SURVEY AND EVALUATION OF WEB STYLE GUIDES FROM PEER
INSTITUTIONS
We queried Google for ―web style guide‖ and took the first 70 non-duplicating
results for higher education institutions. We collected a few more than 70 web style
guides, but we limited it to a number that would allow us split the workload evenly
given the time we had available for the project. We created an Excel spreadsheet to
evaluate each with the following criteria: web standards, accessibility, and date stamp.
We chose these three criteria in an attempt to quantitatively determine which guides
were worth emulating. We also recorded our initial impressions of the style guide and
overall website for future reference.
Web standards refer to a collection of technological ―best practices‖ defined by
the World Wide Web Consortium (W3C) to make sure web content is more accessible
and usable. Though the complete scope of web standards is still under debate, a few
widely accepted best practices include separating content from presentation and
behavior, using cascading style sheets (CSS) to style documents, using a document type
definition (DTD) in the X/HTML code, validating the files against the W3C
specifications, and testing the website across multiple browsers (Dickerson). By
adhering to web standards, institutions can ensure that their sites will be viable years
13
down the road.
Fortunately, the W3C created online tools (known as validators) that test web
pages for web standards compliance. We used two of these tools, the W3C Markup
Validation Service and W3C CSS Validation Service to test each website style guide for
X/HTML and CSS compliance. Specifically, we recorded pass or fail on the style guides
displayed as X/HTML; six web style guides in our sample group were PDF files and
could not be tested with the W3C tools.
Because MCC is a county-based institution that receives state and federal
funding, we are required by law to make our website meet certain accessibility
standards. Specifically, this means that we seek to comply with the IITAA, by following
the Implementation Guidelines for Web-Based Information and Applications 1.0. IITAA
also satisfies Section 508 of the 1998 Amendment to the Workforce Rehabilitation Act
of 1973 (specifically part 1194.22) as well as the Web Content Accessibility Guidelines
(WCAG) version 1.0, Priorities 1 and 2, recommendations. These guidelines help
ensure that our web content is accessible to those using alternative browsers, screen
readers, and other assistive devices. Hence, we wanted to follow models that made
accessibility a priority in their style guides. We used the online Web Accessibility
Evaluation Tool (WAVE) to test each web style guide page for accessibility (if images
had alternative text and forms had proper labels, for instance) and recorded the errors, if
any.
The connection between web standards and accessibility is important: if a web
page is created using web standards, it is probably (though not certainly) more
14
accessible than a web page created without regard for best practice. Accessible content
ensures that the user will still be able to get the content of the page, regardless of the
device used: screen reader, mobile device, or web browser. As the Web Standards
Project mission statement summarizes, ―These technologies, which we call 'web
standards', are carefully designed to deliver the greatest benefits to the greatest number
of web users while ensuring the long-term viability of any document published on the
Web. Designing and building with these standards simplifies and lowers the cost of
production, while delivering sites that are accessible to more people and more types of
Internet devices‖ (Web Standards Project).
Beyond web standards and accessibility, we also thought it would be useful to
record the date stamp left on the page. It is common practice to leave a date stamp to
show when the web page was last published. The reasoning was that if a web style
guide was recently updated, say within the last year or so, people were actually using it.
The results of our survey were sobering (see Appendix A for survey results).
Only 25% of the web pages surveyed passed the W3C X/HTML validator, 46% passed
the CSS validator, and only 34% passed the WAVE validator. The oldest web style guide
was last updated in 1997, with the most current updated in 2008, and 48% showed no
date stamp. More disheartening was the fact that out of 64 higher education web style
guides tested, only two passed all three tests: Iowa State University and Penn State
University. On the flip side, 25% of the web style guide web pages tested failed all three
tests.
When we saw these results we realized we were still no closer to finding ideal
15
models to follow. It became obvious that there was no direct correlation between web
page compliance and the quality of the web style guide it displays. Just because a web
page in a guide was validated—passed the W3C X/HTML and CSS coding tests—that
did not entail that the content displayed was very helpful or useful. For example, the
Iowa State University guide was one of two that passed all three tests, but the content of
the web style guide itself was minimal and very brief; the section on accessibility was
lifted almost verbatim from the W3C WCAG ―Quick Tips.‖ The Penn State guide was
equally lacking in content, most of which was tied to two dense policy statements. The
majority of the guides that failed validation were inconsistent because they professed
adherence to standards that they failed to implement.
We came to understand that an institution may have a comprehensive web style
guide in place, but if the department does not have support from upper management to
enforce compliance across the university, the web style guide will have little impact on
web visitors, who are still served poorly formed web content that does not meet current
technical standards. For example, the Southern Methodist University web style guide is
comprehensive and well-organized, yet their website’s homepage failed the W3C
X/HTML validation tool with 77 errors. Did that mean the web style guide itself should
be disregarded for the purposes of our study? The converse was true as well: two web
style guide web pages validated perfectly, yet the contents were sorely lacking. Should
these sub-par-yet-well-formed web style guides become our models to follow?
Another difficulty was the dawning realization that there was no ―ideal‖ web
style guide. Ideal for whom? Best for what purpose? It became a question of audience.
16
The guides in our survey were as varied as the websites that housed them—from the
colors and fonts to the organization and content, no two guides covered the same
territory. For example, Drury University and University of the Pacific only included the
traditional editorial content in their web style guide. Topics such as code, web
standards, or graphics—what we would consider the ―web‖ portion of a web style
guide—never entered the picture. That is not to say they did not have policies governing
those web-specific areas, just that those policies were never referenced in their web
style guide.
On the other hand, the majority of institutions (62 out of 70) had comprehensive
web style guides. These guides included not only editorial and visual preferences but
also web-specific standards: detailed X/HTML markup specifications, code samples,
video and audio standards, privacy and copyright policies, not to mention sections on
mobile devices, search engine optimization, and the correct use of Flash and JavaScript
technologies. Clearly, somebody needed this level of detail documented in the web style
guide for those institutions.
Hence, the idea of finding prototypes or ideal models to follow was abandoned.
Nonetheless, there was still great benefit in the survey we conducted. In the handful of
web style guides that appealed to us, it was their content that we found valuable, topic
areas that we could use or borrow from to start our own guide. We realized we were
looking for the wrong thing: we did not need web style guides that passed the W3C
standards compliance tests, we needed common web style guide content areas.
CHAPTER 4
SURVEY OF COMMON WEB CONTENT AREAS FROM PEER INSTITUTIONS
Using the original 70 web style guides, we went back and listed the table of
contents and, when necessary, the specific details of each guide in an Excel spreadsheet.
For example, if a web style guide listed ―Design Standards‖ as a major heading, we
looked at the specific content to determine what they meant by ―Design Standards‖ and
recorded the specific terms. We were not interested so much in how they classified their
content as much as the content areas themselves. As a final note, when a style guide
linked to web-specific policies, say a web accessibility policy, we counted that as being
part of their web style guide, even though technically it is not. Our reason for doing so
was that these supporting documents inform the web style guide, at least indirectly.
When policies were not prominently linked, however, we did not search for them.
Once we listed the table of contents and supporting details, we made a master
list of possible content areas for web style guides. This was a great step forward, in that
we had a bank of terms to draw from for our content areas. Moreover, we determined
from the master list what terms were common to the greatest number of guides and then
created a starter list of content areas that most higher education web style guides shared.