side innovations and content management solutions. Web programming is a full-scale specialty that requires a command of a number of programming languages, along with skills in visual display and desktop publishing. Individuals who possess this skill set are often referred to as "full stack web developers," and in the San Francisco Bay region, they are in demand.
That demand spans the high-paying tech-startup market and the more staid terrain of banking, business, nonprofits, and higher education. Most organizations need at least one fully skilled web developer at their disposal or a team of programmers in large-scale organizations. Outside contractors can take on this role too, provided that they are ready to step in and address emergencies.
My own perspective on web administration comes almost entirely from higher education, which colors my opinions about what best practices should be. One distinctive aspect of web programming in the education sector is the fact that the newest, best, speediest, and most rock-solid services to maintain online services may not be available due to cost constraints. We do the best we can, which keeps some of the spirit of the artisanal web alive and kicking. It also means that we frequently need to merge diverse web-based resources together in idiosyncratic ways.
This is not always an easy process. In fact, my favorite comeback to big project requests comes from one of my programmers: "The work involved is not exactly trivial!" In this case, we were asked to bring a CMS to life with capabilities for nontechnical end users to enter data. This was before WordPress was released and Joomla! and Drupal were beginning to gain attention. The work was real--and, indeed, it was not trivial.
Some of these challenges continue. Two examples come to mind, which illustrate all four of the new workflows that web administrators must repeatedly face: code writing and review, user consultation, code revision, and user education. These workflows create challenges as well as opportunities. So I'll conclude with the opportunities as a way of sweetening the hard truth that more work is needed, expected, and--most likely --underappreciated.
Merging the Old With the New
The web began as creative commons, and this invited the introduction of new tools for information management. As these evolved beyond HTML and CGI (Common Gateway Interfaces--which could link to databases), CMSs emerged as time-saving platforms. The promise of a CMS is to bring content publishing directly to staff members who have no programming experience through menu-driven screens that provide formatting and metadata. Nowadays, we use many types of CMSs on a daily basis. Web programmers now must ensure that diverse CMSs operate together effectively and seamlessly, easing web users' browsing. It seems that a CMS is a great solution and extends the web further and to more people.
Unsurprisingly, the reality is a bit more complicated. CMSs have with been us for years now, and new developers are jumping in, without much training. In higher education, this might begin with a sharp staffer or student creating a large public website via a blogging platform such as WordPress. But without training, it's possible to set up WordPress sites that look beautiful, but operate with convoluted back-end architectures. These big sites go live, attract attention, and then are delivered complete to professional programmers.
In the case of my own institute, the library team operates as the "fixers." We were recently bequeathed a new website that was built by a public health graduate student. In the hurry to get a polished site live to entice prospective funding, he used WordPress' posting function to create what seem to be static pages--but in the back end, they remain posts. At the present moment, media attention is high, and this "posts-as-pages" strategy has crumbled under the load. At the same time, metadata were always an afterthought, and the project leader (a friend of mine) had to devote her attention to grant making. There was no one available to monitor the site--and keep it robust--before it was gifted to us.
Things came to a head one day when the campus "web farm" relocated a large group of live websites to a new server. News was broadcast and reached the folks who were supposed to be on top of this, but it was not interpreted as crucial. So the site went dark. My group did not have admin rights on the private web host that needed to be updated. We worked with the student and restored the site within 24 hours--as opposed to 30 minutes. Now we have the access we need.
In a perfect world, events such as this would never happen, but in the real world, they can happen at any moment. While CMSs are a good foundation for large websites, there must be programming personnel monitoring how they perform, how they are set up, and how they grow.
Tagging the Old to Make It New
This tale is not the first of its type in my group's job as "web rescuers." Another richly funded project hired outside programmers to build its website, but later dropped the relationship. As elegant and customized WordPress code extensions were neglected, the site went down. Now it is ours to run, and our staffers work very closely with the library team.
This program publishes high-profile briefings and is funded by major organizations such as the Bill & Melinda Gates Foundation. Now that we are on an inside track, we can forestall any web problems. This creates an added benefit: training and teaching. High-profile staff members publish reports that start out as new publications, but later need to move to topical listings. Sounds simple, right? To make that content discoverable, robust metadata are needed from the outset. As part of a full web redesign, my group proposed to tag all reports and establish a database, so that the content can migrate wherever it needs to go. We had just finished a similar project with our largest research group, and now we offer this metadata service to all of our clients. The maturation of the web into a database-driven ecosystem enables us to build closer ties with our clients and create mutual benefit.
This example illustrates the power and potential of pushing library skills such as classification and metadata assignment into the live web. Our colleagues did not see this step as an obvious need, but once we made the connection between shifting content easily and displaying it anywhere that made sense as program goals evolve, their excitement was palpable. And we gained a new team of engaged "data owners" with a stake in discoverability.
Unlock the Opportunities
Throughout the web's growth as a universal medium, we have witnessed a tension between the urge to "lock down" design into CMSs and the impetus to be creative and unfettered. For example, the popular LibGuides' CMS allows skilled designers to be as creative as they wish, but if a local webmaster/design "czar" decides to lock down LibGuides' design options, then all of the data owners using it are limited. But the desire for design freedom persists. At the same time, everyone would like to believe that CMSs free us from much or all of the labor of programming.
My examples demonstrate that we can save some time for everybody involved, but programmers must be in the driver's seat to forestall disasters and reduce unnecessary labor. At the top of the food chain, high-powered domains such as Pinterest offer popular platforms in which users have voted to accept the guidelines they work within. That's a great trend, but my guess is that we will continue in a mixed environment in which the small and artisanal takes our attention.
The opportunities for info pros to prove their value proposition are twofold. First, by "auditing" the information environments we work in, we can discover problems with a CMS and propose solutions. My example of metadata assignment is hardly rocket science, but it points out that knowledge workers are not always remembering this crucial step. Second, and most important, we gain a teaching role in helping content managers do a better job. As that role advances, we become agents who bring stability to the diversity of the web platform. This reaffirms my long-held premise that web coding, content management and display, and discoverability on the open web are fundamental competencies for the profession.
Director of Library and information Resources institute for Research on Labor and Employment University of California-Berkeley
Terence K. Huwe is director of library and information resources at the Institute for Research on Labor and Employment at the University of California-Berkeley. His responsibilities include library administration, reference, and overseeing web services. His email address is firstname.lastname@example.org.
|Printer friendly Cite/link Email Feedback|
|Title Annotation:||Building Digital Libraries|
|Author:||Huwe, Terence K.|
|Publication:||Computers in Libraries|
|Date:||Mar 1, 2016|
|Previous Article:||Leveraging Google analytics: helping users find what they think they want.|
|Next Article:||Going the third way: developing custom software solutions for your library.|